Can robots do it?

JonClarke · Apr 22, 2008

Here is a challenge for all those roboticists out there. A common claim is that robots are faster, cheaper, and better than humans for space exploration. Exactly what these robots are like is never specified, but they are apparently capable of extreme mobility, dexterity, adaptability, and flexibility, with near infinite power, immense sensor and analytical capabilities, and high levels of autonomy and creativity and able to resist all environmental extremes, all for quite modest mass requirements.  As such they appear to be good candidates for technology-indistinguishable-from-magic, and have more in common with robots in the imagination of Isaac Asimov than actual machines currently in service, being developed, or conceivable with foreseeable technology.In the real world robots are designed for specific missions with a particular set of capabilities.  They need power, have to conform to mass limits, and their control systems need to be protected from environmental extremes of temperature, pressure, radiation, dust, etc.  They carry specific tools and instruments and have set boundaries to their autonomy.  Budgets set their own constraints on what can be done.The claim that robots are faster, cheaper, and better than humans is often made specifically in connection with Mars missions. There have been extensive studies on the likely capabilities, power requirements, time frames, masses, and science and exploration capabilities of crewed Mars missions. There have been many studies on the requirements of human Mars missions, but very none that have produced the unmanned missions that can exceed what these crewed missions can deliver.So here is my challenge to the proponents of the idea that robots are faster, cheaper, and better than humans for exploring the Martian surface.  Design a mission (or series of missions) that can meet the following parameters:1 Mission parameters1.1 Likelihood of success – 97%   1.2 Distance1.2.1 2000 km of traverses1.2.2 100 km radius from landing site    1.3 Max depth drilled – 100 m   1.4 Total depth drilled 200 m   1.5 Mass returned samples – 400 kg 1.6 Mission duration1.6.1 18 months prime mission 1.6.2 operational life - 5 years Mars surface, 10 years in orbit1.7 Deploy 2 tonnes of scientific equipment on the Martian surface. 1.7 Discipline areas investigated 1.8.1 Geology1.8.2 Geophysics 1.8.3 Geochemistry 1.8.4 Mineralogy 1.8.5 Glaciology 1.8.6 Hydrology 1.8.7 Meteorology 1.8.8 Aeronomy 1.8.9 Astrobiology 1.8.10 Astronomy 1.8.11 ISR potential1.8.12 ISRU technology1.8.13 Field engineering 1.8.14 Space systems engineering 1.8.15 Materials science 1.8.16 Microbiology2 The conclusions must be expressed in terms of the following mission specifications:2.1 Number of spacecraft required to complete the mission   2.2 Mass to carry out the mission (both mass in Mars orbit and on the martian surface) with 20% margin   2.3 Power requirements (including power requirements in Mars orbit and on the martian surface) with 20% margin   2.4 Time required to complete the mission 3.0 Only technology currently in service, under development or at the laboratory testbed stage can be used.Note: the following research themes are commonly identified as being Mar of any mars missions but cannot be carried out withour human presence.1.8.17 Plant growth and physiology1.8.18 Physiology1.8.19 Partial gravity adaptation1.8.20 Radiation medicine 1.8.21 Nutrition and diet1.8.22 Human factors   Jon <div class="Discussion_UserSignature"> 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 Arthur Clarke </div>

qso1 · Apr 22, 2008

I was always the mixed bag type. I think humans and robotics ought to always be utilized, especially for Mars. The one thing about mars missions I would say is best done by humans by far, is the confirmation and analysis of potential life on mars. <div class="Discussion_UserSignature"> My borrowed quote for the time being:There are three kinds of people in life. Those who make it happen, those who watch it happen...and those who do not know what happened. </div>

JonClarke · Apr 22, 2008

Replying to: <DIV CLASS='Discussion_PostQuote'>I was always the mixed bag type. I think humans and robotics ought to always be utilized, especially for Mars. The one thing about mars missions I would say is best done by humans by far, is the confirmation and analysis of potential life on mars. Posted by qso1</DIV>Me too.  But that is not what I am asking here.  And I certainly don't want this to be another huans vs robots stoush.What I want is the robots only advocates to come up with a plausible set of missions that can achieve what humans can with probable technology, with mass, power requirements etc.  This has never been done.Jon <div class="Discussion_UserSignature"> 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 Arthur Clarke </div>

nimbus · Apr 23, 2008

Replying to: <DIV CLASS='Discussion_PostQuote'>Me too.  But that is not what I am asking here.  And I certainly don't want this to be another huans vs robots stoush.What I want is the robots only advocates to come up with a plausible set of missions that can achieve what humans can with probable technology, with mass, power requirements etc.  This has never been done.Jon Posted by jonclarke</DIV> Just my minimaly informed guess, but I don't think robots will compete with humans on any truly varied multi-purpose mission.  Not before AI makes some significant progress. <div class="Discussion_UserSignature"> </div>

