Hello: To finish my project I am trying to implement a pure pursuit algorithm. My problem it is that in theory the code should work but when doing a field test it **bleep**s-up. The problem is that the steering angle it calculates are completely wrong. I really want and need to solve it but I don't know how. (I can't even get a straight line in a field test)To know:*The GPS sends the heading(orientation of the front of the vehicle) in radians from pi to -pi*The GPS sends coordinates X and Y in an already defined plane of the corresponding UTM, sending the current position in meters away from (0,0).*The GPS also sends the current speed, yet this is constant and is just for checking that the can message sent is retaining the current speed.*The vehicle is an SUV, front steering, 4WD almost 4meters long 2 wide.* On (0,0) facing forward -Y == right, X == forward,and the heading ==0* Recorded paths by the path recording program have a minimum distance between them of S, S=Kla*VehicleLength/12, (kla(is look ahead point constant)) *The vehicle has a relation from -40 to 40 deg answering to α ≈ 15.26108* δ, where δ(steering angle of the outer wheel) and α (steering wheel angle); resulting in a propulsion gain of 1.*Real Position and heading is received in real time from the bus can 500kbd *The Angle is send to the robot through the CAN BUS (This works the robot receives the wrong angle my program calculates) Note: For it to run, Add the matlab codes attached to your labview matlab directory

Actually it now works, I only have the basic labview and the mathscript I don't think I have the tools you linked Anyway... I attach my solution

LabVIEW

Solved!

Thanks!!

Message 11 of 12

(853 Views)

Author

Thanx for what?

Message 12 of 12

(849 Views)