| Robots have been widely used in military,industrial,scientific exploration and many other fields.Traditional rigid robots are generally connected by rigid modules through motion pairs.This kind of robot has the advantage of accurate motion,but the rigidity of the structure makes its environmental adaptability poor and has high control requirements.The soft robot is made of flexible materials and realizes the interaction with the target through active and passive deformation.It has strong environmental adaptability and high flexibility,but its load force is small,and its moving accuracy is easily affected by the environment.However,the rigid-flexible coupling mechanism combines the advantages of the two.In this paper,therefore,a modular rigid-flexible coupling software manipulator with variable stiffness is proposed.Its structure is composed of multiple identical moving modules.Each moving module realizes the switching between multiple metamorphic states through variable bending stiffness module and regulating module to carry out multiple modes of motion.At the same time,the variable stiffness performance of the variable bending stiffness module and the regulating module,as well as the ability of the moving module to realize a variety of motions and anti-interference are verified.The forward kinematics of the moving module under different metamorphic states is derived.At the same time,the static performance of the moving module is analyzed by using the strain energy density equation and force balance equation combined with ABAQUS simulation analysis.The pose of the moving module is monitored based on machine vision.Through color filtering and contour recognition,the coordinate information of feature points is obtained and transformed into the relevant parameters of the moving module.The moving module is closed-loop controlled by PID control method.The serial communication is established,and the PID control method is used to control the moving module in Arduino UNO.The pneumatic circuit design and control system design are completed,and the experimental platform is built to test the grasping ability of the flexible manipulator.Finally,a real-time grasping condition monitoring method of manipulator based on machine vision is proposed. |