JonClarke · Apr 23, 2008

Replying to: <DIV CLASS='Discussion_PostQuote'>Just my minimaly informed guess, but I don't think robots will compete with humans on any truly varied multi-purpose mission.  Not before AI makes some significant progress. Posted by nimbus</DIV>Perhaps.  For this exercise, only capabilities that exist in at least been tested in the laboratory can be used.  See point 3.0Jon <div class="Discussion_UserSignature"> 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 Arthur Clarke </div>

lildreamer · Apr 23, 2008

Replying to: <DIV CLASS='Discussion_PostQuote'>Perhaps.  For this exercise, only capabilities that exist in at least been tested in the laboratory can be used.  See point 3.0Jon Posted by jonclarke</DIV> Hi Jon - how detailed do you want this to go, is it best we handle each point seperately or combine various elements for simplicity sake? <div class="Discussion_UserSignature"> </div>

ThereIWas2 · Apr 23, 2008

One consideration is that robots do not breathe, eat, poop, get bored, or go nuts after months inspace.   The mass savings in not having to support those activities would be considerable.  The savings could go to smaller launch vehicles or more cargo. <div class="Discussion_UserSignature"> </div>

JonClarke · Apr 23, 2008

Replying to: <DIV CLASS='Discussion_PostQuote'>One consideration is that robots do not breathe, eat, poop, get bored, or go nuts after months inspace.   The mass savings in not having to support those activities would be considerable.  The savings could go to smaller launch vehicles or more cargo. Posted by ThereIWas2</DIV>True, but they still need to be powered, heated, cooled, protected from radiation, sealed against dust, equipped with sensors and instruments.  Theyhave to carry out diverse tasks - communication and control, science and engineering.  All this costs - time, power, mass. The question is, how much mass, power, etc. is required for umanned spaceecaft to achieve the mission specified in the OP?Jon <div class="Discussion_UserSignature"> 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 Arthur Clarke </div>

JonClarke · Apr 23, 2008

Replying to: <DIV CLASS='Discussion_PostQuote'>Hi Jon - how detailed do you want this to go, is it best we handle each point seperately or combine various elements for simplicity sake? Posted by lildreamer</DIV>This is a very good question.I am not sure, a forum is not a good way to do such studies in detail.  I have three agendas here.1) Overal to understand whether it is even possible to achieve the goals with a purely robotic misson2) More generally, to show that using purely robotic spacecraft to achieve the specified goals is not a trivial exercise.3) Specifically, get a ball park figure of what would be required in terms of mass, power etc. to achieve it.Perhaps the first thing we could do is look at the science goals, 1.8.  Are there items here which are redundant, can be combined, or are a priori impossible?cheersJon <div class="Discussion_UserSignature"> 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 Arthur Clarke </div>

danhezee · Apr 23, 2008

i would like to see a human mission to mars that lacked robotics.  robotics as in automated control system and also tele-presence. that can't be done. you should asked for a human-only mission too. lol. <div class="Discussion_UserSignature"> </div>

JonClarke · Apr 23, 2008

Replying to: <DIV CLASS='Discussion_PostQuote'>i would like to see a human mission to mars that lacked robotics.  robotics as in automated control system and also tele-presence. that can't be done. you should asked for a human-only mission too. lol. Posted by danhezee</DIV>It's a given that any human mission will use high level robotics in support.  Autopilots, automated ISRU plants, teleoperated rovers, robotic arms on pressurised rovers, etc.  Just like missions to the ISS, which use the most sophisticated space robots to date.But that is not the issue here.  Let's see whether or not it is possible for a purely robotic mission to achieve the specified mission, and if so, what would its requirements be in terms of mass, power etc.Lots of human (or to be accurate human + robot) missions have been designed.  Nobody has, to my kneowledge, tried to outline a mission that achieves the same thing with purely unmanned spacecraft.  So let's do it!Jon <div class="Discussion_UserSignature"> 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 Arthur Clarke </div>

danhezee · Apr 23, 2008

I see what you mean. I can’t help with mass or power requirements.  I am a software guy.  But I can tell you I feel that the advancements in collective intelligence research will make it possible for level of sophistication you want from your robots only mission.  If we plan a mission around the idea of a bunch of small simple cheap robots we can raise the likelihood of success since a robot can die while the others still work. Also a bunch of robots allow for concurrency with data gathering.  More data in a shorter time!!!! Unlike the current mars rover which are designed to “do it all”, so that means they can only “do it all” at one spot at a time.  Of course this idea will raise the power and mass requirements. <div class="Discussion_UserSignature"> </div>

danhezee · Apr 23, 2008

ohh another thing multiple robots means you are gathering data in all directions at once.  Linear distances wouldn’t be a good measure (2000 km), surface area would be more appropriate, km^2. Of course measuring in surface area also leads to the problem of detecting missed patches, but that is a software engineering problem<img src="http://sitelife.space.com/ver1.0/content/scripts/tinymce/plugins/emotions/images/smiley-smile.gif" border="0" alt="Smile" title="Smile" /> <div class="Discussion_UserSignature"> </div>

