Smooth Sub-Phases Based Trajectory Planning for Exoskeleton System

摘要

Trajectory is a crucial part of exoskeleton robot design. The appropriate trajectory will affect wearer's safety, health and comfort. The trajectory profile of angular position can be generated as same as actual trajectory profile in term of shape. However, the problem of unreachable intermediate points increases the error between the polynomial trajectory profile and real human trajectory profile. This error directly can affect to generate a less accurate profile that can’t be accurately matched to the human trajectory profile. Based on the aforementioned problem, the objectives of this paper is to investigate the effect of having a different number of sub-phases trajectories on the error related to the real human trajectory profile. Also, this paper presents trajectory generation method that is called quintic polynomial segment with 6th polynomial blend (6-5-6 PSPB). The 6-5-6 PSPB trajectory includes the via points at initial and final of each phase in order to solve the problem of unreachable intermediate points. The result shows a good improvement (almost 8%) in term of error based on the proposed 6-5-6 PSPB technique compared to the three categories of gait analysis. The 6-5-6 PSPB technique shows a good result based on the root mean square error (RMSE) and average difference (AD) error of angular position (0.3146 and 0.031310 Degree respectively). At the end, 6-5-6 PSPB can improve the error and generate a trajectory motion profile that accurately matches to human trajectory profile.