Study On Joint Drive,motion Planning And The Matching Algorithm For Humanoid Fast Operation

Humanoid robots have similar characteristics in physical configuration with human beings,which is advantageous in helping humans with daily lives and work in artificial environments.Among the different capabilities of a humanoid robot,operation capability,especially the fast ability to operate in dynamic and non-structural environments,is an important premise to assist humans in tasks.Due to the limitation of the size of its structure,resulting in insufficient joint driving ability,thus it is difficult for humanoid robot to effectively implement the trajectory planning,and achieve fast operation currently.This dissertation will take humanoid robot fast operation as the goal,and presents studies on such key issues as joint drive control,motion trajectory planning,as well as the matching algorithm between these issues.The research issues and results are as follows.First,according to mission requirements and the biomechanical properties of the humanoid robots in human living environment,the humanoid robot model is built.Based on this,mission requirements of humanoid fast operation is analyzed,and proposed the joint drive control and motion trajectory planning matching implementation to realize humanoid fast operation.The humanoid robot control system hardware and software design is completed.Secondly,by analyzing the movement of the body during rapid joint operation,instantaneous power of the drive controller design ideas is proposed for humanoid joint motor drive.For humanoid fast operation demand instantaneous output capacity for the establishment of joint control model,the instantaneous energy of the motor control algorithm based on current integration which can be greatly enhanced by the joint instantaneous power control algorithm.Meanwhile,fuzzy control methods is used to adapt to volatility when high power output state.The joint drive controller hardware and software design is completed for humanoid robot system.Third,take table tennis fast movement as an example,for driving under the premise of joint capabilities,while meeting the requirements of the job tasks and stability,optimal energy trajectory planning method is proposed.Under the premise to meet the mission requirements batting,performed in order to reduce the difficulty of the robot joint trajectories.Using the energy optimal strategy to determine the minimum speed of the robot arm and internal joints configured,so the movement of the robot in the fast hit will has minimum energy consumption.According to the tasks require,the humanoid robot motion initial posture,arm movement time and the trajectory is designed.Then,in order to make full use of joint drive ability to complete the fast motion,matching algorithm of joint drive and trajectory planning is studied.Establish the constraint function by analysis of joint driving capability.Take fast walking as an example,using similarity function to complete trajectory planning based on the ability.Meanwhile,using joint drive performance to implement error compensation method for reduced articulation errors to make humanoid robots more effective to achieving fast operation.Finally,using the simulate joint,table tennis hitting operation and fast walking as an example,the proposed methods for joint drive method,trajectory planning and the matching algorithm are verified on a physical humanoid robot.