I agree. But not the robots we're currently sending, as great as they are.

__________________
"If this were play'd upon a stage now, I could condemn it as an improbable fiction."
Shakespeare, Twelfth Night
Illuminati's Razor-The most complicatedly evil answer is usually the most correct answer. - Fazor
"Every book is a children's book if the kid can read." - Mitch Hedberg
"Distance doesn’t matter much in space, where if you just start a thing off with the right kind of shove, sooner or later it will get where you want it to go." -Frederik Pohl, Mining the Oort

Now show what is wrong with my facts, reasonign, or calculations.

There will be improvements, but what is your evidence that they will improve by two or three orders of magnitude? people have been predicting such performance for 50 years. We don't haved it yet, and we don't know to get it. What if your hope is wrong?

Conversely we do know that with existing and developable technology will put people on Mars at no greater risk than we currently accept.

This is not supported by the evidence. First of all deep drilling is well beyond the capability of existing robots. Despite there being a lot of investment in this technology. Robotic vehicles that can traverse thousands of km are still a pipe dream, despite massive investment.

Robots can't do science. They can collect data, but the science is done at the end of a time delay. that is why, despite nearly 40 years of robotic developments still can't match the capabilities of the teloperated Lunakhods. The Lunakhods were more than 10 times faster than what is predicted for MSL.

Lastly robotic missions are more expensive per kg than human ones. Stationary landers cost about $1 million per kg on Mars, rovers $2 million. averaging everthing to payload sent to Mars robotic missions average $500,000 per kg, almost twice what human missions cost.

For large scale Mars exploration, it is cheaper, faster, and better to send people.

Jon

Which is truly unfortunate. If we plan to send men to Mars, we should start ASAP in looking for base and colony locations with robots. Once we get men on Mars, we can look for signs of life etc a lot more efficiently.

__________________
Reality: What a concept!……………………..><Ç(((Ç°>

Agreed. That’s why we need Goliath and Big Al.

__________________
Reality: What a concept!……………………..><Ç(((Ç°>

I believe the increase was comparative performance, as has been demonstrated over the past 30 years. While computer and robotic advancements are driven by literally hundreds of thousands of potential uses, rocketry advances are extremely limited in application. Since the trend in rocketry applications is unlikely to increase significantly, and since there is every reason to expect the trend in computers and robotics to continue to increase significantly, it is reasonable to conclude that robotics technology will continue to increase far more rapidly than rocket technology.

Only because deep drilling robots have nor had a need to be developed. But all the technologies to develop them already exist. They need only to be combined.

The military now has many robotic vehicles, such as R-Gator, Black Knight. Other vehicles such as Terra Max have shown that autonomous robotic vehicles are possible with current and emerging technology.

And there is no reason a sophisticated robotic rover could not do the same. With an onboard science lab, samples can potentially be subjected to hundreds of different scientific tests.

And Goliath is predicted to be 20 times faster than the Lunokhods.

Logically, robotic missions to Mars would be far less expensive than manned missions. Without all the life support required for humans, and without the return vehicle and fuel, robotic missions would be a LOT less expensive.

That depends upon when the exploration is to take place, and what the purpose of that exploration would be. Initially, large robotic rovers would be far more practical for finding, and perhaps even the initial preparation of colony sites. Later, after people are living on Mars permanently, and can take time away from merely surviving, THEN manned expeditions across the Martian surface will be more practical. IMEHO

__________________
Reality: What a concept!……………………..><Ç(((Ç°>

I believe the increase was comparative performance, as has been demonstrated over the past 30 years. While computer and robotic advancements are driven by literally hundreds of thousands of potential uses, rocketry advances are extremely limited in application. Since the trend in rocketry applications is unlikely to increase significantly, and since there is every reason to expect the trend in computers and robotics to continue to increase significantly, it is reasonable to conclude that robotics technology will continue to increase far more rapidly than rocket technology. [/quote]

Impressive though the advances in robotics have been they have not translated into impressive progress in terms of field robotics on the surface of the Moon or Mars. MSL the most advanced planetary rover so far developed, will have less exploration capability than the Lunakhods 40 years before.

