Title: COMBINATION OF SUPERVISED LEARNING AND REINFORCEMENT LEARNING FOR ROBOT NAVIGATION

Year of Publication: Jun - 2012
Page Numbers: 49-57
Authors: Fateme Fathi nezhad, Vali Derhami, Mehdi Rezaeian
Conference Name: The International Conference on Informatics and Applications (ICIA2012)
- Malaysia

Abstract:


Applying supervised learning in robot navigation encounters serious challenges such as inconsistence and noisy data, difficulty to gathering training data, and high error in training data. Hence, using Reinforcement Learning (RL) that is a powerful interactive leaning has been encouraged. However, RL algorithms are time consuming also have high failure rate in the training phase. Here, a novel idea for utilizing advantages of both above mentioned learning algorithms is proposed. A zero order Takagi- Sugeno (T-S) fuzzy controller with some candidate actions for each rule is considered as robot controller. This structure is compatible with Fuzzy Sarsa Learning (FSL) which is used as a continuous RL algorithm. In the first step, the robot is moved in the environment by a supervisor and the training data is gathered. As a hard tuning, the training data is used for initializing the value of each candidate action in the fuzzy rules. Afterwards, FSL fine-tunes the parameters of conclusion parts of the fuzzy controller online. The simulation results in KiKS simulator show that the proposed approach is able to accelerate the learning time as well as to improve the action quality.