What would your dream Martian robot rover be like?

[j05h]

>JPL/Cat _would_ take up work for nuclear multi-purpose bulldozer, weighing so little that it can be soft-landed on mars with currently available transportation methods ? Like, how-near term does that sound to you ? Which reactor were you planning to use ??<br />Meanwhile, you can go and rent a working ASIMO from Honda right now ( yes, they are commercially leased ) . Its capabilities are pretty much as good as you can program it.<br /><br />I am thinking a starting 'dozer in the 1000kg range, using airbag-retros or Viking-style soft descent. The initial power source would be a DoE standard RTG. Later versions of the vehicle would mass 5-10 tons and use an SP100 reactor or bigger RTG. RTGs are off-the-shelf for NASA and AF, a company with the right connections could get access to them. It'd be a great source of income for Oak Ridge labs. If you can figure it out, go solar and use electric power cables (like Catepilar has) to power the dozers from a central point. There has been academic research into dozer requirements including rough specs. Tele-op dozers are definitely a near-term vehicle, if the market calls for it.<br /><br />As for renting an ASIMO, that is really cool but only for promo purposes. It's no where near space capable. Look to Robonaut for how hard even the simpler part of a human body (torso, head, arms) can fail to be copied. Walking is even harder. I'm not discounting that androids will exist some day, but they aren't going to be installing bases on the moon in the next few decades. If we don't build bases on the Moon soon, it may never happen, ergo the first bases will be built by something other than androids: dozers and familiar space-station modules.<br /><br />I guarantee, if there are going to be any moon bases that require site-preparation, there will be some kind of dozer vehicle included. <br /><br />from willp: /> I did not imply otherwise. I replied to you because of the need to expand on what you wrote. <br /><br />I hope that I was able t <div class="Discussion_UserSignature"> <div align="center"><em>We need a first generation of pioneers.</em><br /></div> </div>

 

[JonClarke]

JC wrote:<br /><br /> Define "almost here". Five years, 10 years, 20 years? Fifty years? <br /><br />In respnse you wrote: <i>five to ten years tops. Quite a lot of relevant technology areas are moving forward in that direction ( MEMS sensors and actuators, embedded computing power, energy dense power storage ) and advent of consumer robotics is accelerating developments even further.</i><br /><br />That is interesting, and no doubt will have a lot of applications, but how does this equate to humanoid robots on Mars? Humanoid robots need 13 times as much power per kg as a wheeled one just to move, communicate, and stay upright. It makes no sense from an engineering perspective to use such an inefficient approach.<br /><br />JC wrote:<br /><br /> In reply to:<br /><br /> What possible use would a humanoid of semi humanoid form be for a Mars rover?<br /><br />You wrote: <i> Climbing into places that wheeled rovers cant get to. Digging. Using all sorts of different tools that humans would. Need a new set of tools for new investigations ? simple, just soft-land a new toolkit nearby, your bot will go pick them up and make use of them. Building. Tending prototype equipment and experiments. Basically, like said, they could perform any acts that humans could, physically</i><br /><br />Wheel rovers can probably reach 90% or more of the places that need to be got. The remaining places are going to be small confined spaces such as caves or ascending/descend cliffs. It would be better to use a specialised robot for this task. <br /><br />As for digging, again a specialised machine would out perform a generalised one. If you want to big you would use a backhoe or bucket or auger, not a humanoid robot with a shovel.<br /><br />Tending equipment and experiments is something that requires not just human level flexibility, and dexterity, it requires adaptability, innovation, perception, in short consciousness. Are you arguing that artificial consciousness is only 5-10 years away?<br /><br />A <div class="Discussion_UserSignature"> <p><em>Whether we become a multi-planet species with unlimited horizons, or are forever confined to Earth will be decided in the twenty-first century amid the vast plains, rugged canyons and lofty mountains of Mars</em>  Arthur Clarke</p> </div>

 

[JonClarke]

