[ODE] Problem with hinge and hinge2 joints

[ron taņeza]

I'm using ODE to simulate wheeled cars.. not the real ones, but toy cars -- it's supposed to be a mobot simulator. The mobot uses a car-drive system with 2 rear wheels for moving forward/backward and 2 front wheels for steering. I used hinge2 joints for all wheels at first, with the rear wheels' axis1 (vertical axis) lostop and histop parameters set to zero. The problem is, the rear wheels rotate around axis1 when the car is turning at high speeds (problem with the joint stops?) 

So I try using hinge joints for the rear wheels. This works fine, but now the car tilts forward (i.e. front part of the car is lower than rear part) even though I placed the wheels at the same height.. it seems the anchor points of the hinge and hinge2 joints are not the same. This is at -9.8 gravity. When I set the gravity to small values (like -1), the tilt disappears. I also tried setting the chassis' mass to small values, and the tilt also disappears even at -9.8 gravity.

All the joints have the same joint erp and cfm parameters (erp=0.4, cfm=0.1). It seems the hinge joint can handle more force than hinge2. Is this correct? Right now I offset the front wheel positions so they're lower than the rear wheels.. this works after several trial and errors, but it doesn't really solve the problem, since the user should be able to set the size and mass of the car chassis and wheels, and gravity, etc., and he'll then have to adjust the offset to remove the tilt. 


2,000,000,000 Web Pages--you only need 1. Save time with My Lycos.
http://my.lycos.com