PDA

View Full Version : Goliath Awaits……… On Mars.



MentalAvenger
2007-Oct-02, 01:18 AM
Goliath Awaits……… On Mars.
That’s what I would like to see the headlines read someday soon.

A long time ago, in a website far far away, I came up with a concept for a Mars Rover called Big Al. Perhaps the name wasn’t catchy enough. So I have decided to call it, Goliath. Compared to the other Martian robotic rovers it would truly be a Goliath. The size of a large SUV, Goliath would be able to carry out some serious research on Mars, especially finding a suitable site for a Martian Colony.

Goliath would be a very large rover, about the size of a Ford Expedition, with many capabilities far greater than anything we have sent, or anything we have planned. I envision the drive mechanism to be eight wheel drive, with individual electric motors for each wheel. For simplicity and reliability, it would be a skid-steer design or centrally articulated.

Goliath would be capable of taking core samples down to at least 30 feet and be able to analyze the contents of the cores in great detail on board. It would be able to set seismic transducers, create significant seismic vibrations, and record seismic waves. Goliath would also be able to traverse open ground at about 40mph, using AI and multiple sensors to move safely and autonomously across open areas of the surface of Mars.

When Goliath and others like him have identified a suitable site for a viable colony, He will send a signal to Earth: “Goliath Awaits the human colonists.”

Nowhere Man
2007-Oct-02, 02:22 AM
And what launch vehicle is going to get it there? Remember, you not only have to loft the rover, but also the landing system and the in-flight hardware.

Fred

01101001
2007-Oct-02, 02:38 AM
Perhaps the name wasn’t catchy enough. So I have decided to call it, Goliath.

Yeah, that name's going to capture imaginations: Goliath... that's the big fat ungodly looser that was stopped by a little stone, right?

Needs some marketing rehab.

Gemini
2007-Oct-02, 03:24 AM
And what launch vehicle is going to get it there? Remember, you not only have to loft the rover, but also the landing system and the in-flight hardware.

Fred

Ares V?

JonClarke
2007-Oct-02, 10:02 AM
And what launch vehicle is going to get it there? Remember, you not only have to loft the rover, but also the landing system and the in-flight hardware.

Fred

Payload on the Martian surface of the largest versions of Delta IV or Atlas V are of the order of 5 tonnes. That is exclusive of EDL systems.

Jon

MentalAvenger
2007-Oct-02, 10:47 PM
Yeah, that name's going to capture imaginations: Goliath... that's the big fat ungodly looser that was stopped by a little stone, right?
Needs some marketing rehab.I guess you never saw the movie Goliath Awaits (http://www.imdb.com/title/tt0082461/) Whatever. Anyhow, I was hoping for a discussion on the feasibility and utility of such a rover, not petty nitpicking at the name. Goliath, Big Al, Big Foot, Tiny, whatever………

01101001
2007-Oct-02, 11:07 PM
Anyhow, I was hoping for a discussion on the feasibility and utility of such a rover, not petty nitpicking at the name. Goliath, Big Al, Big Foot, Tiny, whatever………

The name was only an interesting, significant fact when you commented on it. When I commented on it, it was nitpicking. Petty nitpicking, no less.


A long time ago, in a website far far away, I came up with a concept for a Mars Rover called Big Al. Perhaps the name wasn’t catchy enough. So I have decided to call it, Goliath.

Any other peculiar restrictions to the direction we might take the conversation?


Goliath Awaits……… On Mars.
That’s what I would like to see the headlines read someday soon.

Got a backup plan for a headline?

Neverfly
2007-Oct-03, 12:29 AM
<chuckle>
entertaining you two:p

In the meantime...

I agree. I'm sure NASA agrees too. A buttkicking rover would definetly help and sadly, only size can deliver the power needed for hard travelling and core tapping.

The trouble with it is the energy conversion. The larger the rover- the more energy needed to power it.

MentalAvenger
2007-Oct-03, 12:45 AM
Nuclear power, of course.

Neverfly
2007-Oct-03, 04:17 AM
Hmmmm...:think:

What about maintainance? Such a large investment isn't something you want to wear out or run down too quickly.

MentalAvenger
2007-Oct-03, 05:53 AM
Have it built by Checker Motors Corp.

JonClarke
2007-Oct-03, 07:33 AM
Nuclear power, of course.

I doubt very much whether nuclear is going to be feasible for a vehicle of this size. Much better to use fuel cells, you should be able to get perhaps 1000 km out of a single filling. if you link in with a series of ISPP plants you can get effectively infinite range.

Jon

Ronald Brak
2007-Oct-03, 07:52 AM
Using lasers to power craft has been discussed in other threads recently, so this makes me think that perhaps it would be possible to have a satellite in aresynchronos orbit that collects solar energy and then uses lasers to transmit power to the rover. This might sound like a lot of trouble, but it would save on weight that would need to be landed and the satellite could be used for later missions. If Spirit and Opportunity are still going when it arrives it could give them an energy boost when needed.

IsaacKuo
2007-Oct-03, 01:52 PM
Using lasers to power craft has been discussed in other threads recently, so this makes me think that perhaps it would be possible to have a satellite in aresynchronos orbit that collects solar energy and then uses lasers to transmit power to the rover. This might sound like a lot of trouble, but it would save on weight that would need to be landed and the satellite could be used for later missions. If Spirit and Opportunity are still going when it arrives it could give them an energy boost when needed.

Laser beamed power doesn't scale down well. The mass and bulk of the focusing optics remain the same regardless of how much or how little power is being transmitted. It depends on the distance, and unfortunately geosync (or aresync) orbit is a terribly long distance. Beamed power from space works best when you either need a short burst of LOTS of power (laser thermal rocket launch), or you can afford a collection of power satellites. These can operate from low orbit.

Otherwise, plain old solar power works better. With Mars in particular, there's also the issue of dust storms. These may reduce solar powered vehicles to low power, but they would utterly block lasers.

For a mars ground probe, solar power is pretty hard to beat. The only really good alternative I see is indirect solar power--rocket fuel. You'd love to have a solar powered rocket fuel refinery to supply return missions launching from Mars. All the necessary ingredients are in the atmosphere, you just need the storage tanks and the solar powered refining hardware. As long as you've got this keen little fuel refinery with an endless supply of fuel...you might as well use it to fuel a ground vehicle also.

Still, that limits you to a particular radius from the fuel refinery. Excellent for in detail exploration of a limited area, but solar power is still the way to go for global surveys.

R.A.F.
2007-Oct-03, 02:53 PM
I guess you never saw the movie Goliath Awaits (http://www.imdb.com/title/tt0082461/) Whatever.

Thought that "Goliath Awaits" sounded familiar. Is that why you chose the name Goliath?...so you could make that little "word play"??

MentalAvenger
2007-Oct-03, 03:42 PM
Thought that "Goliath Awaits" sounded familiar. Is that why you chose the name Goliath?...so you could make that little "word play"??Yep. :)

KaiYeves
2007-Oct-04, 12:47 AM
I perfer the tagline for Zathura:
Adventure is waiting
So simple, and yet so powerful.

MentalAvenger
2007-Oct-04, 06:08 PM
So, can we get off the name game and actually discuss the practicality of large, SUV sized rovers for Mars?

IsaacKuo
2007-Oct-04, 06:35 PM
Honestly I like the idea of a larger number of smaller probes better. I think we really need to plop some probes onto the polar caps to find out what's there. Maybe even seed a large area with graphite dust to see if ideas for Martian terraforming based on melting the caps might work someday.

Imagine a radioisotope heated probe that can melt its way down through a polar cap. It can take readings all the way down to the bottom. This "ice digger" could be a precursor proving mission for a Europa ocean explorer. Wouldn't that be awesome? It's not just some embiggened Spirit/Opportunity. It's a new sort of probe, exploring a new part of Mars, and which could lead to an exotic explorer for the only extraterrestrial ocean we know of!

Note that the Martian polar caps may be the ideal place for a Martian colony, offering incredible amounts of easily available water. They might also be crucial for one day terraforming Mars.

JonClarke
2007-Oct-04, 10:23 PM
For starters I would not have 64 kph as a requirement. You don't want to drive at that sort of speed cross country. Especially on Mars. Twenty kph would be more than enough. Remember with your sensor suite the rover is going to spend much of its time sitting still. Just as the current rovers do.

Also I would not have a seismic system on board. Deploying geophones is very labour intensive. I suggest a ground penetrating radar in a snake dragged behind. Or, if you except some inefficiencies, in a belly mounted anetenna.

Mind you, for the complexity and cost of such a mission for only a little bit more you could have a human crew which could do ten times as much. I know which I would prefer.

Jon

Neverfly
2007-Oct-05, 12:54 AM
For starters I would not have 64 kph as a requirement. You don't want to drive at that sort of speed cross country. Especially on Mars. Twenty kph would be more than enough. Remember with your sensor suite the rover is going to spend much of its time sitting still. Just as the current rovers do.

Also I would not have a seismic system on board. Deploying geophones is very labour intensive. I suggest a ground penetrating radar in a snake dragged behind. Or, if you except some inefficiencies, in a belly mounted anetenna.