"I should note that the MERs are somewhere in the gray area between being "tele-operated" and autonomous."<br /><br />I agree. And this is always going to be a limit because whether you are looking at science or engineering tasks you will always want human oversight. This is not an issue with truly automated tasks like ISRU or environmental monitoring or simple assembly, but for field science or engineering the longer the time delay the slower the task.<br /><br />At least until the robots become self conscious and start setting their own goals......<br /><br />Jon <div class="Discussion_UserSignature"> <p><em>Whether we become a multi-planet species with unlimited horizons, or are forever confined to Earth will be decided in the twenty-first century amid the vast plains, rugged canyons and lofty mountains of Mars</em>  Arthur Clarke</p> </div>

 

[JonClarke]

" am terribly sorry, maybe its my reading comprehension problem ( not a native english-speaker ) "<br /><br />Wow! I would never have picked that. You communicate exceptionally well! What is your native language, BTW?<br /><br />Jon <div class="Discussion_UserSignature"> <p><em>Whether we become a multi-planet species with unlimited horizons, or are forever confined to Earth will be decided in the twenty-first century amid the vast plains, rugged canyons and lofty mountains of Mars</em>  Arthur Clarke</p> </div>

 

[JonClarke]

"I'd hazard that JPL and Catepilar would take up the challenge if money was available. "<br /><br />There is a lot of industry-funded work into robotic mining equipment at the Centre for Field Robotics at the University of Sydney.<br /><br />The equipment need not be large for Mars. If you wanted to completely bury a cylindrial base module 4.5 m in diameter & 12 m long, assuming uncohesive regolith you would need to excavate 141 m3 and then back fill it, 282 m3 in all. An excavator the size of a small Bobcat might have a bucket of 0.25 m3 and move 1 m3 an hour. Assuming only 8 hour operation for reasons of light this could excavate and then bury the module in 36 days. You could use a smaller vehicle but there is a minimum size needed to shift more recalcitrant materials.<br /><br />Jon <div class="Discussion_UserSignature"> <p><em>Whether we become a multi-planet species with unlimited horizons, or are forever confined to Earth will be decided in the twenty-first century amid the vast plains, rugged canyons and lofty mountains of Mars</em>  Arthur Clarke</p> </div>

 

[JonClarke]

"how would you go about designing a robot that would land near lunar poles and determine the form of buried hydrogen there ?"<br /><br />I don't think entry descent and landing is a problem - both rocket braking and airbags would work, There are issues with trajectory, I believe, but not insummountable. At least for the northern hemsiphjere, which islow altitude. the southern polar area is very high and could be tricky for landing.<br /><br />We know from the neutron data of MGS and radar from MEx that there is ice with 1 m of the surface north and south of 60 dedrees. We also know there is ice in the polar caps themselves and some craters.<br /><br />For exposed ice I suggest an IR spectrometer (SWIR and TIR bands). For shallow subsurface ice a suggest a passive netruon detector and a ground penetrating radar mounted on the belly of the rover. One was considered for MSL apparfently, but not selected. of you could go to an active neutron beam to probe the subsurface, as MSL will do. Now that's a technology that can only be used on an unmanned rover. If you value your gonads that is!<br /><br />On that note I must go to work....<br /><br />Jon <div class="Discussion_UserSignature"> <p><em>Whether we become a multi-planet species with unlimited horizons, or are forever confined to Earth will be decided in the twenty-first century amid the vast plains, rugged canyons and lofty mountains of Mars</em>  Arthur Clarke</p> </div>

 

[willpittenger]

<blockquote><font class="small">In reply to:</font><hr /><p>from willp: > I did not imply otherwise. I replied to you because of the need to expand on what you wrote.<br /><br />I hope that I was able to clear it up.<p><hr /></p></p></blockquote><br />In the future, please reply to the specific post. That would help with understanding what you are talking about. You might also use the Quote tag. <div class="Discussion_UserSignature"> <hr style="margin-top:0.5em;margin-bottom:0.5em" />Will Pittenger<hr style="margin-top:0.5em;margin-bottom:0.5em" />Add this user box to your Wikipedia User Page to show your support for the SDC forums: <div style="margin-left:1em">{{User:Will Pittenger/User Boxes/Space.com Account}}</div> </div>

 

[willpittenger]

