[X] My Assistant

[RNeuhaus]

Squyres last comments on Athenas on October 4, ...Right at the dog-leg there's a pretty steep step, which we're not certain we can get down. So we're going to descend the upper portion of the ridge, right to where the step is, and assess the situation. ...

On May 6 (Sol 136), when the Oppy was on the rim of Endurance Crater:
I have cut and past a phrase from the Oppy's experience before doing down into the Endurance crater.

Brian Cooper, leader of JPL's squad of rover drivers for Spirit and Opportunity, said the initial view of the crater doesn't settle accessibility questions yet. "The slope right in front of us averages 18 to 20 degrees. Getting into the crater is no problem, but we have a lot more work to do to assess whether we could get back out. That depends on soil properties and slippage, as well as slope." The planned circuit around the rim will also require careful navigation. "If you don't go close enough to the lip, you can't look in, but if you go too far, you could fall in," he said. "We're going to have a very interesting few weeks."

http://marsrovers.jpl.nasa.gov/gallery/all...4P1214L0M1.HTML

From the above picture, ones can see the bottom of crater. However it is very hard to estimate the inclination of slope from the above perspective since the rovers' camera remains on the same horizon surface view.

How do the rovers controllers calculate the slope inclination degree on the way toward to Haskin Ridge? Does the MER have any balancing instrument: Gyroscopy?

Rodolfo

[helvick]

From the above picture, ones can see the bottom of crater. However it is very hard to estimate the inclination of slope from the above perspective since the rovers' camera remains on the same horizon surface view.

How do the rovers controllers calculate the slope inclination degree on the way toward to Haskin Ridge? Does the MER have any balancing instrument: Gyroscopy?

Rodolfo

I'm not 100% sure what you're asking but I'm assuming that you mean how can they figure out what the rovers orientation to the horizontal is?

They use both the targetted sun sightings and the images of the Sundial at the end of each traverse (well after each major traverse has finished but before moving again, and possibly only when they need to know the exact orientation). That's enough to figure out the tilt from the surface normal in the x,y and z planes. Once you have that the Navcam stereo imaging will gives you the 3D layout of the land around the rover.

[RNeuhaus]

I'm not 100% sure what you're asking but I'm assuming that you mean how can they figure out what the rovers orientation to the horizontal is?

Yes, you guessed it well. I haven't tought about the usefulness of Sundial to determinate the rover's tilt. It is a very simple instrument (no weight) and well enough for that purpose. No need of a sophisticated gyroscopy.

Thanks of your explanation.

Rodolfo

[RNeuhaus]

What is the best way to go down? The best way is of course of minor slope, but if there is no another way, the next decision is to select the land which of the following characteristics:

1) Sandy land, a fuffly sand is even better. On rocky land or firm land with lots of pebbles are the worst ones since they make to rover to be very sleepery.
2) Go down as slow as possible, avoid in stopping during the descending until reaching the bottom flat if the slope is greater than 20 degree for a sandy land or less for another characteristics lands (slipery). Must go very down since the greater down speed, it would be harder to stop.
3) Always go down toward the gravity and not in diagonal to the gravity.
4) All wheels must be rotatting at the same time with the rotating speed to be as close as possible to the land. These offers the best bracking.
5) Never brake the front wheel instead of back wheel. The back brake helps to the rover to straight its position toward the gravity by having the effect of back wheels pull.

Hope that the Spirit rover has implemented software alogaritm to compute the percentage of slippage as a new feature of slip-detection. That will help she to have better control when she is descending on steep step.

Anyone are welcome to comment your opinions and critics to help the Rover team to be better prepared and avoid the accident that would be the first recorded in Mars

Rodolfo

P.S. Thanks to Pando, I left a link to Rover's wheels capabilities : http://marsrovers.jpl.nasa.gov/mission/spa...ver_wheels.html

This page says that the rover is designed to withstand a tilt of 45 degrees in any direction without overturning. However, the rover is programmed through its "fault protection limits" in its hazard avoidance software to avoid exceeding tilts of 30 degrees during its traverses. Its tilt capability is very similar to SUV.

