Reinforcement learning-based method for autonomous navigation of mobile robots in unknown environments: an experimental demonstration
Trần Đức Chuyển, Roan Van Hoa, Nguyen Duc Dien, Tran Ngoc Son, Tung Lam Nguyen · International Journal of Advanced Mechatronic Systems · 2021
Reinforcement learning (RL) is a subset of machine learning that deals with learning decisions from rewards given by the environment. The model classic reinforcement learning algorithms are usually applied to small sets of states and an action. However, in real applications, the state spaces are of a large-scale, and this causes the problems in the generalisation and dimensionality. In this research, the authors integrate neural network with reinforcement learning method to generalise the value of all the states. The simulation results on the Gazebo and experiment results software framework show the feasibility of the model proposed method algorithm. The robot can safely navigate in an unprotected work environment and becomes a truly intelligent system with the ability to learn and adapt itself to the model.