I'm trying to build an r/c car controlled by pc / arduino (http://mechariusprojects.com/thunderbolt). The planned max speed is around 50km/hr, and one of the things I wantto measure is wheel slippage. From a bit of research i think am easy way to do this may be comparing wheel endcoders with an optical mouse sensor rigged up to measure xy displacement.
There are a lots of optical mouse / Arduino interfacing tutoris out there so I think I'm ok with that. My question is more around the quality of the sensor - looking at the specs for many optical mice (gigs-byte ones specifically) they list a top speed of 28 inches / sec or around 2.5/hr. Has anybody had any experience of these sensors running at faster speeds? Would it work or should I be considering a different approac
No I don't think so. The mouse sensor is a ccd camera, like the one you have on your cell phone, only less pixels, about 40*40 only. A digital signal processor compares subsequent frames to determine motion. If you want higher speed measurement, like near 50km/hr, can you use a gps sensor? You get updates every second, not very fast.
Thanks for the replies. The point of the project is to learn as much a possible, and to try different things out. The brief is to control engine performance from the Arduino, and whilst Gps would give a good picture of position and velocity etc (and I plan to put it in at some point) I would like to measure wheel slip (if any) to help calibrate the gear changes, work out "throttle maps" for acceleration etc.
Im still working through it but I'm fairly certain only the rear wheels will be driven. I assume i could get the same effect if i just measure the difference between the front and rear wheel speeds?
I would like to measure wheel slip (if any) to help calibrate the gear changes, work out "throttle maps" for acceleration etc.
Your project sounds like you are trying to re-invent ABS (anti-lock brake system). ABS typically monitors the speed of individual wheels to determine if any of them are slipping (locked up).