[general]

Anyone are welcome to comment your opinions and critics to help the Rover team to be better prepared and avoid the accident that would be the first recorded in Mars 

Rodolfo
[/i]

I thought the first recorded accident on Mars was when Sojourner hit a rock named Yogi and got stuck.
http://www.cnn.com/TECH/9707/12/mars.update/

Attached image(s)

 

[Tesheiner]

What is the best way to go down? The best way is of course of minor slope, but if there is no another way, the next decision is to select the land which of the following characteristics:

1) Sandy land, a fuffly sand is even better. On rocky land or firm land with lots of pebbles are the worst ones since they make to rover to be very sleepery.

I would say the opposite. Compare the northern ridge (sandy) with the southern one (rocky); Spirit gave up trying to climb by the northern side due to slippery terrain. In addition, the MER team was marveled by the good traction observed when climbing by the southern ridge.
Ok, Spirit will be going downslope and not climbing *but* the obstacles in the terrain are quite unknown and it may happen that in order to overcome such obstacles a small detour (drive back [upwards] and forth [downwards]) would be needed.
Imho, Spirit would be in deep sh... going downslope without any hope/capability to climb again.

[RNeuhaus]

Compare the northern ridge (sandy) with the southern one (rocky); Spirit gave up trying to climb by the northern side due to slippery terrain. In addition, the MER team was marveled by the good traction observed when climbing by the southern ridge.

To climb, the better land is rocky land covered by a paved land than on the sand. To climb a sand needs great velocity and momentum to climb. Otherwise, the rocky land, needs less velocity and momentum for the same height as the sandy land. Hence, the MERs team has found a way to climb by going along on the southern side.

Imho, Spirit would be in deep sh... going downslope without any hope/capability to climb again.

Of course, always there is a "line of decision to go down or not". If the decision to go down means that there is solution in either way: backward or downward but with good exit to another place.

Hence, the thoroughly view toward the down slope until a safe point, the decision would be correct ones since the path toward down is well studied and sure without any hestiate go backward but on the firm way on the toward. It is like life or death. Any mistake ones is almost alike a way to the death...

Rodolfo

[Marcel]

2) Go down as slow as possible, avoid in stopping during the descending until reaching the bottom flat if the slope is greater than 20 degree for a sandy land or less for another characteristics lands (slipery). Must go very down since the greater down speed, it would be harder to stop.
5) Never brake the front wheel instead of back wheel. The back brake helps to the rover to straight its position toward the gravity by having the effect of back wheels pull.

Rodolfo,

2) Speed is of no importance considering MER: they move SLOW (in order of an inch a second). If they go considerably faster, this would mean that the vehicle rolls or slides down, ending up in an uncontrolled crash downhill.
5) There's no need to keep head down all the time i think. Traversing (in fact: deviating from the route straight down) might even be unavoidable due to other restrictions (solar angle, rocks in the way, etc.). It also probably costs less energy per day to traverse down instead of heading straight down the slope: more force on the actuators, so more power dissipated in the motors. Besides: using only rear wheels restricts the angle attackable: there's 6 wheels that can be braked: they're better be used all of the when the "going gets tough"'. Letting just the rear wheels brake and let the other 4 mill around......might end up in a speed record on Martian land (and a pile of expensive rubble downhill).

[RNeuhaus]

Rodolfo,

2) Speed is of no importance considering MER: they move SLOW (in order of an inch a second). If they go considerably faster, this would mean that the vehicle rolls or slides down, ending up in an uncontrolled crash downhill.

I agree with the point 2. This case is discarded for the case of MER-x.

5) There's no need to keep head down all the time i think. Traversing (in fact: deviating from the route straight down) might even be unavoidable due to other restrictions (solar angle, rocks in the way, etc.). It also probably costs less energy per day to traverse down instead of heading straight down the slope: more force on the actuators, so more power dissipated in the motors. Besides: using only rear wheels restricts the angle attackable: there's 6 wheels that can be braked: they're better be used all of the when the "going gets tough"'. Letting just the rear wheels brake and let the other 4 mill around......might end up in a speed record on Martian land (and a pile of expensive rubble downhill).

It depends on the slope angle. If it is lower than the 30 degree, Spirit may transverse but with less safety margin, the other words, there is no tolerance to rise any of the wheels over a small stones. Yes, I agree that going down, it is better to keep all 6 wheels to mill about the same time on the land but if the rover is out of control, the back brake must exercise greater control until the rover is straighted down and apply the brake to all 6 wheels. I realize that this case would not apply for the MER-x since it will go very slow.

