The development on obstacle avoidance design for a humanoid robot based on four ultrasonic sensors for the learning behavior and performance

摘要

A behavioral strategy designed for a humanoid robot for the purpose of obstacle avoidance based on four ultrasonic sensors for the learning behavior and performance is proposed and implemented with an autonomous humanoid robot. A mechanical structure with 4 degrees of freedom is designed so that a small-size humanoid robot named ARSR is able to accomplish three types of walking motion. One experiment is presented to illustrate ow the the proposed bipedal structure lets the ARSR move forward, turn left and turn right. Four ultrasonic sensors are mounted on the ARSR to obtain environmental information and detect obstacles. Based on the information obtained from these sensors, a decision tree method is proposed to decide upon one behavior from three possible types of movement: walk forward, turn left and turn right. An experiment is carried out to show how the robot can autonomously avoid obstacles to effectively arrive at its destination. Based on the mechanic design and development, we will get a good agreement for the learning behavior and performance.