Mind you, for the complexity and cost of such a mission for only a little bit more you could have a human crew which could do ten times as much. I know which I would prefer.

Jon


To both IsaacKyo's post and yours.
I agree and disagree.

IsacKuo, smaller probes would be beneficial and useful. So would would one big large fat one. Why not use both to fulfill all your needs?

Jon, although most of the sensor readings would be taken at a stand still, being able to move quickly to other targets and further targets can be an asset.

I agree with your statement about human interaction. Landing on Mars. However a probe can also stay longer and therefore do more during that time. Again. Given time and planning- maybe both are the most practical.

Ronald Brak
2007-Oct-05, 09:27 AM
IsacKuo, smaller probes would be beneficial and useful. So would would one big large fat one. Why not use both to fulfill all your needs?

Money, money, money, must be funny, in a rich man's world.
- Old Swedish proverb

Ronald Brak
2007-Oct-05, 09:38 AM
A rover would benefit from aerial survallience of the surrounding area of higher resolution than an orbiter could likely provide. I don't know if this idea is impractical or just plain stupid, but a camera that gets a few minutes flight using compressed martian air could be useful. But compressing that much air would take a fair bit of energy.

JonClarke
2007-Oct-05, 10:19 AM
Jon, although most of the sensor readings would be taken at a stand still, being able to move quickly to other targets and further targets can be an asset.

The improvement is surprisingly small. A rover is likely to spend about 90% of its time standing still, especially if ti is going to do detailed site investigations. This rule seems to hold true for both crewed and robotic rovers. Imagine a goliath rover doing a 100 km traverse. If it could travel at 10 kph it could cover this in 10 hours, 20 kph 5 hours, 50 kph 2 hours. But it isn't simply driving, it is working. let's stay it investigates 10 sites on the way, 10 hours at each, that's 100 hours. Total time to do the traverse at the afore mentioned speeds is 110, 105, and 102 hours respectively.

The faster you go, the more power you need, the more sophisticated the AI, and the more rugged the construction. These are extensive things to build into your rover. All four a saving of a few percent in overal time.

Furthermore if you are collecting data, say EM, GPR, magnetometer, gravity, spectrometry, etc. when driving you will want to move slowly, probably walking pace. You don't want your rover bouncing about, it upsets the instrumentation.

Note that 10 kph on Mars is the equivalent of 17 kph on earth, equivalent terrestrial speed increases by a factor of 1.7 for every 1 kph of speed. Note that 62 kph on Mars is the equivalent of 107 on Earth. I have done a lot of off road driving in remote areas, and the rule is slow and careful. Ground as rough as much of Mars I would not want to go over 20 kph on if it were on Earth. On Mars I would take it even slower.


I agree with your statement about human interaction. Landing on Mars. However a probe can also stay longer and therefore do more during that time. Again. Given time and planning- maybe both are the most practical.

I agree. These are not either-or choices. Note too that a crewed pressurised rover will carry at lot of built in instrumentation and have ample power reserve. It would almost certainly be capable of robotic operation and could do a lot of work before a human crew arrive and after they leave.

Jon

JonClarke
2007-Oct-05, 10:23 AM
A rover would benefit from aerial survallience of the surrounding area of higher resolution than an orbiter could likely provide. I don't know if this idea is impractical or just plain stupid, but a camera that gets a few minutes flight using compressed martian air could be useful. But compressing that much air would take a fair bit of energy.

It's hard to imagine a case where more than 30 cm resolution of HiRISE is needed. But a tethered aerobot with a small camera might fill the gap.

Jon

MentalAvenger
2007-Oct-05, 03:58 PM
IsaacKuo,
As we have seen, small rovers are extremely limited in what they can do. We need more than pictures samples of the surface to determine optimum sites for settlements. Only a large, robust rover with heavy equipment can do that.

While the polar caps may be a good source of water, they would definitely not be a good location for settlements. Remember, the solar energy that reaches Mars is about ˝ of that that reaches Earth. Solar energy will probably be the main power source for Martian colonies, so optimizing the location for maximum exposure will be one of the criteria.

MentalAvenger
2007-Oct-05, 04:13 PM
For starters I would not have 64 kph as a requirement. You don't want to drive at that sort of speed cross country. Especially on Mars. Twenty kph would be more than enough. Remember, at 144,798,465 km˛, the surface of Mars is nearly the same as the land surface of the Earth (148,939,100 km˛). That is a LOT of territory to cover, and much of it will be open country with few if any interesting (or at least important) features. If you surveyed one square kilometer ever day, it would take 39, 671 years to survey Mars.


Remember with your sensor suite the rover is going to spend much of its time sitting still. Just as the current rovers do. All the more reason for high speed travel between testing sites.


Also I would not have a seismic system on board. Deploying geophones is very labour intensive. I suggest a ground penetrating radar in a snake dragged behind. Or, if you except some inefficiencies, in a belly mounted anetenna. Good suggestion. I will incorporate that into my design.


Mind you, for the complexity and cost of such a mission for only a little bit more you could have a human crew which could do ten times as much. I know which I would prefer.I disagree. A human mission would cost many times more. Humans are fragile, high maintenance machines that require frequent rests. And while they are self-replicating, the manufacturing cycle is quite long, and the new machines cannot perform useful work for many years and require even higher maintenance. :)

MentalAvenger
2007-Oct-05, 04:21 PM
Manned missions to Mars are inevitable. IMO, they cannot begin too soon. However, it is very important to find the best settlement sites before the first humans arrive for several reasons. First, survival will be tough enough with a base to work from. Just getting everything set up and functioning will take up all of everyone’s time. There won’t be time to go looking for a suitable site.

Second, all the supplies required for the colony will have to be sent ahead. It would be impractical to build the colony anywhere other than where the supplies are landed, especially if the robotic cargo vessels are pre-fitted and used as habitats.

Third, without such a base to work from in the first place, manned missions would have to be round trip, which more than doubles the cost.

MentalAvenger
2007-Oct-05, 04:26 PM
A rover would benefit from aerial survallience of the surrounding area of higher resolution than an orbiter could likely provide.An orbiter does not require much energy to operate once it is in orbit. Also, the platform is more stable than any flying vehicle. A sufficiently high resolution camera mounted on an orbiter could do the job.

MentalAvenger
2007-Oct-05, 04:43 PM
Imagine a goliath rover doing a 100 km traverse. If it could travel at 10 kph it could cover this in 10 hours, 20 kph 5 hours, 50 kph 2 hours. But it isn't simply driving, it is working. let's stay it investigates 10 sites on the way, 10 hours at each, that's 100 hours. Total time to do the traverse at the afore mentioned speeds is 110, 105, and 102 hours respectively.That is an unlikely scenario for the first large rovers. We know that there are great expanses on Mars that are essentially flat, open ground. The purpose of the rover will not be detailed examination of the entire surface, that will come much later when we have colonies there. The purpose of the initial rovers will be detailed examination of potential settlement sites. These will almost certainly be near mountains and ravines, not on open flats. So there will be no need for the rover to make all those stops along the way.


The faster you go, the more power you need, the more sophisticated the AI, and the more rugged the construction. These are extensive things to build into your rover. All four a saving of a few percent in overal time. Again, due to the mission of the rovers, the savings in time will be great. As for construction, Goliath had better be built damn tough. Failure is not an option. And of course, it would travel at a speed compatible with the terrain (marrain?) Flat open desert


Note that 10 kph on Mars is the equivalent of 17 kph on earth, equivalent terrestrial speed increases by a factor of 1.7 for every 1 kph of speed. Huh?????


I have done a lot of off road driving in remote areas, and the rule is slow and careful. So have I. Hills, gullies, and rocks require slow and easy. Flat open desert is great for 80mph.


These are not either-or choices. Note too that a crewed pressurised rover will carry at lot of built in instrumentation and have ample power reserve. It would almost certainly be capable of robotic operation and could do a lot of work before a human crew arrive and after they leave.Check our your current vehicle. Most of the room and features are designed for the passengers. Having a rover that had to have food, water, air, interior room, and other features that humans need would either reduce the usefulness to half, or double the size.

IsaacKuo
2007-Oct-05, 05:19 PM
IsaacKuo, As we have seen, small rovers are extremely limited in what they can do. We need more than pictures samples of the surface to determine optimum sites for settlements. Only a large, robust rover with heavy equipment can do that.

I didn't suggest just more of the same kind of rover, but probes to explore different parts of Mars--so far unexplored. And my focus is on more than just manned settlement of Mars (which isn't going to happen until someday when launcher technology is up for it and after we've demonstrated the required technologies on Earth).

As for your assumption that a heavy equipment rover can help determine sites for settlements--why, exactly? Because you think that all we need to do is drill in the right place and it'll tap into water? Maybe so. Maybe not. Certainly it's worth checking out, but so far the most promising evidence for underground water is around cliff faces too treacherous for a large wheeled rover to investigate. Honestly, a manned mission involving ropes and climbing equipment may be our best shot at finding underground water on Mars. Designing a robot to operate in this sort of environment has proven difficult, here on Earth (most famously, the difficulties the robot Dante has had in volcano exploration). If we can't design a reliable robot to handle cliff faces here on Earth, we're definitely not sending one to Mars.


