Shape Control And Heart Motion Tracking Control Of Soft Robot

Inspired by mollusks and organs,bio-inspired soft robots have attracted more and more attention of the researchers.Compared with the rigid robot,the soft robots,which are usually made of flexible and elastic materials,have better deformability,environmental adaptability and interaction security.Due to the performance of motion and operation,soft robots are suitable for applications in dynamic,uncertain and unstructured environments,such as medical operations,disaster relief,military reconnaissance and scientific exploration.The shape of soft robot studied in this paper like the octopus arm or elephant trunk.The material of the soft robot is silica gel.According to the muscle contractions,four cables are embedded inside the soft robot and attached to a metal ring.Using servo motor to change the length of cables,the robot can bend into a circular arc.How to control soft robot to finish specific tasks(such as heart ablation)is a challenging work.In the process of moving of soft robot,a three-dimensional shape control algorithm is proposed for avoiding the obstacles by controlling the shape of soft robot.Based on the constant curvature and torsion of the soft robot's centerline,the shape parameters which include curvature,torsion and deflection angle are defined.The cylindrical helix equation is introduced for complete description of the shape of the manipulator.Then,the functional relationships between the lengths of four cables and shape parameters are obtained.The inverse Jacoby matrix which can represent the relationship between the change rate of cables' length and the change rate of the shape parameters could be obtained.Based on the matrix,the controller for shape control is designed.The experiments verify the validations the algorithm.After soft robot reaches the target position,an iterative learning controller(ILC)with heart motion compensation is proposed for tracking the target point on the heart surface.Using the deformability of soft robot,the robot can keep a fixed depth distance between the end-effector and the target point on the heart surface and make the end-effector toward the target point.Based on the sensors' information and robot model,the real-time coordinate of the target point in the base coordinate system can be obtained,which represents the heart motion information.A weight linear frequency combiner is used to model the heart motion.Based on the heart motion model,the real-time compensation can be calculated.At the same time,the period of the heart motion is obtained by the peak detection of the motion trajectory.The obtained period is used to modify the iteration period of the iterative learning controller.The final output is composed by the output of the ILC and the motion compensation.The experiments prove the effectiveness of the heart motion tracking algorithm.