Research on planning and optimization of trajectory for underwater vision welding robot

Abstract

To achieve welding torch trajectory continuous smooth and welding time the shortest, which of the underwater visual welding robot end, a method of interpolation the location-time discrete points of each joint based on the five b-spline curves were proposed. The start-stop continuous have been smooth for welding by this method, then the sequential quadratic programming filter trust region method was used to optimize the time by the constraint condition of the kinematics of robot, so the efficiency of welding was improved. Finally, the simulation experiment showed that continuous smooth trajectory of each joint for the vision robot, meanwhile welding was used less time to accomplish, and the properties of motive controlling and trajectory tracking were improved for the welding robot.