JonClarke · Apr 23, 2008

Hi Dan  Replying to: <DIV CLASS='Discussion_PostQuote'>I see what you mean. I can’t help with mass or power requirements. </DIV> That is a pity because most things in space exploration come down to basics like mass and power. Replying to: <DIV CLASS='Discussion_PostQuote'>I am a software guy.  </DIV> I’m only a geologist.  But everyone has something to contribute.  Replying to: <DIV CLASS='Discussion_PostQuote'>But I can tell you I feel that the advancements in collective intelligence research will make it possible for level of sophistication you want from your robots only mission. </DIV> We have to be careful relying on generalised promises of advancement in technology.  They are unpredictable (still waiting for the robot housecleaner I was promised a decade or so back). So for this mission we need to rely on technology that exists, at least in the laboratory.  See point 3.0 in the OP. Several questions arise from your comment that a software person like yourself might be able to answer. 1)      The mission will require a range of tasks, from spacecraft operations to exploration to running experiments. A) which of these can/should be run autonomously? B) Which of the autonomous tasks are possible with technology to hand or under development?2)      What tasks require collective “intelligence” and which don’t?3)      What level of hardware is needed to run such software.  My understanding is that unmanned spacecraft run on 286 equivalent chips.  The most sophisticated computers are the laptops on the ISS which are 486.  The Pentium age has not yet arrived in space.  One reason is the vulnerability of higher capability ICs to radiation.  To run Pentium level ICs in space would need shielding, which mass Feel free to restructure these questions from your perspective. Replying to: <DIV CLASS='Discussion_PostQuote'>If we plan a mission around the idea of a bunch of small simple cheap robots we can raise the likelihood of success since a robot can die while the others still work. Also a bunch of robots allow for concurrency with data gathering.  More data in a shorter time!!!! </DIV> I think it is given that a mission of such complexity would require a whole fleet of robots.  However  am not sure whether they will all be small and simple.  I find it hard to imagine how small and simple robots can drill 100 m drill holes, carry out 2000 kms of traverse, or do sophisticated geochemical analyses (which require complex sample preparation).   So the question for you is: which tasks can be performed my small, simple robots, working in cooperation, and which cannot? Also, how will such robots be powered? Replying to: <DIV CLASS='Discussion_PostQuote'>Unlike the current mars rover which are designed to “do it all”, so that means they can only “do it all” at one spot at a time.  Of course this idea will raise the power and mass requirements.</DIV> Actually the MERs can do very little.  They carry only 6 instruments (Pancam, MiniTES, RAT, Mossbauer, APX, and MI).  There are many things they can’t do that need to be done.  Collect samples, carry out sensitive, high precision analyses, do geophysical surveys, carry out meteorological and radiation obs, determine orientation of geological structures, look for biosignatures.  Etc.  In many ways they a approximate a reconnaissance rover.  The problem is they are painfully slow.  You would need hundreds to carry out just the scouting of the design mission. Maybe I need to draw up a list of tasks that need to be done?  Then we could see how they could be distributed amongst different robots. Cheers Jon   <div class="Discussion_UserSignature"> 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 Arthur Clarke </div>

JonClarke · Apr 23, 2008

Replying to: <DIV CLASS='Discussion_PostQuote'>ohh another thing multiple robots means you are gathering data in all directions at once.  Linear distances wouldn’t be a good measure (2000 km), surface area would be more appropriate, km^2. Of course measuring in surface area also leads to the problem of detecting missed patches, but that is a software engineering problem Posted by danhezee</DIV>This is a very important point, and a difficult one to answer!A lot depends on the assumed exploration method.  Will the explorers traverse to sites of interest, with only cursory examination along route?  or will they lawnmower the surface?  if the second, at what scale will the lawnmoving take place?  Km, 100 m, 10 m, 1 1, 10 cm?  Almost certainly both approached will be used.  Traverses to sites of interest, as the MERs do at present andd by explorers on the Moon and on Earth, past and present. Lawn mowing when investigating specific sites of interest with a particular method technology - grid geochemical sampling, grid drilling of an ice body, or geophysical surveying of a tight feature like a fossil spring or vent. But the number of sites and the types of survey carried out on each will depend on the site chosen and the nature of the features of interest.  Something we can't quantify at present.That is why I stuck to distance traversed, a measure that is used for the MERs, Lunakhod, Apollo, and desert and polar explorers on Earth.But if you could work out a meaningful measure of area to be put into the parameters, then great, I will inlcude it.I have thought about having an operational radius from a central point as an additional parameter.  Say 100 km. What do you think?cheersJon   <div class="Discussion_UserSignature"> 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 Arthur Clarke </div>

