Deep Reinforcement Learning for Time Optimal Velocity Control using Prior Knowledge
share
While autonomous navigation has recently gained great interest in the field of reinforcement learning, only a few works in this field have focused on the time optimal velocity control problem, i.e. controlling a vehicle such that it travels at the maximal stable speed. Achieving maximal stable speed is important in many situations, such as emergency vehicles traveling at high speeds to their destinations, and regular vehicles executing emergency maneuvers to avoid imminent collisions. Traditionally, time optimal velocity control is solved by numerical computations that are based on optimal control and vehicle dynamics. In this paper, we show that a deep reinforcement learning method for the time optimal velocity control problem outperforms a numerically derived solution. We propose a method for using the numerical solution to further improve the performance of the reinforcement learner, especially at early stages of learning. This result may contribute to the optimal control of robots in applications where some analytical model is available.
READ FULL TEXT