ArduPilotOne ("ArduRover") Live Test of Unmanned Ground Vehicle with Waypoint Guidance/ Gain Changes via QGroundControl / Also Hardware in the Loop via Mavsim

All of this is open source and freely available for download. You can find the core autopilot here. We developed it using eclipse, but the environment is fully arduino compatible as well thanks to the hardwork of Mike!
This repository is for rover specific configuration files and a wiki site for ardupilotone based rovers.
To run the hardware in the loop you will need the mavsim scicoslab toolbox here.

Views: 2149

Comment by Randy on April 11, 2011 at 5:08pm

congratulations!  It's very smooth and seems very accurate, i'm surprised that the inherent gps inaccuracy does cause the car to turn a lot more instead of following such a straight line.


Also why is it jumpy on grass?  Is it measuring it's wheel speed and noticing that it's slowing down, giving it more power which causes it to break free and speed up for a bit before slowing down again?

Comment by James Goppert on April 11, 2011 at 6:15pm
The magnetometer helps out a lot and we have the xtrack gain relatively low, but we were getting a very good gps signal, later we increased the xtrack gain and it still worked well. We also noticed the lurching on the grass, it is using the gps velocity, we believe it is from the integrator on the PID controller. The behavior goes away for faster velocities as there is enough momentum that the grass doesn't slow the car down significantly.

Comment by healthyfatboy on April 11, 2011 at 7:42pm
What hardware did you use for this? It looks like it was the APM (with or without oilpan?), a magnetometer, GPS, I assume XBee to do telemetry. I have an r/c truck ready for some brains but wanted to see what was out there before putting something on.

Comment by Randy on April 11, 2011 at 7:50pm
presumably this is the basis for the ArduRover code?

Comment by Jason Short on April 11, 2011 at 7:51pm

Pretty Awesome James! See at Sparkfun then!


Comment by James Goppert on April 11, 2011 at 8:22pm
Yes, this is the ArduRover code, which is based off of ArduPilotOne. As of now ArduRover is just a config file for ArduPilotOne. This is using the standard APM + Oilpan hardware. We are using a 900 Mhz Xbee. Also, the power jumper SJ1 has to be unsoldered to allow the board to be powered off a separate (11.1 V) 3 cell LiPo. It looks like 3-4 of us will be piling in for the 18 hour drive to Boulder, See you all there!

Comment by Mark Colwell on April 11, 2011 at 9:17pm

James Congratulations!

I will try to port over ArduRover  to my new MultiPilot 32 with a few cores running OpenTLD for vision assisted guidance, with 2 or maybe 4 cameras. Chassis is a mobility GEO_Chair, designed carry a person. It folds up and fits in a trunk too.



Comment by James Goppert on April 11, 2011 at 9:28pm
That looks great Mark! I don't know if you've seen it, but there is a ROS package for mavlink, and APO speaks mavilnk, so that is another option too if you want to have an onboard ubuntu linux box. Porting APO shouldn't be too bad, I have everything abstracted as much as possible, for instance no calls to Serial.print. Everything goes through the hardware abstraction layer class.
Comment by Jim on April 12, 2011 at 3:20am

What do you think the chances are of getting this code running with the old ArduPilot board and IMUv2?  I, probably like many others have an old AP sitting in the draw gathering dust now that I'm flying with an APM.  I'm guessing the answer is not easily as you are using the APM libs, but here's hoping!


Comment by Jonathan Simmons on April 12, 2011 at 9:26am
What ground station are you using ? As a Linux user I find ground stations are few and far between.


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

Join DIY Drones

© 2019   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service