First step done!

Highly doubtful, but add duct tape and I'm sure you're fine.

Tommy

Reply to
Tommy Thorn
Loading thread data ...

I don't think solar cells will recharge your batteries quickly enough to justify the amount of time you will have to "rest". Plus, you are at the mercy of the weather, in that you hope you have enough sunlight during your rest period to make the solar cells useful...

Have you considered some type of fuel cell to recharge the batteries, or even to supplement your power supply? Granted, they are expensive, but they might be the answer to your problem and there are a few "toy" fuel cell kits on the market to allow you to experiment with them on your test rc car.

(You can even devise a way to jettison the fuel cell once it is spent, to help reduce your vehicle's weight - assuming the fuel cell assembly is heavy enough to justify the effort.)

Just a thought...

~WEC

Reply to
W.E.Cole

It's really interesting. I started a similar project with similar goals (low-cost re-rash of the DARPA GC) and similar components (R/C car, cameras for vision, PC-based brain). You can take a look at the current status here:

formatting link
It's great to see that others think along the same lines as well! I however have some doubts about your goal of the 10 mile range. My measurements and experiments show that I won't get more than half an hour of operation out of two 7.2V 3Ah battery packs. Do you think you can navigate with an average

20mph over rought terrain? Or do you think that you can get significantly more operating time out of these batteries? Or do you use a different power source?

Regards, Andras Tantos

Reply to
Andras Tantos

Yeah, nowhere near the full coarse. I'm pretty sure (90%) the record so far is that red hummer at 7.somethin miles. Appearantly it's not an easy thing :-)

--Andy P

Reply to
Andy P

Sandstorm (the red Hummer) did go the furthest last time at seven miles. According to Red Whittaker when he came to talk to our company about the race last year (we're a partner in the project) Sandstorm got hung up on the last major obstacle. Most of the rest of the course would have been relatively straight forward desert driving. I expect at least one or two machines to finish this year.

Catman

Andy P wrote:

Reply to
Catman

Do you have enuf battery power on the bot to operate this board for very long?

Reply to
dan

"Andras Tantos"

Hi Andras,

Very nice work! Any special reason you chose the stampede over the e-maxx?

Regarding the 10 mile range, after initial testings we are doubting that will be the case, but it will be at least 5 miles. This week I got my shipment of 64 cells of GP3700 nimh batteries. While some of them are going to power the computer and the sensors, 48 of them (4 packs of 12 in series) are reserved for powering the two 540 14.4VDC motors. (14.8Ah or 213W).

For the 10 miles range, we initially though about lithium polymer (lipo) batteries, but they are still too expensive. We would have spent about 1.5K only on batteries. Also, they require special care on recharging.

I made some onboard video going over rough terrain this past weekend and I saw that it's going to be difficult to achieve 20mph over rough terrain. So we decided to go ahead and work with whatever limitations we find. Hopefully we will be able to navigate this thing until the battery stops, and if we do so, we can get another grant to upgrade the platform... a real sized one? who knows.

I quickread your page, I'll bookmark it and read with more care, as I'll probably stumble over many of the problems you went through.

Padu

Reply to
Padu

At least in theory, if I didn't screw up my calculations.

Reply to
Padu

Hi Padu!

Thanks. The main reason for me to go with the Stampade was price. I will also have to replace the springs just you did due to the additional wheight but I thought I will not need the extra power of the second motor.

Talking about weight, the 64 nimh cells must be quite heavy. My aproach is to eventually reduce power consumption of the electronics on board by moving to something other than a PC and extend the range that way.

I was thinking about using some high-speed accelerometers or gyros to capture the orientation of the cameras at the time the images were taken. That way I will be able to compensate for the bouncing around on the terrain. Kind of like a home-brew image stabilizer.

There are quite a number of problems with high-speed navigation apart from the mechanical rigidity problems, and so I didn't at least for the beginning think about doing that.

One obvious thing is that the processing requirements will be quite high. My initial testings show that the processing of a stereo pair of images on a

1GHz PIII takes about 200ms at least with the algorithm and resolution I choose. That's not counting the map building, vehicle localization, path planning, etc. This means that I don't think I can get faster than 4-5 frames per second. If you go 20mph, or around 30km/h, your vehicle travels 5 feet or 1.6m between each frame. It might be OK for a large 4x4 because the mechanical time constants are quite high anyway, but for a 1:10 model car such separate samples will make controlling really hard. Though from your higher number of batteries I would guess that you can get ~1.5 hour operating time and thus the ~3mph average speed can get you to the 10mile range. With that speed 5fps sampling might be adequate.

The other problem is that with traditional HW (USB cameras) it's impossible to take the frames from the left and right cameras at the same time. If you move, the time error will amount to a location error, and will make your processed data less precise. You could for example use a synronized mechanical shutter to prevent that but than you would have to have some kind of an iris as well to control the exposure. What are your plans for dealing with this problem?

