Humanoid Robot Gait Planning Principles And Methods Of Research,

Research on humanoid robots is currently one of the most exciting topics in the field of robotics and there are many ongoing projects. Development of humanoid robots with natural and efficient movements presents many challenging problems to the humanoid robot researchers. For all humanoid robots, gait planning and trajectory generation are the core problems that mainly contribute to the quality of their movements. As for humanoid robot with limited carrying energy, gait planning with least energy consumption is an important research content for humanoid robot. A gait optimization method, which is based on three energy consumption index, i.e. average power, average power deviation and mean torque consumption, is proposed in this dissertation, three typical motion gaits of humanoid robot are studied deeply and systemly based on above method.Gait parameters of humanoid robot is optimized by introducing three energy consumption indexes, calculation formulas of driving torque for each joint in humanoid robot are derived based on Lagrange dynamic equation, mathematic models for gait optimization are established. A set of optimal solutions for gait parameters are obtained utilizing genetic algorithm in matlab environment. Relationships between energy consumption and gait parameters are analyzed. Numeric examples for gait parameters optimization and analysis of humanoid robot are given for illustration. Results of numeric examples show that the optimized gait of humanoid is similar to that of real human being, with minimum consumed energy.Aiming at gait planning problem of climbing barrier humanoid robot, a stability judgment criteria, which is based on the value of ZMP(Zero Moment Point), is introduced. Mathematic models for gait optimization of climbing barrier humanoid robot are established. These models regard least energy consumption and maximum stability as optimization goals. An on-line gait planning algorithm, which is based on improved BP (Back Propagation) neural network, is proposed in this dissertation, comparison and analysis between real-time gait planning results and off-line gait planning result are also carried out.Focusing on gait planning problem of running humanoid robot, based on single leg bounces dynamics model and double legs bounces dynamics model, an improved bounces dynamics model and an sine-parabola method for gait planning of running humanoid robot are proposed in this dissertation, object function, which regards least energy consumption and maximum stability as optimization goal, is established. Numeric examples for gait parameters optimization of running humanoid robot are given for illustration. Analysis and comparison between optimized gait planning results and off-line calculation results are also carried out.At the end, the gait control of humanoid robot was designed based on a DSP (Digital Signal Processor) controller. We analyzed the significance and function of the basic motion controller in control system and elaborated the controller design thought. According to the idiographic request of the basic controller, we proposed the design method based on the TMS320LF2407 DSP chip. It made emphasis to Narrating the controller software and hardware design method. In addition, the control algorithm has been investigated and the project realization of the digital PID (proportion integral derivative) control algorithm is conducted.