Rocketry does not need to advance to get people to Mars. The technology of today (and 30 years ago), is more than adequate.

Conversely the last 30 years have seen major advances in most of the other space flight necessary for humans to go to Mars. Long duration human spaceflight, management of multi year missions, large scale solar power production, orbital assembly, high efficiency life support, knowledge of the surface of and atmosphere of Mars. The main areas of development pending are EDL, rovers and suits. None of these require break though science, simply application and development of what we now know.

There has been extensive research on such technology. It has failed to deliver anything reliable beyond a couple of metres. Which is a pity, because there have been times when i could have used such a rig. Whereas a simple human operated rig can drill hundred metres.

Terra Max crashed spectacularly as I recall. Not exactly a useful system as yet.

In autonomous mode R-gator carries out simple repetitive tasks, not complex exploration. It can move at only a quarter the speed it can be driven by a human operator. It can run for only 8 hours and needs for than a kW to run its electrical system.

Furthermore it operates in close cooperation with people. In fact, it is an excellent example of the type of robot vehicle that would support a human mission as an unpressurised rover with some specialised autonomous capability. it's not a question of humans or robots, but using both to their best ability.

What is the evidence that sophisticated robots will be able to do sophisticated tests? You will always need scientists in the loop deciding which sites need to be visited, what samples should be collected and which tests should be done on them. With the time lag this will always be slower than with people on site. Even advanced robots can’t circumvent the speed of light.

I hope you are not trying to highjack the discussion onto your hobby horse.

This is not logic, this assertion. It is a fact that robotic missions cost twice as much per kg as human ones. For example:

MGS cost $538 million and massed 1,031 kg in LEO, and therefore has an LEO cost per kg of $522,000

MO cost $348 million and massed 725 kg in LEO, and therefore has an LEO cost per kg of $480,000

MRO cost $720 million and massed 1,180 kg in LEO, and therefore has an LEO cost per kg of $610,000

Phoenix cost $386 million and massed 680 kg in LEO, and therefore has an LEO cost per kg of$567,000

The combined MER cost was $900 million and they massed 2,126 kg in LEO, and therefore had an LEO cost per kg $423,000

MSL has an estimated cost of $1.7 billion and a LEO mass of 3,400 kg in LEO, and therefore has an LEO cost per kg $500,000

(All costs adjusted for inflation to 2006-2007 $)

Average cost per in LEO (exclusive of the EDS) is therefore $500,333.

For comparison the average per kg cost of the ISS and the entire Apollo program is $235,955, just over half the cost of unmanned missions.

This is what the current generation of rovers is doing, in part. Testing the ground and preparing the way for future crewed missions.

Jon

That is no reflection on the possible use of robotics, but rather a reflection on how the PTB chose to apply the technology. Your argument does not address the practicality of robotics for Mars missions.

Adequate, yes. But isn’t the inefficiency (and therefore cost) of current rocketry often cited as one of the main reasons we are not already on Mars?

I have been supporting that viewpoint for years. But it is irrelevant to this question because the same can be said for robotics. We can build them. We have the technology. We simply need to apply it properly.

Goliath could carry a rig that could drill 20 feet in a single pass. The only restriction on drilling deeper is robotically adding sections of drill pipe. We have a machine here that bores 50-100 ft under streets to lay cable. The only thing the human operator does is drop another section of pipe onto a tray. The machine does the rest. A robot could drop a pipe onto a tray.

Irrelevant. Some of our most sophisticated and reliable technology had early failures, some of them quite spectacular.

True. However, the people need to sleep and rest, while the robot can work 24 hours a day. I’d say that more than makes up for the time lag.

That is perhaps one of the most representative examples of the Logical Fallacies of Hasty Generalization and Misapplied Generalization I have seen. You have taken a variety of programs spanning 45 years, with incredible changes in technology, major changes in the driving political and economic climates, and many different types of delivery systems. That isn’t apples and oranges, that is apples and quartz and squid. Adjusting the dollars is irrelevant, because the other variables are so incredibly disparate. The comparison should be the difference between manned and robotic missions using the same base technology in the same time period.

