Automatic Navigation Research for Multi-directional Mobile Robots on the Basis of Artificial Intelligence Application, Q-Learning Algorithm

In this article, research on the applications of artificial intelligence in implementing reinforcement learning is a subset of machine learning that deals with learning decisions from rewards given by the environment. Classic reinforcement learning algorithms are usually applied to small sets of states and actions. However, in real applications, the state spaces are of a large scale and this will bring the problems of the generalization and the curse of dimensionality. In this paper, we integrate neural network into reinforcement learning methods to generalize the value of all the states. Simulation results on the Gazebo framework show the feasibility of the proposed method. The robot can complete navigation tasks safely in an unpredicted dynamic environment and becomes a truly intelligent system with strong self-learning and adaptive abilities.