KosmicHero · Apr 23, 2008

Replying to: <DIV CLASS='Discussion_PostQuote'>This is a very important point, and a difficult one to answer!A lot depends on the assumed exploration method.  Will the explorers traverse to sites of interest, with only cursory examination along route?  or will they lawnmower the surface?  if the second, at what scale will the lawnmoving take place?  Km, 100 m, 10 m, 1 1, 10 cm?  Almost certainly both approached will be used.  Traverses to sites of interest, as the MERs do at present andd by explorers on the Moon and on Earth, past and present. Lawn mowing when investigating specific sites of interest with a particular method technology - grid geochemical sampling, grid drilling of an ice body, or geophysical surveying of a tight feature like a fossil spring or vent. But the number of sites and the types of survey carried out on each will depend on the site chosen and the nature of the features of interest.  Something we can't quantify at present.That is why I stuck to distance traversed, a measure that is used for the MERs, Lunakhod, Apollo, and desert and polar explorers on Earth.But if you could work out a meaningful measure of area to be put into the parameters, then great, I will inlcude it.I have thought about having an operational radius from a central point as an additional parameter.  Say 100 km. What do you think?cheersJon  Posted by jonclarke</DIV> I am usually a hard-core manned space kinda guy, but I like a challenge. The biggest challenge for a mission of this type is that robots aren't that sophisticated yet.  A large (greater than 2 or 3) fleet of robots working autonomously in tandem is beyond current technology.  It is also beyond the ability to remote control a fleet like this (an aerial platform would be best for cursory lawnmowing (i.e. mapping/surveying) but a Mars plane cannot be flown from Earth due to the time lag).  But I will have to contend with this requirement a little.  In every scientific/technological endeavor, there are things that have never been done, and answering these challenges require new tech development and innovation.  So I think a little leeway here would help our chances of making this work.  I would like to work on the systems in at least enough detail to make sense of them, but I feel, as you do, that a forum might not be the right place.  Mostly I think this is because, by the time anyone has one vehicle done the thread will be buried.  Regardless, I would like to attempt this.  Since my background is engineering, I'm not familiar with what the science requirements will entail (please advise).  Last point: spacecraft go crazy after several months in space too ... but there is usually nothing that can be done for them when it happens. <div class="Discussion_UserSignature"> kosmichero.wordpress.com </div>

JonClarke · Apr 24, 2008

Replying to: <DIV CLASS='Discussion_PostQuote'>I am usually a hard-core manned space kinda guy, but I like a challenge.</DIV>Me to, which is why I thought this might be a fun and usual exercise.Replying to: <DIV CLASS='Discussion_PostQuote'>The biggest challenge for a mission of this type is that robots aren't that sophisticated yet.  A large (greater than 2 or 3) fleet of robots working autonomously in tandem is beyond current technology.  It is also beyond the ability to remote control a fleet like this (an aerial platform would be best for cursory lawnmowing (i.e. mapping/surveying) but a Mars plane cannot be flown from Earth due to the time lag).  But I will have to contend with this requirement a little.  In every scientific/technological endeavor, there are things that have never been done, and answering these challenges require new tech development and innovation.  So I think a little leeway here would help our chances of making this work. </DIV>These are good points. So, if we are doing to allow leeway, how much to de allow?  We don’t want, through arm-waving, to allow technology indistinguishable from magic.Alternatively, maybe we should plan to have not more than 2 or three robots interacting at any one time.Replying to: <DIV CLASS='Discussion_PostQuote'>I would like to work on the systems in at least enough detail to make sense of them, but I feel, as you do, that a forum might not be the right place.  Mostly I think this is because, by the time anyone has one vehicle done the thread will be buried.  Regardless, I would like to attempt this.  Since my background is engineering, I'm not familiar with what the science requirements will entail (please advise). </DIV> I think your input would be of great value.  I suspect a lot of the mass would be what could be called engineering mass (structure, power, payload etc.) rather than the electronics. Place holder masses can be assigned to different science packages.As a minimum I see:At least one, probably several, large survey rovers that can carry out geophysical surveying, image the surface, characterise the surface, and document and collect the samples. Small robots to climb or descend cliffs, sampling and documenting the geology.  Also to enter caves.UAVs and aerobots.At least one drill rig, with 100 m capacity, along with logging and sampling tools, all vehicle mounted. These will require a lot of power, probably supplied by chemically stored energy (IC, batteries, fuel cells).  This means some kind of fixed ISPP plant and a power supply, and a way to deploy that power.Some type of fixed laboratory to analyse the samples.  Your guess is as good as mine as to how much this will mass and how much power it will require.  At least one, and probably several, fixed geophysical/astronomical/meteorological stations.  A spacecraft to load and carry the  400 kg of samples to orbit, and another one to return them to Earth.Satellite constellation providing imaging, communications, and navigation support.Perhaps you would like to have a crack at these.  Once we have rough estimates we can then work out how many we need to achieve the specified coverage in the given time.Replying to: <DIV CLASS='Discussion_PostQuote'>Last point: spacecraft go crazy after several months in space too ... but there is usually nothing that can be done for them when it happens. </DIV>This is another good point.  There will have to be some kind of self repair capability on missions so complex.  Is it possible, even with technology at the prototype stage, to repair and service robots autonomously?  Or would it need teleoperation?  Is such complex teleoperation feasible given the time lag?Remember that impossible, unlikely, and very difficult are acceptable answers to all questions in this discussion!CheersJon <div class="Discussion_UserSignature"> 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 Arthur Clarke </div>