Logically, robotic missions to Mars would be far less expensive than manned missions. The food, water, oxygen, and living facilities alone would make human missions more expensive. If you leave the crew on Mars, they have to have huge amounts of supplies. If you bring them back, you have to have a return vehicle, more supplies, and fuel. That makes the manned mission even MORE expensive.

Hardly. None of the current generation of rovers is going to verify a settlement site, which will be the most difficult and most critical part of early exploration. There is very little more that we need to know before simply sending a flags and footprints mission to Mars.

__________________
Reality: What a concept!……………………..><Ç(((Ç°>

There are two main issues here:

(1) Putting humans on Mars just for the sake of doing it, and/or for colonization, etc.

(2) Scientific study of Mars by robots vs human astronauts.

For the former, the relative science benefit and pros/cons of humans vs robots is less a factor. Humans are going there anyway in this scenario.

For the latter, the issue is how to maximize the science return per invested dollar, independent of whether humans go to Mars.

It seems clear robots are far more cost efficient than humans if science results per dollar is the primary goal.

We must remember little actual science will be done on Mars by either humans OR robots. Rather the data is captured and most of the actual scientific work is done on earth.

Humans are basically fragile biological robots in that role. On orders from earth, they deploy instruments, turn knobs, throw switches, take samples, etc. Robots can do similar things without the vast overhead needed to support humans on Mars.

The projected timeframe must also be specified in any comparison of human vs robot science return per dollar. This is because of the continuing rapid progress in computer and robotic technology. IOW, a recent time estimate for a realistic manned Mars mission is 2031 or 2037:
http://news.bbc.co.uk/2/hi/science/nature/7116834.stm
http://afp.google.com/article/ALeqM5...uI4rNSCpCl3TTQ
http://www.nasaspaceflight.com/content/?cid=5048

We know what humans will be like in 2037: about like today. Rockets will be somewhat similar. By contrast robotics and computers progress very quickly and will be much more advanced. Already IBM's Blue Gene/L has simulated half a mouse brain: http://news.bbc.co.uk/1/hi/technology/6600965.stm. Blue Gene/P is about 6x as fast: http://en.wikipedia.org/wiki/Blue_Gene. We obviously won't be sending a supercomputer to Mars, but what supercomputers do today, a smaller computer can do in a few years. There are even informed projections that computers may mimic aspects of human intellect by the 2030s: http://www.jetpress.org/volume1/moravec.htm

On the robotic front, Honda's Asimo gives some idea of where things are headed. Anyone who thinks "robots couldn't do this or that on Mars" should examine this closely:
http://video.google.com/videoplay?do...267825427&q=as

Now, there are various videos showing Asimo falling down and malfunctioning. But he's a prototype -- not remotely intended for production use. The key is NOT what can he do today, but at the current rate of computer and robotic progress, what could his successors do in 2037?

Re deep drilling, innovative robotic methods are now being developed for use on various planetary bodies, including Mars: http://www.honeybeerobotics.com/idds.html

Re the limitations of CURRENT unmanned terrestrial vehicles, this has little to do with an advanced Mars rover in 2037, for several reasons:

(1) Progress is rapid in computers and robotics. Limitations or failures today mean little for 2037.
(2) Terrestrial unmanned vehicle trials such as the DARPA Grand Challenge (http://en.wikipedia.org/wiki/DARPA_Grand_Challenge) are totally autonomous. The ground rules don't allow human intervention -- at all. By contrast a Mars rover is under earth control, using automation for limited periods. If it encounters problems it just waits for assistance. Of course by 2037, a Mars rover might need little assistance.

Re a manned mission being able to take more data, more images, etc, that's simply not true. An example of this is the Mars Reconnaissance Orbiter. It takes vast numbers of high-resolution images. Putting a human in Martian orbit to control the imaging would not improve things.

Whether in orbit or on Mars' surface, robots can stay there so long and take such high quality data precisely because humans ARE NOT there. The payload isn't squandered on humans and their support needs but can be dedicated to the scientific mission.

Putting a human on the surface would have some benefit in gathering data. But it is titanically expensive. For the cost of a manned mission (and probably at a fraction of that), a series of unmanned missions could gather far more and higher quality data.

