Gait Planning Research Of BRHL Artificial Leg Based On Central Pattern Generator

The research purpose of humanoid robot is to make it like a human covering the entire fields. While the research of intelligent prosthetic just needs such a highly anthropomorphic experimental platform. Integrated the study of intelligent prosthesis and humanoid robot, a new-style robot called biped robot with heterogeneous legs (BRHL) is proposed. BRHL can simulate amputees dressed with intelligent prosthesis very well, and provide a research platform for intelligent prosthesis. Based on BRHL, gait planning of its artificial leg is researched. The main research contents in this paper are as follows:(1) Following the elaboration of BRHL research background and research purpose, BRHL overall structure is designed. Four-bar closed chain knee joint has been adopted by BRHL. BRHL hip joint is driven by motors. The knee joint of artificial leg is actively driven by pneumatic muscle actuator and the knee joint of bionic leg is semi-actively driven by magneto rheological damper. The structure design of BRHL has laid a foundation for gait planning of BRHL artificial leg.(2) In order to make intelligent bionic leg follow the gait of artificial leg, gait recognition of bionic leg is researched. Gait sensor system of intelligent bionic leg is designed and the selection of sensors is done. Gait recognition model is established based on artificial neural network and L-M training algorithm is adopted for training of BP neural network.(3) Learning from gait planning methods of biped robot, central pattern generator is adopted for gait planning of BRHL artificial leg. CPG control network is established based on Hopf nonlinear oscillator model and parameter setting is done based on genetic algorithm. The artificial leg gait is gained by MATLAB simulation.(4) Gait planning research of BRHL artificial leg based on Matsuoka nonlinear oscillator model is done. Parameter setting is done based on method of single-parameter analysis and planning gait is gained by MATLAB simulation. The gait motion simulation and combined control simulation are done based on ADAMS, etc.Finally, the work in this paper is summarized. The shortcomings are analyzed and future research priorities are proposed.