Cyclic motion generation for intelligent robot by evolutionary computation

In this paper we propose a method for motion generation of intelligent multi-legged robot using evolutionary computation. Legged robot can walk in various complex terrains such as stairs as well as in flat environment. However, setting the robot's behavior to adapt to various environments in advance is very difficult. The robot can mimic the movement of organisms based on computational intelligence. In this study we apply steady state genetic algorithm for generating the motion sequence of a six-legged robot modeled by forward kinematics. The number of intermediate positions of the motion sequence is adapted to the environment and optimized as well. We use a computer simulation environment before we apply our method in real robot. This can reduce the time spent on finding the optimal parameter settings and the solution itself for the optimization problem. The solution is evaluated mainly on the moving distance of the robot. Experiments were conducted to confirm the proposed technique.