Research On QPSO-Algorithm-Based Multi-target Trajectory Planning Of Dual Robot Coordinated Assembly

With the continuous advance of assembly automation process, the tasks of assembly robot to be completed are becoming increasingly complex, single robotic assembly has been increasingly unable to meet current mission requirements. In order to adapt to the increasing complexity and continuously improving the intelligence of the assembly tasks, it is needed for dual robot assembly operations coordinated research. Although dual-robot coordination assembly system has good flexibility and adaptability advantages, it improves the complexity of the coordination exercise program of dual robot system. The problems of robot obstacle avoidance algorithm, dual robot trajectory planning and optimization algorithm should be solved quickly. In this dissertation, dual 6-DOF serial robot coordinated assembly trajectory planning is studied to provide an effective solution for the key issues of the dual robot coordination assembly in the fully automated production line. The main research work is as follows:(1) An assembly position error and assembly attitude error are defined in this dissertation. In order to improve the success rate of the assembly by analyzing the error generated in the assembly process, the geometry models of assembly position error and assembly attitude error is proposed, which are applied to the dual robot coordinated assembly trajectory planning.(2) A planning strategy of dual robot coordination assembly trajectory is proposed in this dissertation. Through analyzing the lack of trajectory planning's method of single robot assembly, combined with the analysis of the dual robot kinematics coordination, a trajectory planning method which has B spline interpolation error compensation function in dual-robot coordination assembly is given. The results show that compared with single robot to complete the same assembly task, the trajectory planning method of dual robot coordinated assembly presented in this dissertation can more effectively eliminate the spline interpolation error.(3) A deep-searching obstacle avoidance strategy of dual robot is proposed in this dissertation. After a collision is detected in dual robot performing tasks, the collision-free path is obtained by the deep-searching strategy to achieve the purpose of dual robot obstacle avoidance. In deep-searching obstacle avoidance path process, after artificially determine the approximate range of the robot obstacle avoidance through the middle point, this dissertation uses PSO algorithm to perform heuristically search for the middle point of obstacle avoidance, guiding the robot to avoid obstacles, which can greatly reduce search space of the algorithm to improve the success rate of avoiding the collision.(4) Multi-objective optimized trajectory of dual robot coordinated assembly is solved by QPSO algorithm. In order to avoid the robot jittering and improve the efficiency and reliability of the dual- robot coordination assembly, high-order B-spline curves to construct joint trajectory of dual robot is used in the dual-robot coordination assembly process. Considering the kinematic constraints of the robot, task executing-time and the smoothness of the assembly trajectory as optimal objective, the best collision-free trajectory of dual robot coordinated assembly is obtained by the QPSO algorithm.