We have already transitioned to mostly unmanned robotic exploration of the deep ocean. Decades ago, we'd send a manned deep submersible. We no longer do that, rather we use unmanned remote vehicles. It's not as exciting, but it's cheaper, safer, and we get excellent results. Likewise there's no serious plan for manned exploration of Venus, Mercury, Jupiter, etc. The entire solar system will be explored exclusively by robots, with the exception of earth's moon and Mars. We could probably put a human on Mercury or Venus, it would just be incredibly expensive. But we get better science return per dollar by using robots. The same is true for Mars.

Well said, joema. Thank you for all the great information.

That Asimo video is almost spooky to watch. It makes you think there must be a person inside that suit.

__________________
Reality: What a concept!……………………..><Ç(((Ç°>

Evidence? Certainly robotic missions can collect some data about Mars more cheaply than human missions. But for detailed scientific studies of Mars or the Moon human presence is far superior.
We need to send people… But I believe firmly that the best, the most comprehensive, the most successful exploration will be done by humans. Steve Squyres Discover magazine interview June 2004
http://www.discover.com/issues/jun-0...over-dialogue/
We are constantly aggravated by the fact that all the questions we have about Mars could now be answered by Ken and me if we could just walk around on the planet for a few days….But for about two years now Malin and I have been absolutely convinced that we're going to have to send people there. Ken Edgett National Geographic, February 2001.
…human exploration coupled with sample return unquestioningly results in the most profound scientific understanding of a planet... they are essential. ...the "New View" of the Moon that has resulted from the entire range of exploratory tools and cross-disciplinary studies has been eye-opening. It points compellingly to a model for how future planetary exploration should be conducted. Because we now know how to do it right: what kinds of missions and (in hindsight) in what order…. There is no substitute for the ultimate mobile sensor: a human... if the Apollo experience taught us anything it is that the human ability to recognize interesting features quickly and then independently act to follow up on that information can lead to important discoveries... Ross Taylor et al., Reviews in Mineralogy and Geochemistry 60, 657-704, Mineralogical Society of America, 2006.
Robots do no science at all, they just collect observations. For field geology and biology they collect that data slowly, clumsily, and inefficiently compared to what human crews could do. This is a documented fact from the Apollo missions and from terrestrial comparisons. Despite 30 years of advances autonomous robots still cannot do as well as teleoperated machines either on other planets or on Earth. Another 30 years might bring autonomous performance up to teleoperated levels. It might not. Even then they will not perform as well as humans on the spot because of the time lag between the sensors and the people actually doing the work. Having humans on the spot will not only collect better data faster but allow on the spot science to be done, as well as much better science back home because of more and better quality data, and larger, better selected and contexturalised samples.

Robots have limited dexterity, mobility, dexterity, adaptability to unexpected surfaces compared even to a gloved astronaut. One reason why the robotic Hubble servicing mission was dropped in favour of a human one was because it would do less with a lower probability of success.

How big an overhead will a human mission need? More than a robotic mission certainly. But robotic missions have huge overheads too. About 4000 people are or have been involved in the MER missions. A human mars mission might have 10 times or 100 times as many people working on it. But it will be at least a thousand times as productive.

We know what humans will be like in 2037: about like today. Rockets will be somewhat similar. [/quote]

Spacefaring will be considerably advanced in 30 years time over what it is at present. We don’t need more advanced rockets than we have already to get to Mars.

All very interesting, but what they mean for actual robot performance on the surface of Mars is just speculation.

Mouse brains don’t do science. A computer of Blue Gene performance would enhance the performance of a human crew even more that it would a robotic mission. In fact computers of such capability would generally require a protected environment to operate – like a human habitat. Fast computer speed is good, but it does not do science, only provide a better tool for people to use.

Predictions of mimicking the human intellect have been made before, so far with little to show for it. And what does “mimic” actually mean? AI research has given us search engines which mimic the human intellect’s ability to search and correlate information but much faster. But it still does not mean your PC does research. What it does is give better tools for people to use.

Completely irrelevant to Mars exploration. No payload. Can only move on flat surfaces. Needs near 400 watts to run. Has an endurance for less than an hour. Such a machine is useless on Mars.

