Gait synthesis for a three-legged robot using Learning Automata

摘要

This article presents a solution to the problem of gait synthesis for a three-legged robot with three actuators in each leg. In other words, it is shown how to coordinate the robot leg actuators in order to maximize the robot walking speed. The position in time of each leg actuator is described by a periodic function that is found by using a reinforcement learning technique called Learning Automata. MATLAB/Simulink and the SimMechanics Toolbox are used to simulate possible solutions for the problem and the simulated robot response is evaluated at the end of each trial. After the reinforcement learning algorithm converges to a solution, it is applied to the real robot that was built using the Bioloid Comprehensive Kit, an educational robot kit manufactured by Robotis. The response of the real robot is then evaluated and compared with the simulated robot response. It is shown that proposed solution generates a quite satisfactory gait for the real robot.