The third issue, is that at high-speed travel you will probably have to have a more precise physical model of your car to prevent slip, roll and in general keep it in the safe operating region. Again, more processing power. Because of the long stopping distance you would have to indentify obstacles sooner and plan ahead longer. For that you would need a more precise model of your surrounding which probably results in higher resolution requirements or zoom-lenses. One makes the processing power requirement sky-rocket, the other is expensive.

One thing that high speed helps in is that since you can't make sharp turns anyway, you don't need wide angle lens - something that's hard to deal with because of its high cusion-distortion.

Anyway, all I'm trying to say is your project is really ambitious. I hope you succeed. Good luck with it and keep us poseted!

Andras

Reply to
Andras Tantos

"Andras Tantos"

I'm already using a via eden, and it really has a very low power consumption compared with a regular PC. My main problem is the energy to make the motors go on.

In this sense, I'm planning on installing some simple sensors on the shocks to measure activity. It should be possible to "learn" that some types of surface allow for greater speeds without much shock activity (and less vibration to the cameras)

I know that, and I hope to tackle those problems. If you keep the grand challenge in mind, the problem is not only traversing the desert, but be the first one to traverse. So speed is a very important component in the equation. Although I am aware that 50 mph on a hummer is very different from

20 mph on a RC car.

We won't be using stereo, which saves some computing time already. And as stated above, one of the main points of my research is to "find out" where it is possible to speed up. If the rover is on a flat surface, then it may afford to travel one or two meters "blindfolded"... Of course these are only theories in my mind... I may be completely wrong.

As stated above, we won't use stereo. One camera (slightly wide angle) will be fixed looking ahead. The other one will be on a pan-tilt base, together with a range finder (sonar). Think on how we drive. Before making a 90 degrees turn, we turn our head first, and then we turn the wheel. Kind of the same.

I think that will be the biggest challenge... with the increased center of gravity, the car flips very easily. I believe we will have to impose some limits on speed anyway, or else we would have to change the platform too much.

I know, and it will probably get many simplifications along the way. But in order to break through in anything, you have to dream high. You only have to be wise enough to realize when it is time to step back or be persistent when big challenges come in your way.

Cheers

Padu

Reply to
Padu

different from

if you keep the past results of the challenge in mind, you could take as long as you want to get to the finish. as long as you got there, you'd win. Everyone got stuck.

Rich

Reply to
aiiadict

Hi!

My eden board died (on the selve!!) before I could measure its power consumption. OTOH for my purposes I'm afraid that board would have been too slow anyway.

That is actually a very good idea! I might steal it, if you don't have a patent on it :-)!!

That's true. My approach is however to tacke the 'getting there' part first and once I have a reliable solution for that, worry about speed.

Sorry, I don't know where I got that you're working on stereo vision.

Yip, that's an issue. And making a pan-tilt base to work reliably on a moving platform is hard. I don't really have a plan yet for that problem. On idea I've been toying with was to split the image into multiple segments and use mirrors to project multiple views into a single camera: front, left, right, rear. Not simple either and has its own drawnbacks but at least no moving parts.

Regards, Andras

Reply to
Andras Tantos

One advantage of a smaller vehicle is you may be able to rig up some sort of re-inversion device to right an overturned vehicle. If you saw the pictures of the 1st Darpa challenge, you'll see big-red upended on the course with it lidar [??] dome crushed in. Prolly takes a crane to turn over a hummer :). ===========

Whereas big-red had just about every sensor known to man on their hummer, the following guys indicated that their stereo vision system alone [back in 2004] was almost adequate to run the entire course ... note the last sentence below ...

formatting link
D.A.D. POSTMORTEM:

formatting link

3/16/04 (Following Tuesday after the race): Here is a "final results and post mortem" update for Team DAD at the DARPA Grand Challenge.

What you probably know is that we finished third, making it 6.4 miles, behind the Red Team and Sci Autonics II teams made it just over 7 miles and 6.8 miles respectively. We were paused and then when we tried to restart we got hung up on a rock for 2 hours then were disabled. Here's some more info you might not have heard.

Dave and I redrove the route yesterday morning, largely in autonomous mode. We think we found the rock we hung up on. The GPS tracking looks like it led us right over it, and once they paused us the rock acted like a wheel chock. When we restarted the speed limit was only 5 MPH and we didn't have the smarts to accelerate until we moved. We only gave enough gas to go 5 MPH and that wasn't enough to get over the rock. Once we were disabled, the chase team just tapped the gas and we were over the rock no problem. D'oh!

