Rover Navigation - How should it be done?

Phobos · 2002-06-22 14:47:58

What would be the best ways for a long range rover to navigate about? I was thinking of maybe setting up two very tall towers some distance from each other that would have a beacon attached so the rover would be able to triangulate it's position and record with some accuracy points of interest to return to. I guess the ultimate would be to have some kind of Martian GPS system but that probably won't be in the budget for awhile. Anyways, with such navigation systems the rover could do a lot of the driving itself. When you want to go home just press a button and let the rover do all the driving. It could also store areas in it's computer bank that the crew could just bring up and have the rover automatically drive to.

Byron · 2002-06-22 16:19:10

I think GPS will be pretty much the only way to go on Mars... From the first landing on, I think it'll be necessary to have at least 3 satellites in orbit for communication and navigation purposes.

A land-based nav system could work without the use of towers, landmarks, or whatever, as all you need is an accurate chronometer, a comprehensive atlas of Mars, and a real-time fix on at least two of the objects that cross Mars' sky daily..i.e., Phobos , Deimos, and the Sun. With this info in hand, it'd be a snap to determine your location almost as accurately as GPS.

But nothing beats GPS when it comes to navigation, here on Earth and on Mars, and I think putting the satellites in orbit should be a priority even before we land on Mars.

B

Pat Galea · 2002-06-23 02:22:28

I think GPS will be pretty much the only way to go on Mars... From the first landing on, I think it'll be necessary to have at least 3 satellites in orbit for communication and navigation purposes.
A land-based nav system could work without the use of towers, landmarks, or whatever, as all you need is an accurate chronometer, a comprehensive atlas of Mars, and a real-time fix on at least two of the objects that cross Mars' sky daily..i.e., Phobos , Deimos, and the Sun. With this info in hand, it'd be a snap to determine your location almost as accurately as GPS.
But nothing beats GPS when it comes to navigation, here on Earth and on Mars, and I think putting the satellites in orbit should be a priority even before we land on Mars.
B

Yeah, GPS is definitely the best way.

But given that Mars is pretty much deserted, there shouldn't be much problem just scattering the surface with land based transmitters. Drop 'em all over the place, let each one work out exactly where it is (using some other means), and then braodcast a GPS-style signal.

They don't have to be perfect, and they don't have to last forever. Just enough to keep everyone from getting lost until the real GPS is set up.

Phobos · 2002-06-23 11:50:58

With all of these probes NASA plans to send to Mars in upcoming years, I wonder if it would be possible for them to design the orbiters not only to do their primary science functions, but also function as GPS satellites. It might be a cheap way to go about installing a GPS system around Mars since you wouldn't need extra launches. Once the satellite completes its primary task it could just be moved into the appropriate location.

Gibbon · 2002-09-01 05:10:16

Why not launch the satellites now? There are thousands or Russian ICBMs that are available for the launching of small satellites. The satellites would have an ION engine and would have about 20 years to reach Mars. I think that the cost of an IRIDIUM satellite was only a few million dollars. A launch is about a million. I think that you could have a very good, budget, GPS and communication system for around 100 million US (Iridium put up HEAPS of satellites and spent about 5 billion US, but this would only need them on about half the planet, which is smaller anyway). It would be a good investment because they could also be used for sending data from the surface to an orbiting station or even Earth. Of course, relay satellites would have to be put in place.

turbo · 2002-09-01 09:35:22

What ever happened to all the commsats that Motorola was putting up? I know the project was cancelled beause technology for cellular went ahead. If some of those sats are assembled and ready for launch, maybe they could be bought less expensively than the IRIDIUMs.

Russian ICBM boosters would be fine, although I still think getting ahold of the old Titan boosters would be great.

turbo

Phobos · 2002-09-01 17:45:27

Are GPS satellites small enough that they could piggyback as secondary payloads on rocket launches? Considering that there's an amatuer satellite group that plans to send a payload to Mars like this I wonder if Martian GPS satellites could be launched in a similiar manner and save the cost of independent launches.