KosmicHero · Apr 24, 2008

Concerning the need to avoid technology-indescernable-from-magic we could employ the 'technology readiness level' or TRL approach.  This is a scale from 1 to 9 that indicates how mature a technology is (1 indicates that the basic principles are observed and a 9 indicates that an actual system is 'flight proven' or essentially  off-the-shelf).  When doing near-term systems design that will have some component of technology development a 6 or 7 (prototype demonstrated in relevant environment) is considered feasible.From your list of science objectives I count a minimum of 4 systems: at least one rover to satisfy drilling, sample collection, and surveying requirements (more likely 2, although in 5 years a robust rover could cover a lot of territory especially if working in tandem to avoid obstacles with other platforms), a sample return vehicle, a science base, and at least one orbital and/or aerial platform.  This excludes cliffhangers and cave-divers, which are special cases that I don't think a rover or typical aerial platform could accomplish. So anywhere between 4 to 8 systems should be to accomplish the core objectives of this mission.  There are still a lot of question marks, like how these systems will operate together (or if they can), and of course, mass, power, time, questions.  Also, I'm not sure what you mean by your last science requirements 1.8.18 to 1.8.22.  These seem to be only accomplished with a human presence. <div class="Discussion_UserSignature"> kosmichero.wordpress.com </div>

JonClarke · Apr 24, 2008

Replying to: <DIV CLASS='Discussion_PostQuote'> Concerning the need to avoid technology-indescernable-from-magic we could employ the 'technology readiness level' or TRL approach.  This is a scale from 1 to 9 that indicates how mature a technology is (1 indicates that the basic principles are observed and a 9 indicates that an actual system is 'flight proven' or essentially  off-the-shelf).  When doing near-term systems design that will have some component of technology development a 6 or 7 (prototype demonstrated in relevant environment) is considered feasible. </DIV>   This is a good idea.   Replying to: <DIV CLASS='Discussion_PostQuote'> From your list of science objectives I count a minimum of 4 systems: at least one rover to satisfy drilling, sample collection, and surveying requirements (more likely 2, although in 5 years a robust rover could cover a lot of territory especially if working in tandem to avoid obstacles with other platforms), a sample return vehicle, a science base, and at least one orbital and/or aerial platform.  </DIV>   I generally agree.  But I do think the drill rig needs to be separated off.  Terrestrial rigs with 100 m capacity weight at last a tonne, sometimes more, depending on the method used.  It would need a separate vehicle to the other rovers.  You don’t need a rig on every vehicle either.  So 2 one tonne exploration rovers and one four tonne drill rig mass less than two four tonne combined rovers.   Replying to: <DIV CLASS='Discussion_PostQuote'> This excludes cliffhangers and cave-divers, which are special cases that I don't think a rover or typical aerial platform could accomplish. </DIV>I agree. Would envisage these as rub rovers or parasite robots on the main exploration rovers.  Analogous to the ROVs that manned submersibles sometimes deploy for use in cramped areas.  Replying to: <DIV CLASS='Discussion_PostQuote'> So anywhere between 4 to 8 systems should be to accomplish the core objectives of this mission.  There are still a lot of question marks, like how these systems will operate together (or if they can), and of course, mass, power, time, questions. </DIV>    The four to eight systems is not counting what is needed in orbit, of course.Do you think we can assign maturity to these technologies?   </DIV>Also, I'm not sure what you mean by your last science requirements 1.8.18 to 1.8.22.  These seem to be only accomplished with a human presence. </DIV>Well spotted.  One goal of Mars exploration is to assess its habitability.  This indeed requires human presence for at least some of the work needed.  So we have to make a decision.  Either we abandon this goal or must obtain the information by other means (e.g. through space station or lunar surface research).  Either way it shows that unmanned cannot do everything better than crewed ones.  Whether determining the habitability of Mars is a worthwhile goal for exploration is a subject beyond the scope of this thread!   CheersJon   <div class="Discussion_UserSignature"> 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 Arthur Clarke </div>

