Research On Trajectory Tracking And Force Control Of Continuum Robot Based On Model-free Algorithms

This article is oriented to the specific control requirements of continuum surgical robots in actual surgical scenarios.Aiming at the current situation of traditional control algorithms with limited accuracy and poor anti-interference ability,based on multiple intelligent algorithms and combining the characteristics of continuum robot systems,two model-free position controllers and a model-free force controller have been developed for position control and force control in a complex surgical environment.The designed algorithms have the advantages of high motion accuracy,strong feasibility,good versatility,and low calculation cost.The detailed research content is as follows:Firstly,according to the structural characteristics and driving modes of the continuum surgical robots,two representative cable-driven single-section continuum robots were designed as prototype prototypes,and a screw-type driving platform was designed.Based on the assumption of constant curvature,the forward kinematics model of the continuum robot is derived,and the differential kinematics relationship is derived based on the Jacobian matrix,and the kinematics controller based on the constant curvature model is built.Finally,through simulation,the accuracy of the model-based controller is easily reduced by factors such as inaccurate parameters and external interference.Secondly,to solve the problems of the above-mentioned traditional control methods,two model-free position controllers are designed to get rid of the dependence of traditional methods on accurate models.This type of algorithm updates the pseudoJacobian matrix of the continuum robot by acquiring the real-time input and output data.The designed algorithm has the advantages of low computational cost,high control accuracy and anti-interference ability.Then,a model-free distal contact force controller of the continuum robot was designed,and the distal contact force control of the robot in a complex environment was realized through the model-free adaptive control algorithm.This method incorporates the unmodeled environmental effects and the robot’s systematic errors and uncertainties into the estimated value of the partial derivative matrix,without relying on the robot’s specific kinematics/mechanical model and prior knowledge of the robot’s contact with the environment,which has good versatility.Finally,on the basis of the above prototype design and algorithm design,experimental research was carried out.First,a preliminary simulation of the designed algorithm was carried out on the simulation platform,and a related experimental platform was built to verify the performance of the algorithm.Experiments show that the designed model-free position control algorithm can reach a position accuracy of1.18mm(0.59% of the length),which is 49% higher than the traditional model-based controller.Subsequently,the feasibility and stability of the force control algorithm were verified by platform experiments.