| Artificial jaw robot can reproduce the chewing motion and occlusal force of human jaw and have a wide range of applications in the fields of dentistry,food science,biomechanics and medical rehabilitation.An artificial mandibular robot with four degrees of freedom and six cable drives is proposed,which is based on the bionic principle of the masticatory muscles of the mandibular system and the temporomandibular joint.Spatial analysis and functional chewing motion planning,dynamic model,optimal distribution of each branch chain driving force and motion control research are stuided.Firstly,based on the characteristics of the human mandibular masticatory muscles and temporomandibular joint,a cable-driven redundant mandibular robot with point contact and high pair constraint is proposed.According to the driving characteristics of the mandibular driving muscles,the three main driving masticatory muscles of masseter,temporalis and lateral pterygoid muscles are determined.The coordinates of the attachment points of the three muscles are determined.The movement form of the condyles on both sides are analyzed,and the symmetrical mandibular fossa articular grooves and condylar rods are designed to simulate the left and right temporomandibular joints of the human body.The overall design of the artificial jaw mechanism is introduced.The cable-driven linear pulley mechanism is used to simulate the masticatory muscles.Secondly,kinematic modeling and performance analysis of the designed artificial mandibular robot.Static and dynamic coordinate systems are separately established at the center point of the base and the center points of the left and right condyle balls.According to the design of the point contact high pair on both sides,the robot has four degrees and the four independent positions of the moving platform are selected.Pose parameters,and inverse kinematics and inverse Jacobian matrix derivation are determined.According to the displacement limitation of the drive rod,Hook hinge and ball pair of the designed mechanism,the mandibular incisal point is selected as the observation point of the workspace measurement.The mechanism working area of the mandibular incisal point is calculated.The model of the robot is established by ADAMS to simulate the common functional movements of the human jaw.The expansion and contraction of each muscle branch is analyzed.Then,the dynamic analysis of the designed artificial mandibular robot is carried out through the first kind of Lagrangian equation.Therefore,the driving force of each branch chain of the moving platform should be calculated.Except the four independent generalized coordinates of the moving platform,six constraint parameters of the chain are introduced to determine the calculation equation of the six branch chains.For the redundant driving characteristics of the artificial mandibular mechanism,the obtained coefficient matrix is not full rank.Therefore,it is necessary to optimize the distribution of the driving force of the six branches according to the specific optimization objective.Then,the dynamic analysis of the mandibular robot is performed through ADAMS,and the dynamic calculation of the movement trajectory is programmed.Through the co MParative analysis,the accuracy of the established dynamic model is verified.Finally,construct the artificial mandibular robot test platform and conduct motion control experiments.Firstly,the common three-loop control model of brushed DC motor is introduced.The driving function of the motor is calculated by the motor manual.A torque feedforward compensator for motor control is proposed according to the dynamic model of the jaw mechanism.Then,control models are established for the six motors in Simulink.The required motion and torque parameters are obtained according to the planned opening and closing motion.The control results of individual position speed control and adding torque parameters are calculated respectively.Finally,the motor software and hardware control system of the imitation mandible mechanism is constructed.The optimization effect of the force feedforward control model is verified. |