Thanks. Remember that a settlement might have anywhere from 5 to 100 modules depending on the mission and size of the crew. (The 100 modules would be mainly for colonization missions.) However, not all would have to be in place when the first humans arrive. However, you would not want to be in modules that were not 100% buried. I would also suggest staying out of modules that did not have at least one more module or airlock between you and the outside atmosphere in case the module interconnects leak. That would be far more likely where soil or rock is falling around the future and unfinished joint. <div class="Discussion_UserSignature"> <hr style="margin-top:0.5em;margin-bottom:0.5em" />Will Pittenger<hr style="margin-top:0.5em;margin-bottom:0.5em" />Add this user box to your Wikipedia User Page to show your support for the SDC forums: <div style="margin-left:1em">{{User:Will Pittenger/User Boxes/Space.com Account}}</div> </div>

 

[willpittenger]

Then it might be like <i>The Matrix</i> or <i>Terminator</i>. If they get that good, what do they need us for anymore? <div class="Discussion_UserSignature"> <hr style="margin-top:0.5em;margin-bottom:0.5em" />Will Pittenger<hr style="margin-top:0.5em;margin-bottom:0.5em" />Add this user box to your Wikipedia User Page to show your support for the SDC forums: <div style="margin-left:1em">{{User:Will Pittenger/User Boxes/Space.com Account}}</div> </div>

 

[willpittenger]

I think it is time to give up on this troll. He does not appear to have an open mind or back up his arguments. <div class="Discussion_UserSignature"> <hr style="margin-top:0.5em;margin-bottom:0.5em" />Will Pittenger<hr style="margin-top:0.5em;margin-bottom:0.5em" />Add this user box to your Wikipedia User Page to show your support for the SDC forums: <div style="margin-left:1em">{{User:Will Pittenger/User Boxes/Space.com Account}}</div> </div>

 

[no_way]

JC, we are basically arguing over specialized robots vs. general purpose ( and we are both stretching arguments this and other way to fit our purposes )<br />Like you saying 5-10 years to land a new set of tools for your old robot .. what happened to 6 month mars transit time ? <br /><br />Yes, specialized tools are more efficient in power usage, they get these specific tasks done faster etc. However, they are not very efficient in mass, for a given set of tasks.<br /><br />But my impression here is that there is so much different stuff to be done ( even before humans get there ) that deploying a different robot for each of few tasks would be highly inefficient, launch mass wise. If launching mass to mars wouldnt cost anything, we wouldnt be even having this discussion.<br /><br />So what would a general-purpose martian exploration robot, no, call it a pioneering robot doing basically site preparation, be like ? Most likely it wont be a humanoid, but a six-wheeled rover-thingy wouldnt be very useful either. We will probably see something in-between.<br /><br />Logically, you would want to make the robot as flexible as possible to perform as wide set of tasks as possible without adding extra mass, because getting mass to mars is expensive.<br /><br />I'll answer couple of your points separately ( im not going to pick through all the text because we largely agree on the point that specialized tools are more efficient at specialized tasks )<br /><br /><blockquote><font class="small">In reply to:</font><hr /><p>Tending equipment and experiments is something that requires not just human level flexibility, and dexterity, it requires adaptability, innovation, perception, in short consciousness.<p><hr /></p></p></blockquote><br />Yes, and thats where human teleassistance comes into play. Adaptability and innovation can be imported in truckloads by pack of highly paid engineers staring at their screens. Like its done for current rovers when they hit snags.<br />(note that perception is something that

 

[j05h]

<blockquote><font class="small">In reply to:</font><hr /><p> > In the future, please reply to the specific post. That would help with understanding what you are talking about. You might also use the Quote tag.<p><hr /></p></p></blockquote><br /><br />Still learning UBB code. I sometimes condense my replies into one post, so as not to spread out the threads- it's weird to see 3 entries in a row from the same user! <img src="/images/icons/smile.gif" /><br /><br />I still like the idea of at least one type of rover being simple (like a super-pressure, solar heated balloon) so that the probes can be left to run semi-autonomously for years at a time. I'm heading for vacation tomorrow, so probably won't be posting much for the next week. <br /><br />Josh <div class="Discussion_UserSignature"> <div align="center"><em>We need a first generation of pioneers.</em><br /></div> </div>

