Dexterous Hand Design And Optimization Based On Cable Drive

As an end effector of a humanoid robotic arm,a dexterous hand determines the level of anthropomorphism and intelligence of the robot.The ubiquity of dexterous hands and the complexity of the work environment place new demands on their development.The development of a highly cost-effective,compact,highly humanoid general dexterous hand has become one of the research hotspots in the field of bionic robots.This thesis mainly innovates the transmission method and grasping plan of dexterous hands.The combined transmission method of single and double lasso and torsion spring effectively reduces the volume of the dexterous hand,and the gripping plan based on the normal force ensures the dexterous hand's gripping performance.The thesis focuses on the structural design,optimization and grasping planning of dexterous hands.The main research contents of the thesis include:(1)The development status of dexterous hand design and grasping planning at home and abroad is introduced.Through the development trend and bionic analysis of the dexterous hand,a flexible cable transmission method with an external motor is used to determine the mechanical structure.The five fingers adopt a modular design,the overall size is equivalent to that of a human hand,and it has the characteristics of small size,light weight and high flexibility.(2)Based on the transmission model and experimental analysis,the transmission characteristics of single and double Tendon-sheath are studied.Based on the mathematical model of the Coulomb friction criterion,the transmission error of the system is calculated,and the minimum preload of the system is determined.The constructed single and double Tendonsheath joint experimental platform verified the correctness of the model,summarized the law of the factors affecting the system transmission error,and performed position compensation for the double Tendon-sheath transmission.(3)Based on the kinematics equation of the finger position,the thumb position is optimized.By analyzing its workspace and differential kinematics,a joint Jacobian matrix is established.Using the condition number of this matrix as an index,the size of the thumb is optimized.The shape closure principle and geometric principle are required to realize the optimization of the thumb position.(4)Through the dynamic calculation of fingers and grasping simulation,the grasping ability of dexterous hands is verified.Calculation of finger dynamics based on Lagrange method.Then use ADAMS to simulate the precise grasping and strong grasping of fingers.Later,joint simulation of MATLAB and ADAMS is performed to realize single joint position control of fingers.(5)A grasping plan based on the normal minimum force is proposed,and the results of the differential evolution algorithm are used to capture the minimum singular value of the matrix to verify the reliability.The flexibility and physical grasp experiments are conducted on the dexterous hand experimental platform to verify the dexterity and load capacity of the dexterous hand.