Mechanism Design And Motion Control Of A 2D Variable Stiffness Snake-like Robot

Biological snakes have better flexibility and adaptability in a complex environment.Lots of snake-like robots have been developed and applied in the fields of emergency rescue,forestry,military,civil and medical treatment.With the development of robot technology,robot joint actuation can be divided into two types: the rigid actuation joint and flexible actuation joint.Although the rigid actuation joint has better control precision,it reduces the environmental adaptability of the snake-like robot.Therefore,this paper studies the motion control and gait optimization of a two-dimensional variable stiffness snake-like robot.The main research contents are given as follows.Firstly,based on the kinematics model and dynamics model of the snake-like robot with traditional rigid actuation joints,the kinematics model and dynamics model of the variable stiffness snake-like robot are established in combination with the structural characteristics of the flexible actuation joints.Based on the control variate method,the analysis of model parameters on the effect of the movement of energy consumption and speed of snake-like robot is discussed.The results show that the phase difference and the angular of the snake-like robot had a greater influence on the movement of energy consumption,and the energy consumption can be reduced and the influence on the motion speed is small by controlling the phase difference.The angular frequency has great influence on both the movement of energy consumption and the motion speed.The joint stiffness has no effect on energy consumption and motion speed.The joint parameters have the optimal control points on the energy consumption and motion speed of the snake-like robot.Secondly,based on the design concept of series elastic actuators,this paper designs a flexible joint driven by two motors for a snake-like robot,and based on the joint stiffness control requirements of the two-dimensional snake robot with variable stiffness,a motion controller for the snake-like robot was proposed based on the two-layer Central Pattern Generator(CPG).The two-layer CPG included rhythm layer and pattern layer.However,the traditional motion controller of a snake-like robot based on the Central Pattern Generator(CPG)only can control the joint trajectory and phase.The joint stiffness control of the two-dimensional variable stiffness snake-like robot cannot be implemented.Therefore,a motion controller for the variable stiffness snake-like robot based on double-layer CPG is proposed in this paper.The double-level CPG contains rhythmic neuron layer and pattern neuron layer.The model of CPG neurons is based on the Kuramoto nonlinear oscillator.The CPG neurons in the rhythmic neuron layer determine the driving phase of the motor units on both sides of each joint.The CPG neurons in the pattern neuron layer can determine the phase relationship,output trajectory and joint stiffness of the joint.Finally,in order to verify the analysis results of the kinematics model and dynamics model of the two-dimensional variable stiffness snake-like robot and the control performance of the motion controller based on the double-layer CPG,a co-simulation environment based on Simulink and MSC.Adams is established in this paper.Meanwhile,a prototype of a two-dimensional variable stiffness snake-like robot is developed.Different co-simulation test conditions and different experimental test conditions were respectively designed for verification.The movement performance of the two-dimensional variable stiffness snake-like robot under different conditions of angular frequency,stiffness and amplitudes was analyzed.The simulation and experimental results show that the double-layer CPG proposed in this paper can realize the motion control of the two-dimensional variable stiffness snake-like robot.The motion energy efficiency of the two-dimensional variable stiffness snake-like robot can be effectively improved by optimizing the joint parameters and control parameters.By adjusting the joint stiffness,the energy consumption of the snake-like robot can be reduced by 30%,and the speed can be doubled.Compared with the rigid snake-like robot,the energy consumption of the variable stiffness snake-like robot can be reduced by 27%.By adjusting the amplitude of joints,the motion efficiency of the snake-like robot is reduced by 65%,and the motion speed by four times.By adjusting the phase difference,the energy efficiency of the snake-like robot can be reduced by 72%,and the snake-like robot can reach the maximum speed of 1.94cm/s.