 

[j05h]

<blockquote><font class="small">In reply to:</font><hr /><p>But my impression here is that there is so much different stuff to be done ( even before humans get there ) that deploying a different robot for each of few tasks would be highly inefficient, launch mass wise. If launching mass to mars wouldnt cost anything, we wouldnt be even having this discussion. <p><hr /></p></p></blockquote><br /><br />This is why a multi-tool bulldozer makes sense, at least for base preparation. I'm arguing that we'll be prepping sites on Mars long before humanoid robots are up to the task. Which is why studies done on robotic site prep on other worlds have assumed boxes-with-wheels. A large rover (the dozer), it's landing shell being a service station, and a small array of attachments (backhoe, bucket, blade, jackhammer) can be designed and built today, flight-weight and ready to fly. <br /><br />The in-between vehicle appeared in some promotional literature a few years ago: it's a "Robonaut" humanoid torso on a space-rated Segway scooter! It's a great idea, except that this version of Robonaut isn't getting built. <br /><br />Asimo is really cool, so was Robonaut, but then, so was the Mad Turk in it's day. None of these is "flight ready". If they'd put a shovel instead of spectrometer, the MER rovers would already be a prototype of what I'm describing. I look forward to the days when you can rent a Honda Builder (tm) and run it around the Moon, but it's a long time away. I also look forward to exoskeletons and neural interfaces, but recognize what can be done with today's tech. <br /><br />On the difference that keeps coming up over exploration versus construction, I think it will prove best to separate the two. Current exploration strategies (rovers, balloons, aircraft) are all very useful and can be extended into future explorer-robots. These all can evolve into various Mars-based tools later, your ground station (pathfinder clone) might evolve into local network of weather/seismo stations. If a pra <div class="Discussion_UserSignature"> <div align="center"><em>We need a first generation of pioneers.</em><br /></div> </div>

 

[JonClarke]

Just some quick thoughts.<br /><br />On the 5-10 year time lag. I am going on the time taken to develop a new mission. So mission X discovers a critical need for tool W. To develop that tool and fly it will take 5 to 10 years at current funding levels. It could be done quicker if more funding was available and a different planning strategy were used. <br /><br />I would have thought that specialised tools were <b>more</b> effective for given mass. It is generalised tools that are not efficient. And specialised need not mean inflexible or limited in capability either. Planetary rovers are specilised exploration robots but still carry a diverse range of instruments- MER 6, Lunakhod and MSL 12<br /><br />General agreement on telepresence and teleoperation.<br /><br />General agreement on robotic conciousness (or lack of it) <img src="/images/icons/smile.gif" /><br /><br />When it comes to "use" of the current I suppose I am being utilitarian from the perspect of space applications. The real problem with humanoid form is that most of the system resources - power, mass, computing power is devoted to the humanoid aspect. ASMO needs (from memory) 9 times as much power per kg as a MER and does not do anything that would be remotely useful on a planetary surface (apart from simple imaging).<br /><br />The couple of space hardened and field ruggedized ASIMOs on Mars would be cute. But like McKay's rose green house idea. But I would rather see a couple more MERs for the same price <img src="/images/icons/smile.gif" /><br /><br />Estonia? You must be one of the few here. Nice to have you!<br /><br />Jon <div class="Discussion_UserSignature"> <p><em>Whether we become a multi-planet species with unlimited horizons, or are forever confined to Earth will be decided in the twenty-first century amid the vast plains, rugged canyons and lofty mountains of Mars</em>  Arthur Clarke</p> </div>

 

[rogers_buck]

Seems like a LIDAR might be interesting to look at the Martian atmosphere dust content. Perhaps the LIDAR could do double duty in range surveys. You could even use a laser as a designator for a smart bomb style projectile released from an orbiter to perform impromptu excavations. The shock and awe rover...<br /><br />Maybe a powerfull communications laser could moonlight for these purposes when Earth wasn't in the line of sight...<br /><br />

 

[JonClarke]