While the polar caps may be a good source of water, they would definitely not be a good location for settlements. Remember, the solar energy that reaches Mars is about ˝ of that that reaches Earth. Solar energy will probably be the main power source for Martian colonies, so optimizing the location for maximum exposure will be one of the criteria.

We don't actually know if there are any other good sources of water on Mars. If not, then the polar caps may be the best place to go. Lots of solar power with little water? We can do that anywhere. Lots of water and a bit less solar power? That might be the required recipe for growth.

MentalAvenger
2007-Oct-05, 07:48 PM
I didn't suggest just more of the same kind of rover, but probes to explore different parts of Mars--so far unexplored. And my focus is on more than just manned settlement of Mars As Jon pointed out, the kind of exploring you are talking about would be better done by people on the surface. The only time people there will have the time and resources to do any worthwhile exploration is after they have established a viable settlement. The purpose of Goliath and Big Al is to find the best settlement sites, not to explore the planet in detail. First things first.


(which isn't going to happen until someday when launcher technology is up for it and after we've demonstrated the required technologies on Earth).
We have demonstrated all the required technologies here on Earth or in space. The fact that we haven’t demonstrated them all at the same time in the same place does not mean those technologies do not exist or that they have not been tested. Its just that, up till now, we have never had the incentive to put it all together.


As for your assumption that a heavy equipment rover can help determine sites for settlements--why, exactly? Because you think that all we need to do is drill in the right place and it'll tap into water? I doubt that drilling for water with a robot would be very practical, although drilling deep core samples with Goliath might discover some clues that will help us find water. The main purpose of Goliath would be to determine, not just the structure of a given area, but the composition both above and below ground. A robust ground penetrating radar and a deep core sampler, combined with an onboard comprehensive analyzing lab, would give most of the data we would need to determine if a given site is suitable. Don’t forget, at least at first, Martian habitats will be underground. Whether we dig into hillsides with TBMs, or roll robotic supply ship hulls into ravines and cover them over with regolith, the first Martians will live underground. That is why it is so important to know what is under the surface.


Certainly it's worth checking out, but so far the most promising evidence for underground water is around cliff faces too treacherous for a large wheeled rover to investigate. Honestly, a manned mission involving ropes and climbing equipment may be our best shot at finding underground water on Mars. I agree. And once we establish viable settlements, we will no doubt do just that.


Designing a robot to operate in this sort of environment has proven difficult, here on Earth (most famously, the difficulties the robot Dante has had in volcano exploration). If we can't design a reliable robot to handle cliff faces here on Earth, we're definitely not sending one to Mars.Uh……wasn’t that another movie?


We don't actually know if there are any other good sources of water on Mars. If not, then the polar caps may be the best place to go. Lots of solar power with little water? We can do that anywhere. Lots of water and a bit less solar power? That might be the required recipe for growth. You may be right. But from what I have seen, I think there is a good chance of considerable water underground in many place on Mars. Hopefully, Goliath and Big Al can help answer that question.

JonClarke
2007-Oct-05, 11:57 PM
Remember, at 144,798,465 km˛, the surface of Mars is nearly the same as the land surface of the Earth (148,939,100 km˛). That is a LOT of territory to cover, and much of it will be open country with few if any interesting (or at least important) features. If you surveyed one square kilometer ever day, it would take 39, 671 years to survey Mars.

All the more reason for high speed travel between testing sites.

Remember what i wrote earlier.

Whether crewed or robotic, a rover is likely to spend about 90% of its time standing still, especially if it is going to do detailed site investigations. Imagine a goliath rover doing a 100 km traverse. If it could travel at 10 kph it could cover this in 10 hours, 20 kph 5 hours, 50 kph 2 hours. But it isn't simply driving, it is working. let's stay it investigates 10 sites on the way, 10 hours at each, that's 100 hours. Total time to do the traverse at the afore mentioned speeds is 110, 105, and 102 hours respectively.

The faster you go, the more power you need, the more sophisticated the AI, and the more rugged the construction. These are extensive things to build into your rover. All four a saving of a few percent in overal time.

Furthermore if you are collecting data, say EM, GPR, magnetometer, gravity, spectrometry, etc. when driving you will want to move slowly, probably walking pace. You don't want your rover bouncing about, it upsets the instrumentation.

Note that 10 kph on Mars is the equivalent of 17 kph on earth, equivalent terrestrial speed increases by a factor of 1.7 for every 1 kph of speed. Note that 62 kph on Mars is the equivalent of 107 on Earth. I have done a lot of off road driving in remote areas, and the rule is slow and careful. Ground as rough as much of Mars I would not want to go over 20 kph on if it were on Earth. On Mars I would take it even slower.

We have not explored all of Earth as yet, we won't explore all of mars at ground level for centuries. We should not try to do that on many mission. Instead we need to focus on sites of specific interest.


A human mission would cost many times more. Humans are fragile, high maintenance machines that require frequent rests. And while they are self-replicating, the manufacturing cycle is quite long, and the new machines cannot perform useful work for many years and require even higher maintenance.

Advanced robotic missions are much more expensive than human missions per kg. MSL costs about $2.15M per kg. The ISS costs about $212K per kg, a 10th the cost per kg of MSL. Goliath at 5 tonnes is roughly a 10th of the mass of a four person human mission. The missions will cost about the same.

However, the human mission will achieve much more because there will be a whole range of science goals it can address that goliath cannot. These include all those associated with determing how people and their crops and animals can actually live on Mars in the long term.

Humans are fragile, but so are advanced robotics and the computer systems that drive them. Adanced computers require radiation protection, thermal control, sheilding from noise and vibration, magnetic fields, and sudden accelleration. Unlike robots they are also creative, rational, intuitive, flexible and adaptable. Even the most advanced AI will always have to refer back to earth for decisions that exceed its program. This will be a frequent occurrence during exploration of an unknown environment. Exploration by robot on Mars will always be slower than by people on EVA.

BTW this is not an argument against using robots, rather an argument for using robots and humans for the tasks each excell in.

Jon

MentalAvenger
2007-Oct-06, 07:46 AM
Whether crewed or robotic, a rover is likely to spend about 90% of its time standing still, especially if it is going to do detailed site investigations. Imagine a goliath rover doing a 100 km traverse. If it could travel at 10 kph it could cover this in 10 hours, 20 kph 5 hours, 50 kph 2 hours. But it isn't simply driving, it is working. let's stay it investigates 10 sites on the way, 10 hours at each, that's 100 hours. Total time to do the traverse at the afore mentioned speeds is 110, 105, and 102 hours respectively.There you go again with the unrealistic scenario. IMO, the purpose for Goliath and Big Al will be simply to survey possible sites for a viable colony, not to perform a detailed analysis of the entire planet surface. As such, there would be no need to stop at “10 sites along the way” in an open desert. That job can be performed later by colonists when they have the time. The purpose of Goliath and Big Al is to study in detail only those sites determined to be likely candidates for a permanent settlement by orbital surveillance.


The faster you go, the more power you need, the more sophisticated the AI, and the more rugged the construction. These are extensive things to build into your rover. All four a saving of a few percent in overal time We are already using autonomous AI vehicles in rallies in Nevada, following predetermined courses and avoiding obstacles. And this is all being done by amateurs. This is not rocket science anymore.


Furthermore if you are collecting data, say EM, GPR, magnetometer, gravity, spectrometry, etc. when driving you will want to move slowly, probably walking pace. You don't want your rover bouncing about, it upsets the instrumentation. Again, the purpose is not to analyze the entire surface of Mars, but rather to analyze in detail likely sites for settlements. Therefore, 95% of the time will be spent driving from likely site to likely site.


Note that 10 kph on Mars is the equivalent of 17 kph on earth, equivalent terrestrial speed increases by a factor of 1.7 for every 1 kph of speed. Note that 62 kph on Mars is the equivalent of 107 on Earth. Again, WTF are you talking about??????


We have not explored all of Earth as yet, we won't explore all of mars at ground level for centuries. We should not try to do that on many mission. Instead we need to focus on sites of specific interest.That’s the point. Goliath and Big Al won’t be attempting to map the entire surface, but rather only analyze the sites of specific interest for possible settlement sites.


Advanced robotic missions are much more expensive than human missions per kg. MSL costs about $2.15M per kg. You gotta be kidding. Robots don’t need life support and return missions.


The ISS costs about $212K per kg, a 10th the cost per kg of MSL. Please don’t bring up that white elephant as and example. The politics and the NASA extreme inefficiency created a monster that cost at LEAST 10 times what it should have.


Goliath at 5 tonnes is roughly a 10th of the mass of a four person human mission. The missions will cost about the same.Good lord, you cannot be serious. With life support and return capability, the robotic mission would cost FAR FAR less than a manned mission.


