作者: Zhe Tang , Changjiu Zhou , Zengqi Sun
DOI: 10.1007/11539117_37
关键词:
摘要: A humanoid walking gait synthesizing approach, which is able to generate gaits in both sagittal and frontal planes, presented this paper. To further improve the consideration of ZMP (Zero Moment Point) energy consumption constraints, a two-stage optimization method proposed. At first stage, real-coded GAs (genetic algorithms) are used set near-optimal gaits. second as training samples for GA-based NN (neural network) gait. By making use global capability GAs, can solve local minima problem. The proposed approach at any speed feasible range. Experiments conducted verify effectiveness method.