Autonomous Vehicle Navigation Behavior Generation by Reinforcement Learning
Keitaro Naruse* and Yukinori Kakazu* and Ming C. Leu**
*Hokkaido University, Kita-13 Nishi-8, Sapporo 060 Japan
**New Jersey Institute of Technology, Newark, NJ 07102, USA
This paper presents an efficient reinforcement learning algorithm for autonomous vehicle navigation. Efficiency is achieved by identifying the structure of a given problem, and it is represented as a set of behaviors – efficient action sequences for solving the problem. Computational simulations are conducted and the proposed mechanism demonstrate.