However, the human mission will achieve much more because there will be a whole range of science goals it can address that goliath cannot. These include all those associated with determing how people and their crops and animals can actually live on Mars in the long term.Whoa!!! Let’s get the cart back behind the horse where it belongs. First, a round-trip manned mission will not be able to determine any such thing. Look at the windows of return and you will see that there is not time for such experiments. Secondly, ordinary manned missions will have all they can do just to survive on Mars.


Humans are fragile, but so are advanced robotics and the computer systems that drive them. Adanced computers require radiation protection, thermal control, sheilding from noise and vibration, magnetic fields, and sudden accelleration. And all of those conditions have been addressed and have been handled so far.


Even the most advanced AI will always have to refer back to earth for decisions that exceed its program. This will be a frequent occurrence during exploration of an unknown environment. Exploration by robot on Mars will always be slower than by people on EVA.Granted. However, humans require a great deal of continual maintenance and rest that robots do not require. I have worked in the aerospace industry, and helped in making parts used in numerous successful launches. Reliability is a function of the care taken in preparing the components for each spacecraft. Heretofore there has been no oversight in this extremely vital aspect of spacecraft construction. I know, I was there. With a comprehensive oversight on all phases of manufacturing and assembly, and attention to detail, there is no reason Goliath and Big Al cannot be very reliable.

Glutomoto
2007-Oct-09, 01:38 AM
We are already using autonomous AI vehicles in rallies in Nevada, following predetermined courses and avoiding obstacles. And this is all being done by amateurs. This is not rocket science anymore.

What is the life expectancy of those vehicles, most of them don't even complete the course. None of them carries scientific instrumentation, which also needs to survive the ride. So maybe JonClarke has some valid points to make, about the difficulties of doing the same on Mars.
NOVA The Great Robot Race (http://www.pbs.org/wgbh/nova/darpa/team.html)

They are getting better each year, so maybe one day it will be possible.

:)

MentalAvenger
2007-Oct-09, 06:17 AM
As noted, all this is being done by amateurs, with no great incentives and pretty low budgets. It was interesting to see how the different programs handled the same challenges. Actually, AI isn’t necessary for a successful autonomous rover, but it would help. What is really needed is better algorithms. The equipment used probably could have handled the course well with the right programming. With a rover like Goliath or Big Al, no doubt that programming would be provided. It isn’t rocket science.

IsaacKuo
2007-Oct-09, 09:50 AM
Those "amateurs" include military pros and the top experts in the field, with multi-million dollar budgets. In particular, Team Terramax was fielding a military prototype demonstrator for an autonomous military supply truck.

It's not a matter of "providing" software. This IS "rocket science". It's on the bleeding edge of robot technology.

And the robot rally did NOT travel off-road, on unprepared terrain. The relatively high speeds involved were only possible because they were traveling on dirt roads. The robots could assume the existence of a relatively flat dirt road, with obstacles mostly off to the sides. Off-road travel would be orders of magnitude more difficult.

Even with human drivers, travel on unprepared terrain is limited to rather slow speeds. And humans can currently see/comprehend the terrain ahead a couple orders of magnitude further/faster than robots.

MentalAvenger
2007-Oct-09, 06:58 PM
I guess that indicates the competence of those “experts”. I have seen competitions of autonomous aerial vehicles that could pick up objects from one area and put set them down in another, making about 4 separate trips. And all of it totally autonomous.


And the robot rally did NOT travel off-road, on unprepared terrain. Not that one. But there have been others that did travel on roads and cross country without the deliberately placed obstacles.


The robots could assume the existence of a relatively flat dirt road, with obstacles mostly off to the sides. Some of those obstacles were dead center in the roadway. I saw at least one of those vehicles nearly avoid an obstacle, then turn right into it. No excuse for that.


Even with human drivers, travel on unprepared terrain is limited to rather slow speeds. Obviously, you have never driven off road. I drive off road on motorcycles, 4WD SUVs, and ATVs. When I see open ground I take off. It isn’t unusual to drive 40, 50, 60, 70, 80 mph across an open plain, depending upon the terrain. Of course there are places where 2mph is top speed. But on flat open terrain, no one around here drives slow. Not even close.

IsaacKuo
2007-Oct-09, 07:38 PM
I guess that indicates the competence of those “experts”. I have seen competitions of autonomous aerial vehicles that could pick up objects from one area and put set them down in another, making about 4 separate trips. And all of it totally autonomous.

I'm sorry, but it's obvious you are not familiar with robotics technology. Aerial vehicles are by far the easiest robots to implement, due to the ease of control and implementing sensors. There is no need to distinguish between flat and bumpy terrain. There's no need to distinguish between passable ground and an obstacle. It's all obstacle.


Not that one. But there have been others that did travel on roads and cross country without the deliberately placed obstacles.

Umm...okay. You're apparently unaware that you only looked at the qualifying run of the great DARPA race.


Some of those obstacles were dead center in the roadway. I saw at least one of those vehicles nearly avoid an obstacle, then turn right into it. No excuse for that.

Many of the contestants were amateurs and students. But there were a small handful of "serious contenders". Note that only one of the contestants successfully navigated the qualifying obstacle course without hitting anything on the very first try (Standford's "Stanley").


Obviously, you have never driven off road. I drive off road on motorcycles, 4WD SUVs, and ATVs. When I see open ground I take off. It isn’t unusual to drive 40, 50, 60, 70, 80 mph across an open plain, depending upon the terrain. Of course there are places where 2mph is top speed. But on flat open terrain, no one around here drives slow. Not even close.

Yes, it depends on the terrain, but 40mph is suicidal on any sort of rocky terrain. Would you want to go 40mph over this (http://en.wikipedia.org/wiki/Image:Mars_rocks.jpg)?

Tuckerfan
2007-Oct-09, 08:42 PM
Actually, a fleet of small, independently operated rovers is the way to go. First, this gives you redundancy, so if one of them falls off a cliff, your entire mission doesn't go with it.

Second, you get economies of scale working for you as you can amortize the cost over several dozen, if not hundreds, of rovers.

Third, you can broaden the area in which you're covering simultainously. All things being equal, ten rovers can cover more territory in the same amount of time than one can. You can also seed them to different spots on Mars, so while one's exploring the North Pole, another can be crawling around Olympus Mons, a third can cover parts of the equator, a fourth one can explore the South Pole, etc.

Fourth, increased data transfer. Right now, because of orbital mechanics, we can only be in contact with each of the two rovers for about 40 minutes a day (Earth day). This means that we've only got a total of about 80 minutes worth of data (assuming we don't have to send anything to the rovers) to analyze for characterstics we're looking for. Multiple rovers increase the odds that we'll find what we're looking for. We can then program the other rovers to look for similar things, thus increasing the odds we'll find it.

Fifth, we won't be shelling out money to develop a new launch vehicle. Smaller rovers will fit in ones we currently have, that's money which can be spent on building more rovers.

Sixth, if we send the rovers in waves to Mars, then the second group can carrying equipment that can build off of the discoveries made by the first wave, the third wave can build on the second, and so on.

Seventh, longer life due to simplier design. The more complex somthing is, the more likely something will go wrong with it. The space shuttle as compared to the Russian rockets is a perfect example of this.

JonClarke
2007-Oct-10, 11:46 AM
Originally Posted by JonClarke
Whether crewed or robotic, a rover is likely to spend about 90% of its time standing still, especially if it is going to do detailed site investigations. Imagine a goliath rover doing a 100 km traverse. If it could travel at 10 kph it could cover this in 10 hours, 20 kph 5 hours, 50 kph 2 hours. But it isn't simply driving, it is working. let's stay it investigates 10 sites on the way, 10 hours at each, that's 100 hours. Total time to do the traverse at the afore mentioned speeds is 110, 105, and 102 hours respectively.

QUOTE=MentalAvenger;1083460]There you go again with the unrealistic scenario. IMO, the purpose for Goliath and Big Al will be simply to survey possible sites for a viable colony, not to perform a detailed analysis of the entire planet surface. As such, there would be no need to stop at “10 sites along the way” in an open desert. That job can be performed later by colonists when they have the time. The purpose of Goliath and Big Al is to study in detail only those sites determined to be likely candidates for a permanent settlement by orbital surveillance. [/QUOTE]

In what way is this unrealistic? That is the operational reality of both regional exploration and detailed site survey. Speed is largely irrelevant because time spent on traverse in a very small fraction of the whole.

Originally Posted by JonClarke
The faster you go, the more power you need, the more sophisticated the AI, and the more rugged the construction. These are extensive things to build into your rover. All four a saving of a few percent in overal time


We are already using autonomous AI vehicles in rallies in Nevada, following predetermined courses and avoiding obstacles. And this is all being done by amateurs. This is not rocket science anymore.

First of all, as others noted, these vehicles are experimental and operated under extremely contrived conditions. Their actual practicality is yet to be determined. However, even if such vehicles live up to their promise high performance will not be desirable because they will always operate under human supervision and be reporting back to earth for detailed instructions. On Mars that means a time lag.

