Posted: 12/6/2009 11:37:58 PM EDT
|
I really should have written this story up a long time ago. It happened back in June and I had almost forgotten about it but I thought it was a good story so here goes. I build custom robotics as a hobby. This is really subject for another thread but this does need some basic background. I've been working on this robot for the last several years off and on. It uses a stripped-out electric wheelchair as a base although it is pretty unrecognizable at this point. There is a laptop computer on-board that functions as the brain (it runs all custom software written by myself). It communicates with some custom circuitry built by my friend Tim. This controls the robot. It can also be driven by remote control from a second laptop using some long-range high-bandwidth radios and more custom software. It also weighs over 200lbs (many batteries) and has quite a bit of power and speed! On with the story, I had just spent much of the day writing software to interface a simple GPS antenna to the on-board computer's serial port. I also had to write algorithms to find heading based on present and past location then to correct this heading. The plan was that this would be the first step in making the robot able to operate without the remote control for longer periods of time. Not knowing how well this would work out I needed to go and test it. An open field of sufficient size was needed for this test. Lack of people was also a requirement as the robot can get dangerous when it goes out of control. As such we went ahead and put a "danger high voltage" sign on the back just to make sure people didn't mess with it. Who would mess with high voltage? The ground also needed to be very flat and level for the equipment to work properly. Also did I mention that it was now after midnight? I loaded the equipment in the truck with my friends and headed off to find a place to run the test. We finally found a place out in a rural section of the state. It was a park that had several baseball diamonds and 4? soccer fields all on the same spot. Lots of room. It seemed perfect. We get out to start unloading stuff and then notice that there is a group of underage kids out drinking on the roof of the snack bar for this park. We figured that there was more than enough room there for everyone so we just ignored it. No sooner did we make the decision to ignore what was going on than four or five patrol cars came screaming into the park from both entrances. It appeared to be a coordinated assault. It also probably consisted of every officer on duty at the time in that little town! So, the kids start running across the field and the cops all get out and start chasing them. They tackle all of them about halfway out into the field and start dragging them back in. Meanwhile we were sitting on the hood of the truck watching this, drinking some soda and eating some burgers we picked up from the store on the way out. It was like being inside Cops, and probably far more excitement than that little park had seen in a good time. Of course this could happen weekly and I'd have no idea. Then once the excitement was over we went over to the back of the truck to work on getting the robot unloaded. Eventually, one of the police comes over to see who we were and why we were there. We said "to test this robot" (gesturing to robot). We then got a shocked "you should keep away from that. It says high voltage". We then mentioned that we put the sign on it after we built the robot. They said alright and even gave us a hand lifting it out of the truck. Once we got it down and going several of the officers stayed and laughed while the robot struggled valiantly to navigate a programmed course. Overall It was a fun night and the test was relatively successful. However, I've been very busy since then so I haven't really made much progress. I doubt it will ever be finished though. As with most hobbies, you just keep adding and tinkering. There really is no end. One final question for those who might know: This place was out in the middle of nowhere and the only people there were us and the kids. How would the police have known that there were people there to go bust? |
|
One final question for those who might know: This place was out in the middle of nowhere and the only people there were us and the kids. How would the police have known that there were people there to go bust? Where do you think the cops partied when they were kids? |
|
Quoted:
One final question for those who might know: This place was out in the middle of nowhere and the only people there were us and the kids. How would the police have known that there were people there to go bust? Where do you think the cops partied when they were kids? Those are the places I look when I'm working 
|
|
Quoted: Sounds like a great night. Pictures of your robot? I just looked. I can't seem to find them right now.  The robot is with my friend Tim right now so I can't take any new pics. I'm sure there must be some around. I'll keep looking. Regardless, I'm thinking about doing a technical thread about the robot sometime this month. There will definitely be pictures posted there. Stay tuned for updates! In the meantime to keep you entertained I'll go ahead and post pictures another one of my epic redneck-engineering feats!  ETA: This was at my friend Dave's yard.     |
|
Quoted: We need pictures of the robot quickly. We have to determine if you are a threat or not. For all we know it started out as a fun and cool idea. Eventually it will turn into this: http://i48.tinypic.com/dmtmh2.jpg Maybe this: http://i46.tinypic.com/2e6hf09.jpg But if you happen to make something like this: http://i50.tinypic.com/sersxv.jpg http://i49.tinypic.com/2dtrntx.jpg I will take 2. That may or may not be the eventual goal of this project.  |
|
Quoted: Quoted: And an electronic compass so that you don't to be moving to calculate heading?Forget about the cop story, lets talk more about the robot  Any future plans on adding accelerometers so that you don't need a perfectly level surface to play on? Why didn't I do that? Well, I already had the GPS receiver and I don't have an endless supply of cash I want to put into this project. Using the GPS antenna also allows me to know the LOCATION which will become more important when I try to integrate the navigation with some kind of mapping software. So far that is still a ways off though. In fact, I can happily say that so far (excluding the laptops which I use for other things) I have kept the whole thing under $500. Anyway, I may yet add a digital compass at some point to augment the GPS for when the robot is moving slowly or is stopped but not right yet. |
