Exasor

Thanks a lot for your reply. I have tried the original KinematicObservation without the additional variables in both the original coefficient and the Carsim full size suv coefficient using dynamics...

Using the original coefficient, the ego car sways less, but it still can't overtake as well as the kinematic model performs: https://user-images.githubusercontent.com/56293255/113412785-bba9d800-93eb-11eb-9349-2fde81408877.mp4

I think the number of episodes may be not enough, and I will try more episodes.

I set 5Hz of simulation_frequency and pi/4 of maximum steering angle before, I will try a larger frequency and smaller angle. Thanks for your suggestions!

I will try it, thanks again.

Hello, do you know the meaning of op_mask now?