LIDARs are way cool, although my use with them has been confined to surveying for digital elevation models.<br /><br />The use of lasers for for terminal guidance on Mars is certainly novel!<br /><br />Jon <div class="Discussion_UserSignature"> <p><em>Whether we become a multi-planet species with unlimited horizons, or are forever confined to Earth will be decided in the twenty-first century amid the vast plains, rugged canyons and lofty mountains of Mars</em>  Arthur Clarke</p> </div>

 

[willpittenger]

As I understand it, lasers for communication would be useful only for orbiters. Landers would have to get the laser past the atmosphere. <div class="Discussion_UserSignature"> <hr style="margin-top:0.5em;margin-bottom:0.5em" />Will Pittenger<hr style="margin-top:0.5em;margin-bottom:0.5em" />Add this user box to your Wikipedia User Page to show your support for the SDC forums: <div style="margin-left:1em">{{User:Will Pittenger/User Boxes/Space.com Account}}</div> </div>

 

[rogers_buck]

There would be some dispersion even though it is a skimpy atmosphere - otherwise the LIDAR wouldn't be too interesting... If the transmitter had some dynamic dispersion compensation capabilities it shouldn't be all that hard to clean up what the atmosphere does to your pulses. Seems like a tenuous atmosphere like mars might be pretty stable within seasons - but it could be the opposite for all I know.<br />

 

[JonClarke]

Lasers have been successful used to cmmunicate through Earth's to satellites and space probes, so Mars should not be a problem. Just a technical quibble, LIDAR and communications are two very different applications of lasers.<br /><br />Jon<br /><br /> <div class="Discussion_UserSignature"> <p><em>Whether we become a multi-planet species with unlimited horizons, or are forever confined to Earth will be decided in the twenty-first century amid the vast plains, rugged canyons and lofty mountains of Mars</em>  Arthur Clarke</p> </div>

 

[rogers_buck]

True in general, but when your talking about a high-power communications laser producing soliton-like pulse trains, I'm thinking there might be some overlap. A normal LIDAR rig a few years back used a q-switched ruby laser to produce a powerfull but short pulse. I'm thinking that a LiNb modulator on a redish solid state laser hitting an EDFA might have a usefull LIDAR mode (at high zenith angles anyway). The tradeoff would be to find a lambda that could do both jobs. Power and pulse width shouldn't be a problem.<br /> <br />

 

[JonClarke]

I just use the pretty pictures! <img src="/images/icons/smile.gif" /> <div class="Discussion_UserSignature"> <p><em>Whether we become a multi-planet species with unlimited horizons, or are forever confined to Earth will be decided in the twenty-first century amid the vast plains, rugged canyons and lofty mountains of Mars</em>  Arthur Clarke</p> </div>

 

[no_way]

Dragging an old thread back from dead, here's another point of view on humanoid robots in space:<br />http://robotgossip.blogspot.com/2006/08/robot-colonists-on-moon.html<br /><br /><blockquote><font class="small">In reply to:</font><hr /><p>The folks at NASA JPL Evolvable Hardware Laboratory have been working on robot companions for the next round of travel to the moon.<br /><br />They have reasoned that the best mechanical helpers would be humanoid robots. They feel that robots shaped like humans - with arms, legs and stereo vision - are the best partners for construction and everyday living on the moon and Mars.<p><hr /></p></p></blockquote><br /><br />Thats, from the guys who brought you MER. Straight from the horse's mouth, huh ?

 

[JonClarke]

Interesting article. But meanwhile Bill Clancey's mobile agents research is actually doing field trials using non-humanoid robots that actually might be useful in real situations. Although even Bill's robots are a long way from that as yet.<br /><br />I may be cycnical but this illustrates the difference in focus between those to whom humanoid robots are the goal and everything else, including exploring and developing the Moon and Mars is a means to that end, and those to whom working on the Moon and Mars is the goal and robots are one tool among many.<br /><br />Jon <div class="Discussion_UserSignature"> <p><em>Whether we become a multi-planet species with unlimited horizons, or are forever confined to Earth will be decided in the twenty-first century amid the vast plains, rugged canyons and lofty mountains of Mars</em>  Arthur Clarke</p> </div>