Traditional robots locomotion relied on the kinetic analysis to design a set of instructions to control the robot. When the surrounding environment changed, human had to develop a code to deal with the changes in the environment. Using evolutionary algorithm to solve this problem, evolutionary robot can adapt its behavior to fit the current environment immediately, which is the advantage of evolutionary robots in saturations that no human can help.

Evolutionary robot research has become an interesting topic recently and this research specifically focuses on evolution and learning. Evolution is the adaptation of robots to the environment. Learning is a task-oriented process whereby the robot gains the ability to achieve a given goal in the environment. In this paper, we apply traditional GA (Genetic Algorithm) and design the BRMA (Biomorphic Robot Memetic Algorithm) to control the robot. Our biomorphic robots have four legs and each leg has several joints. We also test the algorithms on a partially breakdown robot. Our study is a multi-objective evolutionary task since the robot has to evolve to fit several indexes.

In our experiments, we set up a beacon light as a target, and the robot evolves to move quickly and smoothly toward the target. We adopt online evolutionary algorithms and test them on the quadrupedal robot. The experimental results show that the robot, from totally random behaviors, can adjust its actions to move quickly and smoothly toward the target.