Multi-objective Optimization-based Trajectory Planning of the Distributed Actuation Mechanism for Variable Tasks

Abstract

The distributed actuation mechanism is a redundant manipulator that improves the performance of an end-effector (velocity and force) by relocating the actuating position of the slider along the link without changing its posture. However, due to the higher degree of redundancy, it is difficult for trajectory planning. This study presented a trajectory planning of the distributed actuation mechanism for variable tasks through maximum performance evaluation based on multi-objective optimization. First, a multi-objective optimization was formulated to evaluate the maximum end-effector performance for targeting a three-link manipulator based on the distributed actuation mechanism (DAM-3R). For comparison, an equivalent joint actuation mechanism-based three-link manipulator (JAM-3R) was developed, and numerical simulation was performed by connecting the multi-body dynamics analysis module and the optimization module. Based on the maximum performance evaluation result, trajectory planning was performed for a given variable payload task. Simulation results confirmed that the DAM-3R was faster and can lift larger payloads than the JAM-3R. In addition, by optimally utilizing the higher degrees redundancy of the DAM-3R, the energy consumption of the DAM-3R was 64.7% lower than that of the JAM-3R. These results indicate that the proposed method effectively performed the trajectory planning for the variable payload task for the redundant manipulators.