|
Quoted: Forget about the cop story, lets talk more about the robot Any future plans on adding accelerometers so that you don't need a perfectly level surface to play on? Yes. I would also like to put some optical sensors on the wheels so I know how fast they are spinning. Once I finish the coding for the GPS (there are still a couple of bugs) I will probably start in on one of these two things. |
|
Quoted: Why didn't I do that? Well, I already had the GPS receiver and I don't have an endless supply of cash I want to put into this project. Using the GPS antenna also allows me to know the LOCATION which will become more important when I try to integrate the navigation with some kind of mapping software. So far that is still a ways off though. In fact, I can happily say that so far (excluding the laptops which I use for other things) I have kept the whole thing under $500. Anyway, I may yet add a digital compass at some point to augment the GPS for when the robot is moving slowly or is stopped but not right yet. That makes sense - I certainly understand the lack of an endless supply of cash... former college student here  That said, one thing that would be an additional benefit for an electronic compass would be to program in dead reckoning for areas where GPS signals are a bit weak. Tie the heading in with distance measurement from the wheels to be able to have a good idea of your location based on the last known location from the GPS and how far you have driven in what directions. A lot of integrated car navigation systems use this for tunnels, parking garages, downtown, etc. An upgrade at some point Projects like these never end! |
|
Quoted: Quoted: That makes sense - I certainly understand the lack of an endless supply of cash... former college student here Why didn't I do that? Well, I already had the GPS receiver and I don't have an endless supply of cash I want to put into this project. Using the GPS antenna also allows me to know the LOCATION which will become more important when I try to integrate the navigation with some kind of mapping software. So far that is still a ways off though. In fact, I can happily say that so far (excluding the laptops which I use for other things) I have kept the whole thing under $500. Anyway, I may yet add a digital compass at some point to augment the GPS for when the robot is moving slowly or is stopped but not right yet. That said, one thing that would be an additional benefit for an electronic compass would be to program in dead reckoning for areas where GPS signals are a bit weak. Tie the heading in with distance measurement from the wheels to be able to have a good idea of your location based on the last known location from the GPS and how far you have driven in what directions. A lot of integrated car navigation systems use this for tunnels, parking garages, downtown, etc. An upgrade at some point Projects like these never end!That is a planned upgrade coming with the optical sensors I plan to get for the wheels. The only problem is that the wheels skid quite a bit when the ground is slippery. I still haven't figured out a way to solve that one. Maybe a laser rangefinder pointed at some fixed point in the distance? |
|
Quoted:
Quoted:
Forget about the cop story, lets talk more about the robot
Any future plans on adding accelerometers so that you don't need a perfectly level surface to play on? Yes. I would also like to put some optical sensors on the wheels so I know how fast they are spinning. Once I finish the coding for the GPS (there are still a couple of bugs) I will probably start in on one of these two things.
I'd suggest using Wii remotes. They have all sorts of useful things in them. Accelerometers, and the IR camera in the end does not just see the Wii bar you put by the TV, you can use/track any light source you want. You could make a tracking routine off of the sun, streetlights, or a beacons you put out yourself. Or even just put LED's on the wheel and use the Wii remote's sensor to count it passing by. You could probably hack one or two Wii remotes into a passable INU for your bot. I know working from fixed beacons kind of undermines your goals for increasing the robot's autonomy, however it might provide a way for a return to base system etc. or other tracking possibilites. And I'm pretty sure as people find out more and more uses for Wii remotes, someone will come up with a USB receiver for them and open source code. |
|
Quoted: I'd suggest using Wii remotes. They have all sorts of useful things in them. Accelerometers, and the IR camera in the end does not just see the Wii bar you put by the TV, you can use/track any light source you want. You could make a tracking routine off of the sun, streetlights, or a beacons you put out yourself. Or even just put LED's on the wheel and use the Wii remote's sensor to count it passing by. You could probably hack one or two Wii remotes into a passable INU for your bot. I know working from fixed beacons kind of undermines your goals for increasing the robot's autonomy, however it might provide a way for a return to base system etc. or other tracking possibilites. And I'm pretty sure as people find out more and more uses for Wii remotes, someone will come up with a USB receiver for them and open source code. We have looked at Wii remotes for adding similar functionality. We are currently meeting tonight to discuss using security cameras and lasers to judge the range of distant targets. |
|
Quoted: Alright mother fucker, you've had ten days, where are the pics?  Sorry, I just talked to Tim the other day and he didn't know where the pics were.  I have a couple of weeks off now so I will probably have some time to work on the project. I will take some new pictures when I get a chance but they won't be as good as the ones from testing outside over the summer.  |