I have not, and even though I used to race 1/8 th gas on-road cars, I do not see the point.
There was an earlier comment that this would be a good way of testing the Ardupilot and refining the code, but given the resolution of the GPS, and the speed of the cars, I think it would not be accurate enough for a track at least.
I am planning to do it. Contrary to what other people may think, there is a point to that. I am a Senior Scientist at for a government agency that needs detailed river cross section information. There are about 10,000 stream gauging stations across the US, and, it costs a fortune to maintain that information, since river bottom topography changes quite frequently, especially after major flooding. In addition, there is a need to get total bathymetry for a number of river miles. It cost a fortune ($15,000 - 20,000 per point per year. Just do the math...) to maintain the information on river cross sections, and even more, to update the bathymetry for long river distances. Successful data collection by autonomous or semi-autonomous vehicle would save the taxpayers millions of dollars a year. I may need to use a more powerful processor, to use real-time differential GPS, but, for now, I am using Ardupilot. I am planning to update this site with my progress.
Pedro, I can probably put you in touch with some guys who have done this before. They had a radio controlled boat with RTK single beam echosounder and I think they even used multibeam. I have a video of it somewhere. The problem with stream gauging stations may be the stream velocity vs the hull speed of the the vessel and a planing hull is no good for a bathy survey! I have been looking at a hydro-power site which is somewhat remote - up in the mountains - and an option to survey the existing lake in the corrie (should it be necessary) is to use a radio controlled boat. Not sure what type of vessel you are proposing to use but if its of any size a tiller pilot is cheap and can be controlled by an NMEA GPS or chartplotter so depending on size there would be no real need to develop much new technology.
I presume the system for gauging you are using is the USGS system and indeed it is expensive! For one reason or another it has been thrust on developing countries or was left there as a remnant of colonial occupation. A close friend and professional colleague (aged 78!) is back in Nigeria where we have worked together extensively. He has developed an alternative and very cost effective means of measuring water resources (he is a hydrologist) and is setting up and NGO to promote it - and he has invested his own funds to start setitng it up. At least one State Government has agreed to support this project which is great news.
I look forward to watching the progress on your project and wish you luck.
Inertia Labs (www.inertialabs.com) has built a small catamaran with GPS that surveys harbors and channels. If you were really trying to do this autonomously, you need a combination of an aerial view to establish the limits of the waterway along with a water level device to measure depth. Perhaps you could use satellite photos to establish the aerial perspective and corresponding GPS waypoints. It shouldn't be difficult to find inexpensive digital depth sounders or fish sounders with NMEA interface.
Garmin claim that the GPS60 interfaces with one of their fishfi nders (can't remember the reference but can get it if required). You need to buy a special cable from them. My friend has it out in Nigeria but wasn't able to test it before he left as the cable arrived too late. He is having difficulty getting them to talk to each other.
In the mid-90s we started surveying reservoirs in Africa using a Garmin GPS12 and a cheap NASA yacht echosounder. All of the data was logged by hand by simultaneously marking a waypoint and writing down the depth at that point. Got great results for hydrological purposes. Some of these were almost 40km long and would have been very difficult to survey cheaply any other way. We used an 8' pram dinghy we built from plans in my living room and a 2Hp Yamaha outboard we smuggled in in a suitcase. When the fishing nets became too much of a nuisance for the outboard my driver dutifully hired a dugout canoe with its owner on the paddle and proceeded upstream over 3 days to complete the survey.
Hello Howard, I have seen a small catamaran used to survey river cross sections and, at the same time, obtain water velocity. However, I couldn't find it on the inertialabs place. Could you point me to the specific url? The idea is to use ultrasonic sensors for the depth and velocity.
Hi Mike, Somehow I missed your comment until today. Yes, I'd love to be in contact with the guys that have done that before. I am also a hydrologist, and work at the National Weather Service. The problem of the stream velocity is a real concern, especially during flood conditions. Possibly worse than that is the problem of junk coming down the river (trees, for instance), and trapping the boat. So, if you find the information you had, I'd appreciate it. I'll keep you posted on how my project progresses.
Me too. Have resources and keen to chat. Its good to have a partner in these things. I'm kind of doing this on the side of my main duties at work.
Can you drop me line on your progress to date, need a kickstart? email@example.com