Now, Spirit will probably go down on the East wing and by that side, at the beginning looks have a soft slope and good land (little sandy) and so won't any problem up to the next checkpoint.

Rodolfo

[RNeuhaus]

They use both the targetted sun sightings and the images of the Sundial at the end of each traverse (well after each major traverse has finished but before moving again, and possibly only when they need to know the exact orientation). That's enough to figure out the tilt from the surface normal in the x,y and z planes. Once you have that the Navcam stereo imaging will gives you the 3D layout of the land around the rover.

Helvick

I found details about the MER' s hability to determine its referential stability position.

http://mars.jpl.nasa.gov/mer/mission/tl_su...av_balance.html

ROVER NAVIGATION DURING SURFACE OPERATIONS
Keeping the rover right side up and balanced

The rover's Inertial Measurement Unit (IMU) uses gyroscopes and accelerometers to determine the heading and tilt of the rover. The gyroscopes measure small heading changes very accurately, and the accelerometers measure where gravity is strongest, pulling down on the rover. Having knowledge of where gravity (down) is, the rover can partly assess its orientation. The rover will also use its tilt sensors to prevent rollover.

If the IMU fails or the rover is lost, the flight team can use the Pancam, which is not part of the autonomous system, to try to figure out the rover direction and position relative to the horizon (attitude).

Now, I see that procedure way by analyzing own gyroscopes and accelerometers is much better than doing by visual measurements. However, these are useful for backup purposes.

Rodolfo

[sattrackpro]

JPL says the rovers travel at 2 inches per second. That is:

120 inches or 10 feet per minute - 600 feet per hour.

or

50.8 millimeters per second, which is:

3.048 meters per minute - 182.88 per hour.

Considering that they have to scan the road ahead, convert every inch into data, then use that data to calculate for the presence of dangerous obstacles, and make adjustments to heading that will lead to the objective location, or halt, if they can't see such a path, these things are moving at blinding speed!

[helvick]

Helvick

I found details about the MER' s hability to determine its referential stability position.

http://mars.jpl.nasa.gov/mer/mission/tl_su...av_balance.html

Now, I see that procedure way by analyzing own gyroscopes and accelerometers is much better than doing by visual measurements. However, these are useful for backup purposes.

Rodolfo

This is very interesting - I had thought that the Pancam sun sighting was the definitive source of orientation data. I knew there was an IMU but I thought it was used only for intermediate data and in combination with single sun sight shots to provide accurate oritentation data. Does anyone have any data on the accuracy of the IMU?

[RNeuhaus]

JPL says the rovers travel at 2 inches per second.  That is:

120 inches or 10 feet per minute - 600 feet per hour.

or

50.8 millimeters per second, which is:

3.048 meters per minute - 182.88 per hour.

Considering that they have to scan the road ahead, convert every inch into data, then use that data to calculate for the presence of dangerous obstacles, and make adjustments to heading that will lead to the objective location, or halt, if they can't see such a path, these things are moving at blinding speed! 

Sattrackpro, If MER uses the facility of hazard avoidance, the girl advances only between 40-50 seconds and stops for 10 seconds for reassess the situationand compute the next move.

More detailsMore details about Rover navigation operations

Mission Timeline: Surface Operations

ROVER NAVIGATION DURING SURFACE OPERATIONS
Summary | Understand Distance | Avoid Hazards | Create Maps |
Keep Balanced | Know Direction | Traverse Far and Well

Using hazard avoidance software for a safe journey

To safeguard against the rover getting lost or inadvertently crashing into unexpected obstacles, engineers have developed software to help the rover make its own safety choices and to "think on its own." The rover hazard avoidance software stops the rover on an average of every 10 seconds, reassessing the situation and computing its next move for the next 40-50 seconds, after which it begins traveling again. With the hazard avoidance software, the rover can travel safely an average of 30 centimeters (1 foot).

The MER uses a Motorola PowerPC microprocessor (RAD6000) and 128 MB of RAM and 256 MB of Flash memory which are of old technology for today standars. Hence, MER move very slow.

Rodolfo

[alan]

I entered the target names from the pancam tracking page into google. It appears they are from Kansas.

[MahFL]

I would not have thought the speed of the processor determins the forward speed, its how the software works. No 1 priority is not to damage, or get the rover stuck, wether its in sand or lying on its side, or upside down etc. There is absolutely no point in going fast if its the last drive you ever do.