KosmicHero · Apr 24, 2008

Alright that sounds like a good starting point.  So redundant rovers (approx 1 tonne each), a drill rig (4 tonne), maybe a science station and sample return combined lander (I'll have to run some calculations in the morning to figure this one out), redundant aerial bots (either aircraft or balloon), and a small constellation of satellites. This actually starts to look like a 'multi-layered' system-of-systems a la missile defense.4 ground systems, 2 aerial systems, and 2-4 orbital systems.  If this sounds good, and if someone can come up with a list of instruments to satisfy the bulk of the science requirements, a better estimate of power and mass can be made.  <div class="Discussion_UserSignature"> kosmichero.wordpress.com </div>

holmec · Apr 25, 2008

Nice challenge.Personally I am also a mixed bag guy.  But I will endeavour to point out some differences between humans and robots so far. 1. Robots made for specific finite missions, humans are all-purpose assests.    This is a fundamental issue.  No robot launched has been all purpose.  That is give it a broad set of tools and it can perform a broad set of tasks of different skills.    Problem:  The Mars Rovers have exceeded their origional operational time on Mars.  So NASA was left to explore Mars with a finite set of tasks that could be performed.  If the rovers were designed to be all-purpose, then a lot more would have been learned over the years. 2. Robots can outlast humans    Humans need food and lodging and large amount of support while robots can withstand wind, sand, cold and heat for long period of time while only needing electricity from solar panels and sunlight. 3. Humans are sentient, not robots (at least not yet)    Robots so far have been an extension of human senses (telepresence).  I'm sure more technology for telepresence could be developed, like tactile simulation.  Humans can go to a remote planet and learn much, then write a book.  Robots give data and let humans make intelligable information out of that data.  4. Humans want to come back, robots don't have to come back to Earth    This is obvious. 5. Robots operate slower    Robots move slower.  Look how slow Spirit an Oportunity go.  Part of this is the long distance communication.  But it helps in the exploration process.  Slow is good when it comes to science.  Humans move faster and can think on the ground, but their time is limited and thus missions tend to be on the Rush Rush method. I hope this gives a little insight on the problem/challenge.  I have no doubt that better robots and better missions can be designed.  The real challenge is to think of everything. I really hate it when people say that robots are better than people or people are better than robots.  It shows ignorace, and I know people are smarter than that.  Robots are an extension of ourselves.  In space exploration we will continue to use robots and humans, its unavoidable.  <div class="Discussion_UserSignature"> "SCE to AUX" - John Aaron, curiosity pays off </div>

