Study On Kinematic Analysis And Position Control Of A Rigid-Flexible Robotic Arm

In recent years,with many advantages of high stiffness and precision,rigid robotic arms have been widely used in many fields such as industrial automation,healthcare,aerospace,and so forth.However,with the gradual in-depth understanding of robot application scenarios,it has been found that rigid robotic arms have great limitations in unstructured environments,such as earthquake search and rescue,minimally invasive surgery,and pipeline maintenance.As anew type of robots,flexible robotic arm gradually becomes the focus of attention because of its advantages such as high pliability,flexibility,and human-robot interaction safety.Starting from the design for a rigid-flexible robotic arm,this paper proposes anew model driven pneumatically with wire-pulled variable stiffness,and deep research on its kinematics and position control strategies is further carried out.The main contents of this paper are summarized as follows.A new model of rigid-flexible robotic arm is designed,which has the advantages of not only ensuring the flexibility of the arm but also improving its stiffness under the action of the pulling wire.With higher flexibility and lower cost of manufacturing and maintenance,the corrugated tubular airbag structure of the arm is optimized,and the influence of its structural parameters on the performance of the arm is then analyzed.The process of manufacturing and installation of the arm is also described in detail.Equating the flexible robotic arm to a rigid one with variable rod length,a new flexible-rigid equivalent model is proposed,and the kinematic analysis of the designed arm is carried out.Under the assumption of constant curvature,the designed arm is equated to a 2R-2P rigid openchain mechanism,with four degrees of freedom,and its equivalent model is established by the DH parametric method;meanwhile,the positive and inverse kinematic characteristics of the arm are analyzed.Finally,the proposed equivalent model is verified according to the comparison of the results by the two models based on arc length parameterization method and the D-H parametric method.Based on the technology of machine vision detection,the position control of the designed arm is experimentally studied.A miniature camera was set at the end of the arm,and a binocular camera was fixed at the side of the experimental table to establish a hybrid vision system;The YOLO v3 target recognition algorithm and binocular ranging algorithm were used to obtain the position coordinates of the target,and the input air pressure of the airbag was determined by the analysis of the inverse kinematic characteristics of the arm;the closed-loop control of the end position of the arm is realized through adjusting the input air pressure of the airbag by the PI controller.The experimental results show that the vision-based closed-loop control of the arm is able to realize the effective grasping of the target,and that the pull-wire variable stiffness approach is effective.