evaluate RL agent use setrealtimeSimulation
Posted: Wed Sep 28, 2022 12:57 pm
I created a reinforcement learning robot simulation environment through pybullet,I train in Direct mode,and I set the maximum number of steps to 10,000,During training, the monitor found that everything was fine and the model could pass 10,000 time steps,but when i use setrealtimesimulation in GUI mode ,My model didn't seem to work and could only go a small distance before collapsing,And the number of interactive steps displayed is far less than 10,000.I am very confused about this, and I hope I can get your help 