"MVP" computer vision rover with OpenMV for less than $90

This is the cheapest good computer vision autonomous car you can make — less than $85! It uses the fantastic OpenMV camera, with its easy-to-use software and IDE, as well as a low-cost chassis that is fast enough for student use. It can follow lanes of any color, objects, faces and even other cars. It's as close to a self-driving Tesla as you’re going to get for less than $100 ;-)

It’s perfect for student competitions, where a number of cars can be built and raced against each in an afternoon.

Instructions and code are here

Views: 1253

Comment by kruk on October 2, 2017 at 3:02am

I miss these handmade projects in the DIY forum. Congratulations!

Comment by Patrick Poirier on October 3, 2017 at 6:15pm

Yahooo !!!  My OpenMV racer is ready  :-)

3D Robotics
Comment by Chris Anderson on October 3, 2017 at 10:29pm

Great! You added your encoders, which I've only just now attached. I'll add them to the code this weekend. 

Comment by Patrick Poirier on October 15, 2017 at 3:51pm

Rainy Week-End ;-) ... Time to test new stuff 

I found this very interesting code :  ESP_ROS , that is using the Arduino IDE to implement a full turtle bot  ROS  stack on the functionality to the Esp8266 Car. I never relly liked the way the LUA code worked on the ESP-12E version with NodeMCU and motor shield (L293D). The code includes differential DC motor configuration with encoder wheels :  https://github.com/agnunez/espros

Did implement ROS but based on the examples I could adapt the PID to the encoders  and added the AccesPoint library in order to make an ''easy connection'' so I can change the carious parameter : Steering gain and offset, & PID, using this: https://github.com/esp8266/Arduino

On the OpenMV, I send the steering command using serial write on uart 3 @155200.baud and can monitor the numbers on my PC using telnet over WIFI.

Its not perfect, its more complicated to tune, and it might be never fully operational, but it made my rainy week-end a lot of fun :-)

3D Robotics
Comment by Chris Anderson on October 15, 2017 at 11:01pm

Very cool! I also added the wheel encoders, but to be honest I'm not sure how much utility they add since I'm already closing the loop with vision. Do you find them useful?

Comment by Patrick Poirier on October 16, 2017 at 4:54am

Yes I agree the odometry do not really optimize the  tracking loop process.  Actually I found this code while looking at single tick encoders (there are not many) , and I could not resist to try it. I might try the full ROS stack, but I suspect the estimator will lack precision because the encoder is mounted directly on the wheel hence giving just 20 ticks per revolution (+- 1 cm per tick).

Anyway, its a nice little toys to play with... thanks Chris for having me rediscovered the WiFi RC smart car and giving it a new life :-)


You need to be a member of DIY Drones to add comments!

Join DIY Drones

© 2020   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service