General Dynamics Modeling And Optimal Control Method For Redundant Actuated Manipulators

With the rapid development of the electronics industry,3C products are becoming more miniaturized and frequently updated,and the workload is increasing because most of the assembly work is done manually.In order to better improve the quality of product assembly,many manufacturers are purchasing and using existing high-speed,lightduty robots as auxiliary systems to assist in the manual assembly process.Ordinary industrial robots do not have redundant degrees of freedom for capability optimization,fault tolerance and obstacle avoidance,thus limiting their development in collaborative tasks.To address these issues,thesis develops the following aspects of research with a self-developed planar 4R redundancy robot.First,thesis states the research background and significance of the topic,summarizes the advantages and disadvantages of drive centralized arrangement,total performance optimization,fault tolerance control and obstacle avoidance methods and the current research status,and condenses the corresponding research key points.The structural design of the redundancy degree robot,inspired by the rope drive here,proposes a drive centralized arrangement scheme based on synchronous belt as power transmission,which greatly reduces the robot joint load,improves the operation efficiency and fundamentally reduces the energy consumption.A reasonable experimental platform is provided for the subsequent algorithm.Secondly,the kinematic inverse solution of the pre-defined space reduction method based on the weighted M-P pseudo-inverse is proposed,which lays the foundation for the subsequent fault-tolerant control.The Lagrangian method is used for dynamics modeling and the algorithm is verified by ADAMS,which provides a model basis for the subsequent total performance optimization.Multiple indicators of total energy,total torque and generalized end time are weighted and fused to obtain the optimal end time by particle swarm optimization,and the data before and after optimization are compared to verify the rationality of the optimization results.From the second time,thesis starts from improving the fault-tolerant performance of the robot,and the performance indexes of each moment when the robot is trajectory tracking are processed by principal component analysis to obtain a comprehensive evaluation function,which is used as an evaluation index to reasonably schedule the joints that affect the overall performance the most after a failure occurs,so as to achieve the optimal resource coordination allocation strategy and reduce the impact of the robot on the overall performance after a failure occurs.Again,in order to guarantee the safety of the collaborators,thesis starts from the aspect of improving the collaboration ability of the robot to perform dynamic obstacle avoidance without affecting the end tracking trajectory.By establishing a reasonable loss function and assembling the gradient projection method to achieve redundancy degree robot dynamic obstacle avoidance,the effectiveness of the algorithm is verified through simulation.Finally,the above algorithm is verified by the experimental platform,and the results show that the simulation and experiment verify the effectiveness of the above algorithm.