基于惯性动作捕捉的下肢外骨骼步态生成与仿真验证OA

Aiming at the problems of poor adaptability and insufficient human-machine coordination in traditional gait generation methods for lower limb exoskeletons,this paper proposes a data-driven method based on inertial motion capture and multi-body dynamics simulation.The gait data of healthy individuals and patients are collected by the Noitom system.The joint angle trajectories are generated by using spline curve fitting,and a dynamic adaptation algorithm is designed to achieve real-time adjustment of gait planning.A human-machine joint simulation model is constructed in ADAMS,and the CUBSPL spline is used to drive joint movement for systematic kinematic and dynamic analysis.The results show that the generated exoskeleton gait is highly consistent with the natural gait in terms of joint angle,displacement,velocity,and torque,which verifies the effectiveness of the method.This study provides a complete technical path for personalized gait planning and rapid verification of lower limb exoskeletons,and has reference value for improving the adaptability and safety of rehabilitation training.