Optimization of Quadruped Robot Locomotion Gaits Through a Genetic Algorithm

During the last years research and development on legged robots has grown steadily. Legged systems present major advantages when compared with “traditional” vehicles, allowing locomotion in terrain inaccessible to vehicles with wheels and tracks. However, its energy consumption still lag being these vehicles, existing several aspects that need to be improved and optimized. One of them regards the parameters values that these machines should adopt to minimize the energy consumption. Due to the large number of parameters involved in this optimization process, one way to achieve meaningful results is using evolutionary strategies. Genetic Algorithms are a way to “imitate nature” replicating the process that nature designed for the generation and evolution of species. The objective of this paper is to present a genetic algorithm, running over a simulation application of legged robots, which allows the optimization of several parameters of a quadruped robot model, for distinct locomotion gaits.