Hybrid Path Planning of A Quadrotor UAV Based on Q-Learning Algorithm

The hybrid methodology is an emerging technology for control solution of nonlinear control systems with infinite states, moreover, by utilizing the approach the system can be transformed to a finite one based on discrete abstractions. In this paper, the problem of path planning of a quadrotor unmanned aerial vehicle (UAV) is investigated in the framework of hybrid methodology. With the kinematics model of the Quadrotor UAV and the abstraction of the environment in the form of grid world, the design procedure is presented by utilizing the Q-learning algorithm, which is one of the reinforcement method. In this process, an optimal or suboptimal safe flight trajectory will be obtained by learning constantly to maximize the reward. Matlab software is used for computation, and the effectiveness of the proposed method is illustrated by a typical example.