However, we still wouldn't have gotten out of the hills. There were "switchbacks" and other turns in the road that we weren't programmed to handle - we needed more turning radius at slow speeds and better servoing at high speeds. Not a problem - just a software change - but we weren't ready for it.

All tolled, we were probably only 2 days of programming time away from completing the course.

Reply to
dan

"Andras Tantos"

Go ahead :-) I't probably public domain already, as I've mentioned it in a public medium. I've been looking for a few linear potentiometers (the ones with retractable strings), but they are too precise and expensive (their specification is in the order of counts per micron). I thought about hacking old mouses and using their encoders... if you use that route, please share your design.

I understand, I'm taking that approach too, but keeping the speed goal in mind while I'm designing it.

I won't deny that it is difficult. I first thought about that logitech orbit camera, which has a pan/tilt mechanism (and an excellent image by the way), but it is too fragile. Any bumps generated unecessary vibration in the mechanism. I'll have to design my own pan/tilt mechanism with more robustness and some shock isolation... so far only an idea, I will only know how practical it is when I get my hands dirty.

Cheers

Padu

Reply to
Padu

My thesis advisor posted an additional challenge (as if making this thing to navigate by itself wasn't a big enough challenge) of making the car to "jump" some small obstacles. Although I'm not planning it for the first phase of the project, we put some thought about this requirement and we have considered the idea of a spring loaded mechanism to unflip the car when it is upside down... but I don't know how effective it would be. Maybe designing my own chassis that doesn't have an up side (any side is up side... symmetrical about the horizontal axis) would be a more robust solution.

These types of information are really usefull for us that are just starting. I'm lurking the DARPA GC newsgroup too, there are lots of good information coming directly from the teams that participated last year or that will participate this year...

Cheers

Padu

Reply to
Padu

Don't you only want to monitor shock absorber activity rather than actual position at any moment in time? Also optical encoders attached to the shock absorbers on a vehicle will be subject to quite a lot of dust (could quickly be obscured)..

Why not attach a magnet to the moving part of the suspension and a hall effect sensor on the body close by? That way you would get an indication of suspension arm activity and speed albeit with no (accurate) position information.

JMTPW.

Jon.

Reply to
jon sutton

logitech orbit

As you probably know, some of the consumer-market hand-held video camcorders have self-stabilizing mechanisms, so you might investigate what they use. Besides mechanical devices, you can use deblurring software, autocorrelation rectification, etc. Eg, if you have a

1024x1024 camera, and limit the usable image size to 512x512, then you have enuf margin around the edges to compensate for a 256-pixel jitter/shift in any direction between frames. With a 1-Ghz cpu, you prolly have plenty of time to autocorrelate and re-justify the images. The commercial camcorders don't allow for this much jitter-margin, I think, but for your app, you might do it.

Also, JFTHOI, you might take a look at the techniques they use with the vibrating retina devices ...

formatting link
These are cool and actually modeled after jumping spiders, whose retinae are narrow vertical strips, which they scan back'n'forth horizontally to take 3-D measurements. If you think about it, by limiting sampling to only a 1-D strip and then processing this at successive time increments, the spider brain reduces the level of processing by about the square-root. N(t) instead of N*N(t). Even so, prolly pushing what a spider brain can handle :).

- dan michaels

formatting link
======================

Reply to
dan

information

You might take a look at the websites of the teams that did the best in the Darpa challenge, and pay special attention to what they were using for sensors ... Axion, DigitalAutoDrive, Overbot, RedTeam, Sciautonics. Eg, as noted last time, DAD relied mainly on stereo vision, and got almost as far as big-red, who had about $3-4M worth of every sensor know to man, along with something like 5+ multi-Ghz cpus.

Reply to
dan

"jon sutton"

That's exactly what I'm looking for. There are 3 factors that are very important here:

1) cost 2) size (I'll need to implement at least 4 of such sensors, the car has 8 shocks, but I think I can measure activity of 2 shocks in one wheel using only one sensor) 3) robustness

From your comments, I realize that an optical encoder may not comply with

3rd one. Can you give me more pointers about some hall effect sensors that would fit this applications?

Appreciated

Padu

Reply to
Padu

In the UK RS components stock a Honeywell device - SS495A (RS Part No.

216-6231 at 1.37 UK Pounds) I believe it is in an e-line package and available in a surface mount package. Operates from 5 to 10Volts and gives an analogue output from 0.2 to supply voltage dependant on magnetic flux strength (closeness of the magnet) They also have ruggedised industrial sensors that have a sealed body with a screw thread on the outside for mounting - these are a around £25 each and would probably be too large for your present vehicle, but might be suitable when you move up to a bigger platform.

formatting link
Has a lot of info on this type of sensor, as well as a list of worldwide suppliers.

Hope this helps, Jon.

Reply to
jon sutton

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.