| Resumo: | Different strategies have been adopted for the optimization of legged robots, either during their design and construction phases, or during their operation. Evolutionary strategies are a way to imitate nature replicating the process that nature designed for the generation and evolution of species. This paper presents a genetic algorithm, running over a simulation application of legged robots, that allows the optimization of several locomotion, model and controller parameters, for different locomotion speeds and gaits. Here are studied the model and locomotion parameters that optimize the robot performance, in a large range of distinct velocities.
|
|---|