JonClarke · Apr 26, 2008

Replying to: <DIV CLASS='Discussion_PostQuote'> Alright that sounds like a good starting point.  So redundant rovers (approx 1 tonne each), a drill rig (4 tonne), maybe a science station and sample return combined lander (I'll have to run some calculations in the morning to figure this one out), redundant aerial bots (either aircraft or balloon), and a small constellation of satellites. This actually starts to look like a 'multi-layered' system-of-systems a la missile defense.4 ground systems, 2 aerial systems, and 2-4 orbital systems.  If this sounds good, and if someone can come up with a list of instruments to satisfy the bulk of the science requirements, a better estimate of power and mass can be made.  Posted by KosmicHero</DIV>Here is an instrument list.EXPLORATION ROVERS<ul><li>Thermal emission spectrometer </li><li>APX </li><li>Digital multispectral camera</li><li>Microscopic imager (colour)</li><li>Sun compass and clinometer </li><li>Brush </li><li>RAT</li><li>Sample corer</li><li>Meteorology suite</li><li>Ionising radiation suite</li><li>Videocamera</li><li>Laser Induced Breakdown Spectrometer (LIBS)</li><li>Magnetometer </li><li>Gravitometer </li><li>Electromagnetic sounding instrumentation </li><li>IP array</li><li>Resistivity array</li><li>Ground penetrating radar </li><li>Neutron spectrometer</li><li>Shallow (2 m) drill rig</li><li>Drill core sampler</li><li>Heat flow probe</li><li>Hyperectral hole logger</li><li>Geophysical logging tool (gamma, gamma-gamma, neutron, resistivty, IP)</li><li>down hole camera</li><li>swab sample collection tool </li></ul>It might need to be bigger than 1 tonne...  Any of these techniques not practicable with a robotic rover?DRILL ROVER<div class="O1"><ul><li><div>Digital multispectral camera</div></li><li><div>Meteorology suite</div></li><li><div>Ionising radiation suite</div></li><li><div>Solar radiation suite</div></li><li><div>Videocamera</div></li><li><div>100  m drill rig</div></li><li><div>Drill core sampler</div></li><li><div>Heat flow probe</div></li><li><div>Hyperspectral hole logger</div></li><li><div>Geophysical logging tool (gamma, gamma-gamma, neutron, resistivty, IP)</div></li><li><div>down hole camera </div></li></ul></div>Any of these techniques not practicable with a robotic rover?  GEOPHYSICAL STATION<ul><li>Meterology suite </li><li>Ionising radiation suite</li><li>Solar radiation suite</li><li>All sky day-lnght camera and spectrometer</li><li>Lidar</li><li>Radar</li><li>Magneto-telluric observatory</li><li>magnetic observatory</li><li>Seismic monitoring </li><li>Aerosonde facility </li></ul>Any of these techniques not practicable with a robotic station? LABORATORY STATION<div class="O1"><ul><li><div class="O2">Raman spectrometer</div></li><li><div class="O2">Mass spectrometer</div></li><li><div class="O2">Mossbauer spectrometer</div></li><li><div class="O2">IR spectrometer </div></li><li><div class="O2">Atomic force microscope</div></li><li><div class="O2">SEM microscope</div></li><li><div class="O2">Optical microscopes (polarised, bright and dark field, phase contrast, transmitted & reflected light)</div></li><li><div class="O2">XRF</div></li><li><div class="O2">XRD </div></li><li><div class="O2">Wet chemistry </div></li><li><div class="O2">Ice core analysis instrumentation </div></li><li><div class="O2">Oven for burning off volatiles </div></li><li><div class="O2">Microbial culturing </div></li><li><div class="O2">Basic rock preparation </div></li><li><div class="O2">Thin section, SEM sample, and biological slide preparation facility</div></li><li><div class="O2">Age-dating instrumentation </div></li><li><div class="O2">Organic suite</div></li><li><div class="O2">Metabolism detecting instrumentation </div></li><li><div class="O2">DNA fingerprinting</div></li><li><div class="O2">Mass spectometer for organics, ices, carbonates, and gases</div></li></ul><div class="O2"></div><div class="O2">Any of these techniques not practicable with a robotic station?</div><div class="O2"></div><div class="O2"></div><div class="O2"></div><div class="O2">I have also updated the mission requriments to reflect the discussion previously.</div><div class="O">Cheers</div><div class="O">Jon</div></div> <div class="Discussion_UserSignature"> 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 Arthur Clarke </div>