His successors will be doing well if they could mimic what MER can do now. The MERS are much more power efficient, have unlimited endurance, diverse payload, can cross wide ranges of surfaces – rocks, steep slopes, sand. A much better question will be what MERs successors be capable off in 2037? IMHO, based on extrapolation past progress and current options, is that they might be able to mimic the performance of a teleoperated rover on the Earth of the Moon. But they will still be well below what a human could do driving the same machine.

A contract to research technology does not equate with such technology being actually feasible. It will certainly be very slow. So far the ability to drill even 10s of metres has been unrealised. Two m is about the limit for robot technology. Whereas a trailer mounted rig operated by a human crew can drill 10’s of metres in a few hours.

Progress is indeed to expected. Experimental technology now indicates what operational technology may be capable of it 10-20 year’s time. It might be another 10 years before such technology is space qualified. So hopefully vehicles with autonomous speeds measured in kilometres rather than metres might be practical in 30 years. They might not too. Driverless cars have been actively researched for 30 years, and are still not operational.

Even if this technology is available in 30 years, such vehicles will still need to consult with mission control on important decisions, which immediately inserts 10-40 minute delays even for quick decisions. A decision that would a person on the spot 5 seconds would take 10 to 40 minutes Plus every science operation will require constant supervision from Earth, and that is where the time delay becomes crucial. Operations that would take an astronaut a few minutes would take days.

Autonomous navigation will be very useful for many applications, not least for direct human exploration. But exploration is a lot more than being able to navigate a route. I can do that without much conscious thought. Exploration is an activity that is orders of magnitude beyond simple driving. It requires interpretation, understanding, knowledge, and experience, and something which will always be better done by a person on the spot than with a time lag.

MRO is an excellent example of a mission that is ideally performed by an unmanned spacecraft. This is not the sort of mission I or anyone else who has written on Mars exploration has advocated in the past 30 years. We are talking about surface missions.

The point is the MERs don’t get high quality data. The handful of instruments they carry are roughly equivalent to what an astronaut would carry on a pedestrian EVA to do a quick characterisation of the samples and their context. Compared to a pedestrian EVA they cover only 5-8 metres a day, instead of several km. Not that you could send people to Mars simply to do pedestrian EVAs. They would have unpressurised rovers which could cover tens of km a day, carrying more instruments than MSL. They would have pressurised rovers that could explore hundreds of km from the landing site. They would take samples back to the station and run chemical analyses that include trace elements, rather than just the majors. They could process large samples to get truly representative measurements. They could cut thin sections of the samples for petrographic work. They could run thousands of samples to chose the best samples to bring home – and bring back hundreds of kg. The human payload is not squandered but invaluable.

A human mission to Mars will not be titanically expensive. Using per kg costs and reasonable mass estimates a human mission would cost 70 billion. Spread over a 10 years development period this is the cost that is currently being spent on human missions. Unmanned missions are roughly twice as expensive per kg. I have already shown good reason why this is the case.

In the politically unlikely event of similar moneys being spent on an unmanned, 70 billion might buy you 10 large rovers and sample return missions. These would achieve far less than a single human mission of the type described. They might cover between them several hundred km and return a few kg. The human mission would explore thousands of km and return hundreds of kg. Plus those samples would be characterised by people on the spot with a vast array of instruments, not working through a 40 minute time delay with a limited suite of tools. Plus the human mission would investigate a whole range of issues regarding human settlement potential which no unmanned mission can ever do.

ROVs and AUVs are enormously useful machines that greatly expand our capabilities to explore the oceans. But they have not replaced saturation divers, hard suits, surface ships, nuclear submarines, submersibles, they complement them. Furthermore most underwater robots are in fact teleoperated, taking advantage of real time human decision making and judgement. AUVs do very simple tasks. Teleoperation is going to be very important on Mars, as an adjunct to human exploration, not as an alternative.

As for crewed missions to Mercury of Venus, I agree there is no reason (at present) for human missions there. I don’t know anyone who is seriously suggesting them as a goal. These are missions indeed best served by robots. The Moon, Mars, and near earth asteroids are different. They can be reached using present day or near present day technology. Their environments are relatively benign and full advantage can be taken of relatively unimpeded human presence in every respect.

Jon