Gibbon · 2002-09-01 23:53:42

IRIDIUM went bankrupt and was purchased by GlobalStar for about 200 million US. Considering it cost about 5 billion to set up, this is a considerable saving.
Getting a Titan booster would allow you to launch more satellites at once. The payload area on an ICBM is only about 2 cubic metres.
http://www.astronautix.com/lvfam/makeyev.htm
this website lists the ex-Soviet missiles that are now available for purchase (minus nuclear warheads). The planetary society used an ICBM to launch their prototype solar-sailor into orbit.

Phobos · 2002-09-02 16:59:05

The Planetary Society has already launched their solar sail? I was trying to dig up info on that not to long ago but none of it was current. Anyhow, I never thought about trying to launch all of the satellites at once. If the satellites are small enough that would be the way to do it. The only problem with the satellites themselves might be their life spans. Once they run out of fuel to correct their orbits we'd have to replace them which might make a GPS system impracticle unless I'm missing something.

turbo · 2002-09-02 17:24:42

I dunno Phobos, how big does a GPS sat have to be? I had to do a presentation to my Advanced Communications class about geosynch sats, and the altitude on those are like 22,000+ miles. I don't know if current photovoltaic cells could provide power in a synchronous Mars orbit for a GPS sat. ???

I'll look again, but I doubt my textbook says anything about
how often a geosynch sat needs orbit correction.

turbo

C.COMMARMOND (FR) · 2002-09-03 14:49:35

Hello,

GPS satellites are big: 1816Kg and are on 20200 km circular orbit (12 hours) inclined at 55 or 63? and need 710 W, they were designed for 7.5 years life time, but most of those launches in 90 are always working.

If we want to use equivalent parameters around Mars, they should be on a lower orbit (7000 km?) to achieve the same 12 hours orbit, so they should need less power (since I don't know how much power the emettors uses, I will just guess near to 400W, so the same solar panels could be enough).

The Earth GPS use 24 satelites, but I think a minimum of 6 is needed for a minimum service. The problem with GPS is it need several ground stations to work...

Common life span of satellites around Earth is 7/12 years but it should be less around Mars since there is no Van Allen belt around Mars so maybe we need more shielding.

I think NASA is planning to send some GPS (or equivalent systems) to Mars with their internet network.

Conclusion: if NASA doesn't send a GPS like system around Mars, we should have to set one, but because of the complex design of the Earth GPS, it could be needed to design a simpler (less accurate) system to reduce costs (satelites are too heavy and need several ground stations). I guess synchronisation based on far star (or sun) position with 'top' transmitted from one to the others satelites could be less expensive and use of discontinuous localization (all the sats send a one second burst every 5 or 10 seconds) could dramatically reduce power needed while maintaining a good accuracy for 'slow' vehicles. All this to try to obtain the same weigh than a globalstar (400 Kg) and a low power consumption.

CC

Infos on rockets, constellations, companies can be found on:
www.spaceandtech.com

Phobos · 2002-09-03 19:50:49

Ouch! I had no idea gps satellites were that big! I just kind of jumped to the conclusion that they would be small since they just emit signals, and to my knowledge, don't have anything like cameras or extensive instrument packages that would require such huge sizes. But knowing the military, no telling what they've attached to those things.

C.COMMARMOND (FR) · 2002-09-05 12:46:13

Phobos, the problem with emitting satellites is the smaller the recepter (the pocket GPS), the bigger the emettor (sat)...

And to get 'less' satelites, have to be higher in space so the square root acts to reduce again your efficiency so you need to increase power.

But, all is not lost, the ideas I gave could save GPS on Mars...

CC

Phobos · 2002-09-05 21:39:49

Phobos, the problem with emitting satellites is the smaller the recepter (the pocket GPS), the bigger the emettor (sat)...
And to get 'less' satelites, have to be higher in space so the square root acts to reduce again your efficiency so you need to increase power.
But, all is not lost, the ideas I gave could save GPS on Mars...
CC

Considering that gps satellites would be closer to the surface of Mars than they are on Earth I wonder how much smaller you could make them? Do you think there would be big difference in size?