Originally Posted by JonClarke
Furthermore if you are collecting data, say EM, GPR, magnetometer, gravity, spectrometry, etc. when driving you will want to move slowly, probably walking pace. You don't want your rover bouncing about, it upsets the instrumentation.


Again, the purpose is not to analyze the entire surface of Mars, but rather to analyze in detail likely sites for settlements. Therefore, 95% of the time will be spent driving from likely site to likely site.

I don’t think you realise how far apart likely sites will be on Mars. They will most likely be thousands of km apart, an impractical distance for a rover. I also don’t think you understand the level of investigation needed to characterise a site for a human settlement. For a settlement you will need detailed geological survey of available resources and detailed geotechnical work. That will mean hundreds of line km of close spaced geophysical surveys, several hundred drill holes, complex analyses of thousands of samples, and the recording of meteorological conditions over several Martian years. Your rover will spend most of the time standing still.

Originally Posted by JonClarke
Note that 10 kph on Mars is the equivalent of 17 kph on earth, equivalent terrestrial speed increases by a factor of 1.7 for every 1 kph of speed. Note that 62 kph on Mars is the equivalent of 107 on Earth.


Again, WTF are you talking about??????

Your use of a borderline expletive is noted. What I an talking about is that because of the lower gravity, a particular speed on Mars is the equivalent of a much faster speed on Earth, with a corresponding reduction in traction and stability. It is even worse on the Moon. A safe speed over terratin typical to say the Gusev crater floor on earth might be 20 kph. On mars you could only travel at 11 kph on that terrain.

Originally Posted by JonClarke
We have not explored all of Earth as yet, we won't explore all of mars at ground level for centuries. We should not try to do that on many mission. Instead we need to focus on sites of specific interest.




If you focus on specific sites you don’t need to traverse for thousands of km between sites. This also reduced the need for high speed. Such long range missions are impractical anyway.

Originally Posted by JonClarke
Advanced robotic missions are much more expensive than human missions per kg. MSL costs about $2.15M per kg.


You gotta be kidding. Robots don’t need life support and return missions.

Rather than using an argument from personal incredulity, I suggest you look at the actual numbers. For comparison Cassini (allowing for inflation) cost $723K per kg for an orbiter, the MERs $2.22M per kg, and for complex stationary landers like Viking $532K per kg. Sophisticated robotic spacecraft are extremely expensive and rovers are the most expensive of all. Even without life support and return mission (you have conveniently forgotten sample return missions, another class of high complex and expensive robotic missions.

Originally Posted by JonClarke
The ISS costs about $212K per kg, a 10th the cost per kg of MSL.


Please don’t bring up that white elephant as and example. The politics and the NASA extreme inefficiency created a monster that cost at LEAST 10 times what it should have.

Your erroneous opinions of the ISS are irrelevant. Overpriced compared to what?

The fact remains that the estimated final program cost of the ISS is as I have stated. One 10th the cost per kg of a mars rover mission. By that estimate a human mission would cost about the same as Goliath. Of course, if you are correct, and the ISS cost 10 times what it should have done, then the cost of a human mission per kg becomes even more favourable.

For comparison, adjusted for inflation, Apollo, measured in terms of mass placed in lunar orbit, cost $325K per kg. Not wildly different to the ISS’s costs.

Originally Posted by JonClarke
Goliath at 5 tonnes is roughly a 10th of the mass of a four person human mission. The missions will cost about the same.


Good lord, you cannot be serious. With life support and return capability, the robotic mission would cost FAR FAR less than a manned mission.

My opinion is based on the numbers above and a series of peer reviewed papers I have coauthored on conservative Mars mission study. I did exaggerate, relying on memory The actual masses were 76 tonnes on Mars, so the actual cost of the manned mission will be ~50% more that a 5 tonne rover mission, but much more capable. It will be able to do everything the Goliath can do, and visit to Phobos, return 400 kg of samples, carry our diverse human factors and biomedical research essential to Mars settlement, and have all the advantages of creative, intelligent, flexible, adaptable, intuitive, conscious and aesthetic people on site.

Your opinion is based on what? Incredulity?

Whoa!!! Let’s get the cart back behind the horse where it belongs. First, a round-trip manned mission will not be able to determine any such thing. Look at the windows of return and you will see that there is not time for such experiments. Secondly, ordinary manned missions will have all they can do just to survive on Mars.


Whoa!!! Let’s get the cart back behind the horse where it belongs. First, a round-trip manned mission will not be able to determine any such thing. Look at the windows of return and you will see that there is not time for such experiments. Secondly, ordinary manned missions will have all they can do just to survive on Mars.

What is your basis of this assertion? The first human mission and those that follow will include intensive monitoring of the crew’s health status. Environmental microbiology will also be necessary. All human spaceflights have carried out this research. Do you think that somehow this work will not be carried out on the first mars missions? There will be space and power available for plant growth experiments, and have been a part of research on space missions since the first Salyut. Why should they not be carried out on Mars missions? Animal experiments may be more difficult logistically, but studies of fish and invertebrates should be feasible also.

I agree that it will take more than one mission to answer these questions. But do you really expect that people will launch a Mars settlement program without knowing the answers to them?

Humans are fragile, but so are advanced robotics and the computer systems that drive them. Advanced computers require radiation protection, thermal control, shielding from noise and vibration, magnetic fields, and sudden acceleration.[i]

[QUOTE=MentalAvenger;1083460And all of those conditions have been addressed and have been handled so far.

As they have for human crews.

[i] Even the most advanced AI will always have to refer back to earth for decisions that exceed its program. This will be a frequent occurrence during exploration of an unknown environment. Exploration by robot on Mars will always be slower than by people on EVA.


Granted. However, humans require a great deal of continual maintenance and rest that robots do not require. I have worked in the aerospace industry, and helped in making parts used in numerous successful launches. Reliability is a function of the care taken in preparing the components for each spacecraft. Heretofore there has been no oversight in this extremely vital aspect of spacecraft construction. I know, I was there. With a comprehensive oversight on all phases of manufacturing and assembly, and attention to detail, there is no reason Goliath and Big Al cannot be very reliable.

You miss the point. Reliability is not the issue. Capability vs cost is. For costs of the order required to send Goliath type rovers to Mars you could send humans achieve a lot more faster.

Robotic rovers are best for small comparatively low cost missions like MER and, to a lesser degree, MSL and ExoMars. But they are individually so expensive per kg that they rapidly approach human mission costs as size increases.

If ever a robotic program of such magntiude were undertaken if you be better to have a squadron of small rovers looking at different places. Or, if one sites was the one to be investigated, a "community" of specialised and perhaps cooperating robots that could still achieve mission goals if some failed. Both would be much better approaches than single highly complex rovers which place everything into the one basket.

Jon

Tuckerfan
2007-Oct-10, 04:11 PM
Really, the ideal method of robotic exploration of Mars would be based on this idea. (http://discovermagazine.com/1995/oct/robotbuildthysel569/?searchterm=self-replicating)
According to the vision of Klaus Lackner and Christopher Wendt, a few short decades from now the desert chaparral of what was once the White Sands Missile Range in southern New Mexico will be transformed into a strange new world. For hundreds of miles in every direction the alkali flats will be covered with a blinking array of solar panels. These might look familiar enough, but not the little suitcase-size robots scurrying among the panels on a grid of white ceramic tracks.

The robots, called auxons (from the Greek auxein, to grow), are designed for specialized tasks. Digger auxons scrape an inch of dirt off the desert floor. Transport auxons carry the dirt to a beehive of electrified ovens. Out of these ovens, which work at superhigh temperatures, come useful metals, like iron and aluminum, or the silicon required for making computer chips. Production auxons shape these materials into machine parts and solar panels. Assembly auxons fit them into place. Then the process begins all over again as a new batch of self-replicating automatons rolls into the desert to scoop up another load of dirt.

This electrified grid of tracks and bustling robots grows exponentially across the New Mexican mesas, doubling in size every six months. Though it started out the size of a football field, in ten years it could cover the continent. Before this happens, however, some built-in constraint will tell the system to stop growing. Instead of continuing to reproduce itself, the huge array of solar panels will feed its electricity into the national power grid. This one colony of auxons alone, limited to the test site where the world’s first atomic bomb was exploded, will produce enough power to meet the current electrical energy needs of the United States.A small group of auxons, say 5, lands on Mars (called "Group A") and proceeds to build replicas of themselves until they have 10 of each (making "Group B" a total of 50 auxons). Group A proceeds to go exploring, while Group B begins making 10 copies of each auxon (so "Group C" numbers 500), with the process continuing until the last group hits a pre-determined number of units.

When a group of auxons finds an area that would make a good site for a colony, they start building things like solar arrays, etc. the colony would need when they arrive. Because the auxons would need to be able to analyze soil content, find water, etc., they'd have a wealth of data to transmit back to Earth, so our knowledge of the planet would grow exponentially as the number of auxons grew. In theory, the only thing colonists would have to do when they arrived on Mars was move into their new digs and plant things in their gardens.

