Multi-objective trajectory planning and implementation of a metamorphic palletizing robot

This paper focuses on a metamorphic palletizing robot, elaborating on its working principles and analyzing its working trajectory. The primary aim is to address the complex challenge of multi-objective trajectory planning during the robot's motion, with a focus on minimizing time, energy consumption, and jerk. We present a general formula for optimizing multiple objectives, taking into account transformation characteristics based on actual working conditions. The optimization process employs the Non-dominated Sorting Genetic Algorithm with an elite strategy (NSGA-II), while Particle Swarm Optimization (PSO) is integrated into the optimization progression to identify specific metamorphic points. This approach ultimately produces a set of Pareto optimal solutions. From this set, the solution with the lowest time consumption is chosen as the definitive option for multi-objective planning. The joint driving functions of the robot during configuration transformations and within each configuration are analyzed accordingly. To ensure precision, the joint driving functions employed in the experiment are fine-tuned with pulse compensation values. Subsequently, experimental validation is carried out to verify the accuracy and practical feasibility of the multi-objective trajectory planning method.