Pat Galea · 2002-09-06 03:55:11

Redundancy is the key when you're pioneering somewhere blatantly hostile and uncivilised like Mars.

What I need is a pocket navigator unit with the following features:

* Maps.
* GPS.
* Land-based triangulation.
* Inertial navigation (accelerometer).
* Accurate chronometer.
* Sextant.

It'll use the GPS signal first. If that fails, then it'll pick up the land-based signals. If that fails, then it'll rely on its inertial system (calculating position from the last confirmed fix; very useful when mucking about in tunnels).

As a last resort, it'll use the optical pickup, suggesting various objects (Sun, Deimos, Phobos, stars) for you to look at. Using these fixes with the chronometer, it'll work out where you are, and rely on the inertial system from that point on until the accumulated error gets too big. Then it'll ask for more position fixes.

Whenever it gets a GPS or land signal, it'll correct the chronometer too. In fact, if it's really clever, it'll work out the chronometer error and use this information to correct itself when it's running solo.

Should be in the shops by Christmas.

Phobos · 2002-09-06 20:11:35

Somehow I'd probably still get lost even if I had all of that navigating power in front of me. I wonder how valuable Mars's moons would be though as navigation aids. I don't think I've ever heard of anyone using Earth's moon to navigate by, not saying it's impossible, just something new to me. Anyhow Mars's satellites move across the sky quickly, that might make it difficult to get accurate navigation data from them.

Phobos · 2002-09-06 20:53:46

Place Holder

RobS · 2002-09-07 00:39:48

If you had an ephemeris on your rover's computer giving Phobos rise and set times--and it rises and sets about three times a day--then you could determine where you are, in that when Phobos rose or set you'd have to be in a narrow band running along the Martian surface. You'd need to time sunrise or sunset, or a star, or Deimos, to triangulate your position more exactly.

But the easier solution for navigation would be to place about six GPS (global positioning system) satellites in Mars orbit. By the time people get to Mars, such a system could pinpoint your location with a centimeter or two. It would probably be accurate enough to allow automated driving, where the rover's computer would have an exact terrain map of the route (preferrably one that has already been bulldozed and photographed from a vehicle) and would use GPS to determine exactly where it was on the route. The vehicle would not have to worry about tumbleweeds, deer, or stray children running in front of it, so it probably could trundle on autonomously while the "driver" and passenger watched television or played cards (or did something useful).

--RobS

Pat Galea · 2002-09-07 00:40:40

I wonder how valuable Mars's moons would be though as navigation aids. I don't think I've ever heard of anyone using Earth's moon to navigate by, not saying it's impossible, just something new to me. Anyhow Mars's satellites move across the sky quickly, that might make it difficult to get accurate navigation data from them.

The points used just have to be easy to find in the sky. As long as the computer knows exactly where they'll be at any particular time, then they'll do.

In fact, thinking about it a bit more now, the fact that they are fast moving probably helps to make the navigation more accurate. The position fix changes rapidly with time, so errors show up easier. (I haven't formally proven that, but some quick back-of-an-envelope sketches suggest that'll be true.)

I don't know off-hand how big they appear from the surface of Mars. They would, of course, have to be point-like in order to be useful for nav.

-

EDIT: Added last para about 'point-like' requirement.

Shaun Barrett · 2002-09-07 01:58:45

A quick Google search revealed that Phobos, from Mars, is as bright as Venus is from Earth, and only visible from the surface at latitudes lower than 69 degrees.
Deimos, from Mars, is as bright as Sirius, and visible at latitudes as high as 82 degrees.

Josh Cryer · 2002-09-08 01:24:49

Hmm, would Phobos be visible in the daytime?

C.COMMARMOND (FR) · 2002-09-13 12:36:43

Hello,

why to speak about what could not work (GPS, inertial, satellites...)?

If we put six GPS sats around Mars, we should put telescopes too inside for meteo and other needs and radio transmettors and relays.

If GPS fails, it exists micro accelerometers that a small processor can control to extract exact position. It weight only grams...