MentalAvenger
2007-Oct-11, 12:23 AM
I'm sorry, but it's obvious you are not familiar with robotics technology. Nice Ad Hominem. Does anyone else see a pattern here?


You're simply not getting it.

I'm sorry, but you don't understand basic physics.

You are quite completely wrong about your physics here.

It's becoming increasingly clear that you lack the basic physics knowledge to comprehend what's going on.

I'm afraid you lack that knowledge for now.

What you mistakenly interpret as obfuscation is actually important basic physics distinctions.

but you do not fully "grok" the ways in which they are related.

That's the source of your confusion.

R.A.F.
2007-Oct-11, 01:12 AM
Nice Ad Hominem.

"You don't understand robotics technology because you're stupid" would be an "ad hom". What Isaackuo has posted is not.


Does anyone else see a pattern here?

Yes...the "pattern" being that time and again, Isaackuo, JonClarke and others have pointed out to you why your "giant" rover wouldn't be feasable...and you're having trouble accepting that.

MentalAvenger
2007-Oct-11, 01:12 AM
In what way is this unrealistic? Because it assumes investigating 10 sites in 100km instead of two, one on each end. From what I have seen of the Martian surface, and based upon what I consider to be the requirements for a settlement site, it is
far more likely that it would be two sites perhaps 100, 200, 500 km apart. That will be reinforced by more detailed information from orbiting cameras. IMO.


First of all, as others noted, these vehicles are experimental and operated under extremely contrived conditions. Their actual practicality is yet to be determined. However, even if such vehicles live up to their promise high performance will not be desirable because they will always operate under human supervision and be reporting back to earth for detailed instructions. On Mars that means a time lag. You are entitled to your opinion, but that is all it is. I disagree. Although there will be nearly constant communication between the rover and Earth, much of what the rovers I have described would do does not require direct control. Perhaps your rovers do. You might consider some design changes.


I don’t think you realise how far apart likely sites will be on Mars. They will most likely be thousands of km apart, an impractical distance for a rover. Really? That’s strange. Only a few days ago you said ” let's stay it investigates 10 sites on the way,[ doing a 100 km traverse.] Several orders of magnitude discrepancy there.


I also don’t think you understand the level of investigation needed to characterise a site for a human settlement. Actually I do. That is why my suggestion for a rover is one that is actually capable of that level of investigation. Go back and check some of my original requirements for Goliath and Big Al.


Your use of a borderline expletive is noted. Your noting of such is noted. So what’s your point?


What I an talking about is that because of the lower gravity, a particular speed on Mars is the equivalent of a much faster speed on Earth, with a corresponding reduction in traction and stability. …. yadda, yadda, yadda…. Why didn’t you just say so? Anyhow, how did you arrive at those figures? How do those figures apply to different kinds of rovers. BTW, you forgot about the actual factor that governs this, the difference in inertia.


If you focus on specific sites you don’t need to traverse for thousands of km between sites.
You do if you want to use one very capable well equipped rover, instead of dozens of small rovers with relatively limited capability. Remember your contention about how thorough the survey would have to be. You can’t have it both ways. Unless you plan to use an ornithopter (http://www.technovelgy.com/ct/content.asp?Bnum=44)to move the rovers from location to location.


Rather than using an argument from personal incredulity, I suggest you look at the actual numbers. I have, but the numbers I looked at are at least relevant.


For comparison Cassini (allowing for inflation) cost $723K per kg for an orbiter, the MERs $2.22M per kg…. yadda, yadda, yadda…. Taking figures for radically different designs requirements spread out over some 40 years is irrelevant. As pointed out, the same basic rover, without the systems and supplies required for human extended human life support, could easily carry twice as much useful instrumentation, possibly more. And that does not even include the cost of returning the astronauts to Earth. (which appears to be your viewpoint)


Your erroneous opinions of the ISS are irrelevant. Overpriced compared to what? Good lord, you don’t know? Compared to what it should have cost. And in the process, they built something that has very little actual use other than a political gambit, and costs far too much to maintain.


The fact remains that the estimated final program cost of the ISS is as I have stated. One 10th the cost per kg of a mars rover mission. By that estimate a human mission would cost about the same as Goliath. Perhaps it is that kind of backwards thinking that put so many missions into such incredible cost overruns. Ever think of going into politics?


My opinion is based on the numbers above and a series of peer reviewed papers I have coauthored on conservative Mars mission study. Regardless of how you chose to arrive at those figures, the bottom line is still that the same basic rover, without the systems and supplies required for human extended human life support, could easily carry twice as much useful instrumentation, possibly more. That is so incredibly obvious I cannot imagine how anyone could not see it.


What is your basis of this assertion? The first human mission and those that follow will include intensive monitoring of the crew’s health status. Environmental microbiology will also be necessary. …. yadda, yadda, yadda…. As noted elsewhere, survival on Mars will be difficult, and require most of the time of the participants on short missions. On longer missions, where the crew does not have to return to Earth, and therefore not only has a LOT more time for these experiments, but also doesn’t have to time-share with preparing the return trip, there will be time for such experiments.


I agree that it will take more than one mission to answer these questions. But do you really expect that people will launch a Mars settlement program without knowing the answers to them?We have the ability to answer all the required questions here on Earth. If rovers of significant capability are deployed on Mars in the most likely locations, and do their job properly, we should have enough information to select a suitable settlement site. If properly supplied, and properly prepared, the first mission to Mars should be able to stay there and begin settlement.


You miss the point. Reliability is not the issue. Capability vs cost is. For costs of the order required to send Goliath type rovers to Mars you could send humans achieve a lot more faster. As pointed out, that simply cannot be true.


If ever a robotic program of such magntiude were undertaken if you be better to have a squadron of small rovers looking at different places. Or, if one sites was the one to be investigated, a "community" of specialised and perhaps cooperating robots that could still achieve mission goals if some failed. Both would be much better approaches than single highly complex rovers which place everything into the one basket.In your opinion IMO, more robust rovers, capable of handling the extensive investigation you admit will be needed, will be the best way to find a suitable landing site.

MentalAvenger
2007-Oct-11, 01:22 AM
"You don't understand robotics technology because you're stupid" would be an "ad hom". What Isaackuo has posted is not. Incorrect. Ad Hominem = To the Man. He directed the comment to the opponent, rather than merely stating his viewpoint on the matter. By definition, it is Ad Hominem. Claiming that someone “doesn’t know what they are talking about” has no place in a scientific discussion. It is a meaningless, Ad Hominem comment. Rather than berate the opponent, it is a lot more proper, and a LOT more productive, to simply state your opinion. And, after all, virtually all the comments made here are opinions, if not ours, someone else’s we are quoting or paraphrasing.


Yes...the "pattern" being that time and again, Isaackuo has pointed out to you why your "giant" rover wouldn't be feasable...and you're having trouble accepting that. Incorrect again. Those comments were from an other, entirely unrelated thread that had nothing to do with robots. Before making such judgmental, cutting remarks, perhaps you should check the facts.

R.A.F.
2007-Oct-11, 01:34 AM
...entirely unrelated thread that had nothing to do with robots.

So people are agreeing with your "giant rover" idea?? That isn't evident from reading this thread.

Oh...and I agree with Isaackuo...you are not familiar with robotics technology...so you can add me to your "list". (Your next "move" is to report us to the mods.)

MentalAvenger
2007-Oct-11, 01:59 AM
So people are agreeing with your "giant rover" idea?? That isn't evident from reading this thread. I didn’t say they were. It is a relatively new and revolutionary concept. It may be better, it may not, but it will take a lot more discussion, and a LOT more participants to determine that. Three or four opinions out of 6 billion is hardly conclusive.


Oh...and I agree with Isaackuo...you are not familiar with robotics technology...so you can add me to your "list". (Your next "move" is to report us to the mods.)Why would I do that? You are entitled to your opinion. That doesn’t make you correct, it merely shows your opinion.

BTW, I have designed and built 3 successful robots. Although not terribly sophisticated, two of them did have obstacle avoidance features. Obstacle avoidance isn’t rocket science, it is rather basic. True, the robots can get into situations where they “lock up” because they can’t decide on their own. But, that might be expected when working on a budget of hundreds of dollars, using old power chair motors and acoustic detectors (scavenged from Kodak cameras and 40 ft “laser” measuring devices. But it works.

Control mechanisms are also basic. Proportional controllers from R/C models require more programming, but produce a LOT better control and are easy to implement.

Oh, BTW, how many have you built?

IsaacKuo
2007-Oct-11, 02:43 AM
Dude, you've got some issues. You're not convincing anyone.

Anyway, on rocky terrain the problem isn't "obstacle avoidance" but rather identification of passable terrain. There's a reason why Mars rovers are designed with complex active articulating suspensions--it's to navigate over uneven terrain. No matter how much you just ignore the problem, the people who actually work on Mars missions can't ignore it. They have to solve it.

