INCREASED EFFECTIVE OF CONTROL ROBOT USING REINFORCEMENT COMBINE WITH DEEP LEARNING

Abstract

Although deep learning can solve problems that cannot be done by tradition machine learning algorithms, it requires huge  amount of data, which is not always available in control problems. Reinforcement learning is a good solution in robot control. In reinforcement learning, the data is generated when the agent interacts with environment. Along with the develop of noron networks, many researcher have focused on combine noron network with reinforcement learning to create deep reinforcement learning. In this paper, we propose a new deep reinforcement learning model based on the improvement of the traditional Deep Q Learning algorithm by combining techniques: Fixed_Q Target, Double Deep Q, Prioritized Experience Replay with CNN network (VGG16), apply to control the robot with state and agents designed by using Unity ML-Agents. The proposed model  was tested, compared to the original model on the simulation environment. The results show that the proposed method fix the overestimation q value and fast convergence.