JonClarke · Apr 26, 2008

Replying to: <DIV CLASS='Discussion_PostQuote'> Posted by holmecNice challenge. </DIV>Thanks! And thanks too for a most insightful series of comments and reflections. Replying to: <DIV CLASS='Discussion_PostQuote'>Personally I am also a mixed bag guy.   But I will endeavour to point out some differences between humans and robots so far.  1. Robots made for specific finite missions, humans are all-purpose assests.    This is a fundamental issue.  No robot launched has been all purpose.  That is give it a broad set of tools and it can perform a broad set of tasks of different skills. Problem:  The Mars Rovers have exceeded their origional operational time on Mars.  So NASA was left to explore Mars with a finite set of tasks that could be performed.  If the rovers were designed to be all-purpose, then a lot more would have been learned over the years. </DIV>  I think this touches on several issues.  All robots have been finite purpose machines.  The more options you have the more complex the machine gets, the more power and mass it needs, almost geometrically.  MSL has twice the number of instruments as MER and masses more than four times as much.   So it is just more pratical to design robots for a few tasks. But of course that does limit the capabilities and long term potential, as you say.  It als means that observations cannot be improved upon.The closest general purpose robots to are machines like DEXTRE and Robonaut.  However, these are designed to work along side humans in human environments, using human tools, and are directly controlled by people.  They have no autonomous function.  Whether such general purpose machines are feasible in stand alone operations does not seem to be possible, given present or foreseeable technology.  Of course on mars the time lag kills the teloperation side, unless the people are in orbit. Replying to: <DIV CLASS='Discussion_PostQuote'>2. Robots can outlast humans    Humans need food and lodging and large amount of support while robots can withstand wind, sand, cold and heat for long period of time while only needing electricity from solar panels and sunlight. </DIV> Agreed.  So they are ideally suited for long term and mostly passive operations were only low mass is required.  Replying to: <DIV CLASS='Discussion_PostQuote'>3. Humans are sentient, not robots (at least not yet)      Robots so far have been an extension of human senses (telepresence).  I'm sure more technology for telepresence could be developed, like tactile simulation.  Humans can go to a remote planet and learn much, then write a book.  Robots give data and let humans make intelligable information out of that data.  </DIV> Robot sentience is, at this stage, technology indistinguishable from magic.  So we can't plan missions with it. Telepresence is very useful but it a complete substitute for direct presence.  Misssion systems are still much lower resolution than the human eye, and touch is still rudimentary.  Granted this could be surrounded (I will assume they can be for this discussion), but the time lag is still the killer for Mars. Replying to: <DIV CLASS='Discussion_PostQuote'>4. Humans want to come back, robots don't have to come back to Earth     This is obvious. </DIV> Of course!  But what about missions where the robots do come back?  Sample return for instance? These are staggeringly complex, difficult, and tend to be very costly.Replying to: <DIV CLASS='Discussion_PostQuote'>5. Robots operate slower     Robots move slower.  Look how slow Spirit an Oportunity go.  Part of this is the long distance communication.  But it helps in the exploration process.  Slow is good when it comes to science.  Humans move faster and can think on the ground, but their time is limited and thus missions tend to be on the Rush Rush method. </DIV> Agreed.  But it is worth noting that any Mars mission will be far less rushed than Apollo.  Contrast the tempo of a Spacelab mission to that of the ISS.  People take time off, there is time to repeat experiments, slip in news ones, cancel ones that are not relevant.  The same will be true on Mars (and on long term lunar missions.  [QUOTE]People will have - an must have - the time to think. I hope this gives a little insight on the problem/challenge.  I have no doubt that better robots and better missions can be designed.  The real challenge is to think of everything. I really hate it when people say that robots are better than people or people are better than robots.  It shows ignorace, and I know people are smarter than that.  Robots are an extension of ourselves.  In space exploration we will continue to use robots and humans, its unavoidable.  </DIV>  This certainly has been (and still is) my opinion too. Any specific comments on specific mission goals and parameters, instrument required etc.?  Cheers Jon   <div class="Discussion_UserSignature"> 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 Arthur Clarke </div>

danhezee · May 3, 2008

I think we can we use blimps for the flying platforms to save weight.   we can use the CO2 in the martian atmospere and seperate into carbon and O2. then pump up the blimps with O2.  That also allows us to develop a key ISRU technology for human missions in the future. <div class="Discussion_UserSignature"> </div>

kelvinzero · May 4, 2008

Hi Jon,I just wanted to point out this is a slightly unfair comparison.Landing humans on mars also involves some technology we havent developed yet. I have heard that mars is a difficult planet to land on with anything of greater mass than the current robots due to its faint atmosphere and significant gravity. This technology may not seem as 'SF' as a better robot but it will not be cheap.To compare a human and robot mission, it presupposes that the political will for an expensive one-off mission exists. It would be interesting if we could establish that a mission 10 times as expensive could perform much more than ten times the science, but clearly multiple nations have been able to budget for robotic missions, but none have approached sufficient enthusiasm for a manned mars mission. It is hard to say what a robotic mission with that sort of budget would look like. Some technologies absolutely cannot be tested in a mars mission without people, such as the effects of long term exposure to mars gravity. Rather than saying the cost of using robots to perform a human mission approaches infinity, it would be more fair to ask if performing half the science of a human mission could be performed at less than half the cost.

Can robots do it?

JonClarke

Guest

qso1

Guest

JonClarke

Guest

nimbus

Guest

JonClarke

Guest

lildreamer

Guest

ThereIWas2

Guest

JonClarke

Guest

JonClarke

Guest

danhezee

Guest

JonClarke

Guest

danhezee

Guest

danhezee

Guest

JonClarke

Guest

JonClarke

Guest

KosmicHero

Guest

JonClarke

Guest

KosmicHero

Guest

JonClarke

Guest

KosmicHero

Guest

holmec

Guest

JonClarke

Guest

JonClarke

Guest

danhezee

Guest

kelvinzero

Guest

Similar threads

TRENDING THREADS

Latest posts

Moderators online

Share this page