Again, it's a sensor and sensor interpretation problem. It's not a control problem. As anyone in the field of robotics knows, aerial navigation is the easiest environment for a robot to work in. It's easy to identify obstacles because everything is an obstacle. It's easy to avoid those obstacles because you've got an unlimited sky above to navigate into when in doubt. You can simplify your own robot's body as an abstract spherical cow, with 3 degrees of positional freedom. Anyone in robotics doesn't even have to think twice to answer the question of why we've had fully autonomous military flying robots for half a century, but no autonomous military driving robots even now. Autonomously dealing with the ground in any fashion more sophisticated than crashing into it is non-trivial.

Note that while we've had autonomous military flying robots for half a century, none of these ever had to figure out how to land on passable terrain until the very recently fielded Global Hawk. Before the Global Hawk, no autonomous military robot included any option to "land".

(BTW, my experience with robotics is mostly on pre-built robots and virtual simulations. I've only built one robot of any significant original design; a legged robot which honestly wasn't anything impressive. I'm a software guy, not an engineer.)

MentalAvenger
2007-Oct-11, 04:33 AM
Dude, you've got some issues. Another Ad Hominem, fits your pattern.


You're not convincing anyone. I’m not here to convince anyone of anything. I present my ideas and opinions. It is up to the individual to convince themselves.


Anyway, on rocky terrain the problem isn't "obstacle avoidance" but rather identification of passable terrain. There's a reason why Mars rovers are designed with complex active articulating suspensions--it's to navigate over uneven terrain. No matter how much you just ignore the problem, the people who actually work on Mars missions can't ignore it. They have to solve it.Thank you for highlighting my point. With a two foot rover, even a three inch rock is an obstacle that requires avoidance. Goliath wouldn’t even notice such a rock. Its all a matter of perspective.


Again, it's a sensor and sensor interpretation problem. Of course. And, as I noted, that is a software function. Thank you again for supporting my point.


As anyone in the field of robotics knows, aerial navigation is the easiest environment for a robot to work in. Nice Non Sequitur to the examples of the robots I built.


As anyone in the field of robotics knows, aerial navigation is the easiest environment for a robot to work in. It's easy to identify obstacles because everything is an obstacle. It's easy to avoid those obstacles because you've got an unlimited sky above to navigate into when in doubt. Terrain following radar guidance systems have been in use for a long time. Not only is this used by Cruise Missiles, but human pilots risk their lives on the abilities of that technology. That is only one degree of movement removed from obstacle avoidance.


but no autonomous military driving robots even now. I suggest that they haven’t been developed because there is currently no need for them. With instantaneous control available through satellites, why would it be needed? Technology tends to fill needs.


Before the Global Hawk, no autonomous military robot included any option to "land".Again, there was clearly no need for it.


(BTW, my experience with robotics is mostly on pre-built robots and virtual simulations. I've only built one robot of any significant original design; a legged robot which honestly wasn't anything impressive. I'm a software guy, not an engineer.) Good, then you of all people should understand the critical importance of the correct software.

IsaacKuo
2007-Oct-11, 06:10 AM
It's not just software, but computing hardware also. The sensor interpretation computations are extremely intensive (which you would know if you were familiar with the state of the art). The software algorithms involved are truly bleeding edge and experimental, and they push the very limits of computing hardware capabilities. Binocular vision algorithms are theoretically awesome because in theory they can be very accurate and very fast (it's what humans use, and we're excellent at navigating complex terrain). Best of all, the algorithms are simple and mostly straightforward. Unfortunately, computing hardware simply isn't fast enough for the task, yet.

The other route is to use expensive long range stabilized laser rangefinding hardware. This greatly reduces the computation power required by feeding the computer raw 3d point data. Unfortunately, the sensor hardware itself is expensive, heavy, bulky, and not reliable.

If you look at what Stanford's "Stanley" used, it was an innovative combination of short range nonstabilized rangefinding sensors and a very clever non-binocular vision algorithm. This algorithm was suitable for navigating dirt roads, but not for any sort of uneven terrain. Essentially, it used a color matching algorithm on the assumption that what looks flat at short range will be the same color as what's flat at long range. Clever, and it only requires O(n^2) computation on the pixel data. (Binocular vision pattern matching requires O(n^3).)

This isn't just some matter of providing some well worn software algorithms. It's bleeding edge, and difficult to implement.

Your example of terrain following software is almost laughable. Terrain following software is a simplified version of obstacle avoidance. It's a VERY simple algorithm, even compared to traditional 2d pathfinding algorithms. It's a simplified 2d problem, where the 2d plane in question is turned on the side. There's just one obstacle--the ground--which is a very simple bumpy curve. There's only one degree of motion to worry about--climbing or descending. It's hard to imagine a simpler robotics problem. It almost isn't even a pathfinding algorithm at all. Again, there's a reason why that technology was available for flying autonomous robots decades ago but we still haven't fielded any autonomous military ground robots.

You keep on making statements which are wrong and indicate a lack of familiarity with the subjects you claim expertise in. This argument isn't going anywhere. You say you're not here to convince anyone of anything. Well, good luck with that.

R.A.F.
2007-Oct-15, 11:21 PM
MentalAvenger...are you "done" discussing your "giant rover" idea? It seems you have abandoned this thread.

MentalAvenger
2007-Oct-16, 01:21 AM
When the discussion consists largely of member B telling member A how much member A does not know, there is no point in continuing the discussion. I would like to discuss the idea with someone willing to discuss the actual issues rather than resorting to repeated ad hominems, but I haven’t been too successful so far. And btw, I mean ad hominem in the logical fallacy sense, not the nasty name calling sense.

I consider the idea of a Goliath rover to be a valid option, and possibly even a preferable option, as well as unique amongst rover concepts I have seen. IMO it would do a superior job in the type of scenarios I presented.

R.A.F.
2007-Oct-16, 10:39 AM
I consider the idea of a Goliath rover to be a valid option, and possibly even a preferable option...

This is a science board where posters share their opinions with each other. Some posters here do not agree your opinion concerning the viability of your "giant rover" concept. You seem to think that anyone who disagrees with your opinion is automatically commiting a logical fallacy...they are not.

If you can't stand being disagreed with, then perhaps you shouldn't be expressing opinions on a public opinion board.

MentalAvenger
2007-Oct-16, 01:57 PM
I am rather surprised to get such a response from you. There is a major difference between “opinions”, and Logical Fallacies (http://onegoodmove.org/fallacy/welcome.htm). I welcome differing opinions, they help to find flaws in a concept and assist in refining the concept. In fact, I readily incorporated one of Jon Clarke’s suggestion into my Goliath concept, discarding my original as not nearly as good.

A fallacy is a component of an argument that is demonstrably flawed in its logic or form, thus rendering the argument invalid in whole. (http://en.wikipedia.org/wiki/Fallacy) Logical Fallacies, as they are sometimes used here, do nothing to further either the discussion or the exchange of information. Rather, they are usually an attempt by a poster to either divert the discussion away from the actual point under consideration, or to confuse the issue. Logical Fallacies are often used by someone who finds that their viewpoint is in jeopardy of being found inconsistent with the facts, or when they have made a mistake and don’t want to admit it.

An example of an opinion would be “The rover would stop every 200 yards to do a test”. An example of a logical fallacy might be “We don’t extract oxygen from CO2 here on Earth, so we wouldn’t do so on Mars.” I hope you can understand the difference.

R.A.F.
2007-Oct-16, 02:07 PM
Logical Fallacies are often used by someone who finds that their viewpoint is in jeopardy of being found inconsistent with the facts, or when they have made a mistake and don’t want to admit it.

You've said it all right there...

eburacum45
2007-Oct-17, 07:19 PM
I think that research into artificial intelligence will bring some remarkable results in the next century. I suspect that Martian exploration will be a race between the development of human-rated spacecraft which can take adaptable and autonomous humans to Mars, and the development of adaptable and autonomous artificially intelligent devices which could be delivered by soewhat less expensive spacecraft which are not human-rated.

It may well be that humans are the best bet for Martian exploration if it happens in the first half of the 21st century, but if we delay it too long, humans won't be able to compete with robots.

MentalAvenger
2007-Oct-17, 10:29 PM
You've said it all right there...It’s hard to tell what you are trying to say with your cryptic message. There are several possible inferences. However, the one that seems to be inferred, based on the out-of-context quote and your previous statements, leads me to believe that you are confusing Logical Fallacies with “unworkable ideas”, or something in that vein. AFAIK, I have not used any logical fallacies in my statements. If I have, perhaps someone could point them out so I can correct them. :)

MentalAvenger
2007-Oct-17, 10:29 PM
I think that research into artificial intelligence will bring some remarkable results in the next century. I suspect that Martian exploration will be a race between the development of human-rated spacecraft which can take adaptable and autonomous humans to Mars, and the development of adaptable and autonomous artificially intelligent devices which could be delivered by soewhat less expensive spacecraft which are not human-rated.I would agree, except for the fact that, IMO, robots will be required to find practical sites for the humans, so robots must necessarily proceed humans. And robotic cargo ships would be used before humans arrive to pre-supply the chosen area(s). A lot more supplies could be sent a lot cheaper using unmanned craft sent on the economy route.


