High-speed Autonomous Drifting with Deep Reinforcement Learning
Reference trajectorires for seven maps
Reference trajectories for the maps are located in
traj_0: for map(a), for first-stage training.
traj_1...traj_5: for map(b-f), for second-stage training.
traj_6: for map(g), for evaluation
Trained weights for different models
weights are located in
weights/, where four kinds of models are included. Note that
sac-stg2 are different stages of our SAC controller during training.
sac-stg2 is the final version and
sac-stg1 are only trained on map(a).
sh code/test.sh to test different models on map(g) with various setups. The driving data (timestamp, speed, location, heading, slip angle, control commands, etc.) will be recorded in
code/test/ after the testing process.