If those two systems fails, I think this means you already ar in big troubles... But if this happen, just turn you radio set on and cry...

CC

Bill White · 2002-09-13 12:43:30

Celestial navigation will still work every night - like the sailors in centuries past. Unless there are dust storms, the stars should be quite clearly seen.

Every rover will need a good sextant, ot two, and unless you drive a rover across one, it should not break.

I had a professor who navigated B-17s in WW2. He said they took star shots from the bubble canopy on the top of the plane. Phobos/Deimos might be nice but they are far from necessary.

kalizarin · 2002-09-26 21:19:08

I think GPS will be pretty much the only way to go on Mars... From the first landing on, I think it'll be necessary to have at least 3 satellites in orbit for communication and navigation purposes.

This is particularly true when it comes to communication. Posting comm towers all over Mars is infeasible. Mars has the same land area as the Earth (since much of the Earth is covered by water). Worse, Mars doesn't have the ionosphere that the Earth does, which allows us to bounce radio signals over the horizon. On Mars, radio communication would basically be limited to line-of-sight. You'd need a staggeringly large number of land towers. Orbiting relay satellites would be the best bet.

sethmckiness · 2002-10-03 15:42:05

Some basic GPS info. To truely triangulate with GPS, ie Latitude, Longitude, Altitute you need atleast 4 satellites, the satellites are at around 10,00 miles, I want to say there period is 12 hours, there is currently in the neighborhood of 25-30 in orbit, they work by transmitting two frequencies and the two frequencies refract differently. The difference in refraction means one signal takes longer to reach the reciever then the other giving it a distance from the satellite. I imagine you could do it with three but for realistic accuracy you need four, also, I don't know if there is enough atmospheric refraction in the L-band to allow fo the use of GPS, something someone may want to check out.

We are trained on GPS operation, falls under my Air Force career field.

New Mars Forums

Announcement

#1 2002-06-22 14:47:58

Re: Rover Navigation - How should it be done?

#2 2002-06-22 16:19:10

Re: Rover Navigation - How should it be done?

#3 2002-06-23 02:22:28

Re: Rover Navigation - How should it be done?

#4 2002-06-23 11:50:58

Re: Rover Navigation - How should it be done?

#5 2002-09-01 05:10:16

Re: Rover Navigation - How should it be done?

#6 2002-09-01 09:35:22

Re: Rover Navigation - How should it be done?

#7 2002-09-01 17:45:27

Re: Rover Navigation - How should it be done?

#8 2002-09-01 23:53:42

Re: Rover Navigation - How should it be done?

#9 2002-09-02 16:59:05

Re: Rover Navigation - How should it be done?

#10 2002-09-02 17:24:42

Re: Rover Navigation - How should it be done?

#11 2002-09-03 14:49:35

Re: Rover Navigation - How should it be done?

#12 2002-09-03 19:50:49

Re: Rover Navigation - How should it be done?

#13 2002-09-05 12:46:13

Re: Rover Navigation - How should it be done?

#14 2002-09-05 21:39:49

Re: Rover Navigation - How should it be done?

#15 2002-09-06 03:55:11

Re: Rover Navigation - How should it be done?

#16 2002-09-06 20:11:35

Re: Rover Navigation - How should it be done?

#17 2002-09-06 20:53:46

Re: Rover Navigation - How should it be done?

#18 2002-09-07 00:39:48

Re: Rover Navigation - How should it be done?

#19 2002-09-07 00:40:40

Re: Rover Navigation - How should it be done?

#20 2002-09-07 01:58:45

Re: Rover Navigation - How should it be done?

#21 2002-09-08 01:24:49

Re: Rover Navigation - How should it be done?

#22 2002-09-13 12:36:43

Re: Rover Navigation - How should it be done?

#23 2002-09-13 12:43:30

Re: Rover Navigation - How should it be done?

#24 2002-09-26 21:19:08

Re: Rover Navigation - How should it be done?

#25 2002-10-03 15:42:05

Re: Rover Navigation - How should it be done?

Board footer