It may well be that humans are the best bet for Martian exploration if it happens in the first half of the 21st century, but if we delay it too long, humans won't be able to compete with robots. It depends upon the purpose of the exploration. Finding a settlement site is one purpose, finding resources is another, and mapping the entire planet is a third. Some people might put looking for aliens in there somewhere too.

IsaacKuo
2007-Oct-17, 10:38 PM
It’s hard to tell what you are trying to say with your cryptic message.

It was crystal clear to the rest of us.

MentalAvenger
2007-Oct-17, 10:39 PM
Meanwhile, I still think that Goliath is a viable concept for the initial exploration/search for a suitable settlement site. Sample return missions have been suggested so that samples could be studied in great detail here on Earth in a lab. But what if we sent a complete sophisticated lab to Mars inside a Goliath Rover? Not only would that be ultimately less expensive, but we would get results a years sooner. In addition, we could keep analyzing Mars in detail without having to send additional sample return missions.

MentalAvenger
2007-Oct-17, 10:41 PM
It was crystal clear to the rest of us.Great. Spell it out. Remember, this is in reference to Logical Fallacies (http://onegoodmove.org/fallacy/welcome.htm) of the nature you will find at that link.

IsaacKuo
2007-Oct-17, 10:46 PM
Great. Spell it out.

I already tried that and it's a waste of time.

MentalAvenger
2007-Oct-17, 11:29 PM
I already tried that and it's a waste of time.That is false. You have not even addressed the issue I was referring to. Try to follow the thread, keep it in context. R.A.F. could have meant several things by that statement. It appears to me that he misunderstands the concept of Logical Fallacies (http://onegoodmove.org/fallacy/welcome.htm), or chose to use one to make an invalid argument. Your response indicates the same. Remember, none of this has anything whatsoever to do with whether or not my Goliath concept is workable, or whether or not I agree or disagree with your evaluation of the Goliath concept or the comments you made regarding it. You can choose to respond in context, or you can continue to make irrelevant comments. Whatever.

R.A.F.
2007-Oct-21, 04:27 PM
R.A.F. could have meant several things by that statement.

Isaac had no trouble figuring it out.

You present your giant probe idea, but what you don't present is any actual reason why it would be better alternative to smaller, less expensive, more versatile probes.

Running round "real fast", and digging holes isn't enough. As I pointed out in my first post to this thread, you seem to have based your whole idea on the name of an old movie, but evidently, your reasoning hasn't progressed from that point.

John Jones
2007-Oct-21, 07:34 PM
Oops.

Double post.

My bad

John Jones
2007-Oct-21, 07:35 PM
.... But what if we sent a complete sophisticated lab to Mars inside a Goliath Rover? ...

I doubt that such a thing is even possible. Nevermind the complete, sophisticated lab equipment, but the complete, sophisticated supporting lab facilities won't fit into an SUV.

MentalAvenger
2007-Oct-21, 07:52 PM
Isaac had no trouble figuring it out.Apparently he made the same misinterpretation you did. He figured out what you figured out, not what was actually said.


You present your giant probe idea, but what you don't present is any actual reason why it would be better alternative to smaller, less expensive, more versatile probes. I understand completely. What you have yet to address is the fact that we were discussing “logical fallacies” at that time, and the quote even included the phrase. It appears that you have interpreted “logical fallacies” to be faulty logic, which it is not, as explained in the links I provided. It isn’t just semantics, it is an important distinction. Perhaps you think the logic I used to propose Goliath was flawed. But, in the process, I did not use (AFAIK) any Logical Fallacies as defined by the common definitions.

And, contrary to your claim, I have presented many different, and IMO valid reasons for such a rover. There are some vital things such a rover could do that small rovers simply are not capable of. In some cases, size really does matter. At about 408 pounds (on Earth), Opportunity is limited in what it can carry, especially since so much of it’s mass dedicated to simply getting it around. Compare that to the potential capabilities of a 6000 pound rover, and the variety and completeness of the equipment it could carry.


Running round "real fast", and digging holes isn't enough. Of course not, and that isn’t what Goliath would do. As noted elsewhere, having a very complete lab on board (something the small rovers cannot have) would save sample return missions and speed up analysis by many many years. However, it would also be able to move quickly and relatively easily from site to site.


As I pointed out in my first post to this thread, you seem to have based your whole idea on the name of an old movie, That is not true. The movie title came as a title inspiration after I had written the OP. I chose Goliath originally because “Big Al” has been confused with “Big AI” in the past, and I did not intend the rover to be primarily Artificial Intelligence. Primarily, “Big Al” was a mobile laboratory capable of taking deep samples from a variety of places.


but evidently, your reasoning hasn't progressed from that point. The fact is, that wasn’t my “reasoning”, as I have pointed out in several places. Now, if you wish to discuss specific issues regarding the Goliath concept, I will attempt to politely respond. If I disagree, I will say so, but if I agree with some of your points, I will also say so. I am not afraid of admitting when I am wrong. When I am wrong, I say I am wrong.

BTW, in my opinion, there have been several members here who disagreed with my opinion, but never did supply credible support for their opnions, and since we have surveyed so very little of Mars, virtually all of the comments here have been just that, opinions. I believe that everyone has a right to express their opinion………on the subject at hand. IMO, that right does not extend to making claims about another member’s knowledge, expertise, or intelligence. If someone disagrees, they should provide refuting evidence on the subject of the discussion, without commentary on their opinion of the other members’ credibility or expertise. I don’t think I can be more clear than that.

MentalAvenger
2007-Oct-21, 07:59 PM
I doubt that such a thing is even possible. Nevermind the complete, sophisticated lab equipment, but the complete, sophisticated supporting lab facilities won't fit into an SUV. Point well taken, thank you. However, doesn’t that presuppose a laboratory designed to be used by people on Earth? Rovers already have some analysis capabilities. On Earth we already have totally automated laboratory processes (I ought to know, I worked on the development of such an automated analysis system). Combining what we now use with miniaturization and keeping it all within the requirements for such analysis on Mars, I think it is quite reasonable to expect a 6000 pound rover to be able to carry enough equipment (without having to devote a large percentage to human life support) to do analysis detailed enough to determine settlement sites on Mars. That is something our current rovers cannot do at all. Not even close.

John Jones
2007-Oct-21, 08:09 PM
Point well taken, thank you. However, doesn’t that presuppose a laboratory designed to be used by people on Earth? Rovers already have some analysis capabilities. On Earth we already have totally automated laboratory processes (I ought to know, I worked on the development of such an automated analysis system). Combining what we now use with miniaturization and keeping it all within the requirements for such analysis on Mars, I think it is quite reasonable to expect a 6000 pound rover to be able to carry enough equipment (without having to devote a large percentage to human life support) to do analysis detailed enough to determine settlement sites on Mars. That is something our current rovers cannot do at all. Not even close.


Aren't you presupposing a laboratory design requiring no real-time operation/intervention by people on-site?

Which totally automated lab processes are you refering to?

I have some modest experience in designing automated lab processes myself.

R.A.F.
2007-Oct-21, 08:18 PM
If someone disagrees, they should provide refuting evidence on the subject of the discussion, without commentary on their opinion of the other members’ credibility or expertise.

You would understand why your giant probe idea is simply not viable if you had the "expertise" in the design/construction/purpose of planetary probes. But when anyone points that out to you, you scream "logical fallicies", so there is really no point in continuing this discussion...I'm outta here....

MentalAvenger
2007-Oct-21, 08:24 PM
Aren't you presupposing a laboratory design requiring no real-time operation/intervention by people on-site?Yes, that’s the whole point after all. And of course, real time intervention would be impossible on Mars without a manned presence there, and the whole purpose of Goliath would be to prepare for that manned presence.


Which totally automated lab processes are you refering to? That all depends upon what exactly we would be looking for. For instance, if Goliath could drill down and take a core sample 30 meters down, there are several things we would need to know, and probably many things that would be essentially irrelevant. Tests for water, nitrogen, carbon and other elements would be standard. But the lab should be capable of analyzing for complex compounds that would be relevant to a suitable settlement site. Precisely which compounds to test for would be determined by those responsible for deciding upon a settlement site.


I have some modest experience in designing automated lab processes myself. Then perhaps you could answer some of your own questions.

MentalAvenger
2007-Oct-21, 08:42 PM
You would understand why your giant probe idea is simply not viable if you had the "expertise" in the design/construction/purpose of planetary probes. So you say. But I have yet to see any credible refutations to my design. I provided what I consider to be valid reasons for testing randomly over a large area. The refutations consisted mainly of “You are wrong, they wouldn’t do that”, without any credible or reasonable backup.


But when anyone points that out to you, you scream "logical fallicies", That is 100% false and I highly resent the false accusation. I didn’t know that was allowed here. I pointed out Logical Fallacies only when they were used. I can show, using the links I provided, the credible backup for each and every instance of a Logical Fallacy I pointed out. And the only reason I pointed them out, was that when the Logical Fallacies became so numerous, it became impossible to continue a reasonable, logical, rational debate.


I'm outta here.... Your choice, of course.