Mobile Robot Obstacle Avoidance System Based on GA-Aided OIF-Elman Network
Hongtao Wei, Qing Ye · 2020
Obstacle avoidance is one of the most basic and significant research directions of mobile robot. However, the present obstacle avoidance algorithms of many mobile robots are complex and unable to adapt to complex and changeable environments. This paper proposed an obstacle avoidance system based on GA-aided OIF-Elman neural network. The system can guide mobile robot to complete the movement and obstacle avoidance in the environment with obstacles. Based on the data collected by the robot’s six infrared sensors, the system adjusts their direction and changes the robot’s motion at the next moment. After data training, the information of network is saved as system file. Then put the system file in simulation software for test. The experimental results prove that the system designed by GA-aided OIF-Elman network is more effective for obstacle avoidance, compared with the performance of general Elman network, GA-aided OHF-Elman/Elman network.