A Wearable Sole-Assist Device: Altering the Walking Center of Gravity Trajectory

Studies have shown that exoskeleton robots can enhance the wearer's mobility through joint assistance, typically targeting areas such as the hips, knees, shoulders, and elbows. However, current exoskeleton designs rarely consider assisting the overall center of gravity (CoG) of the human body. During walking, the CoG fluctuates vertically by approximately 5-8 cm, and overcoming gravitational forces during these fluctuations accounts for about 65% of the total energy expenditure. To address this issue, this paper proposes a novel wearable sole-assist device (WSAD) that reduces the amplitude of CoG fluctuations by actively controlling four parallel servo motors and passively balancing with high-strength compression springs. This device aims to alter the trajectory of the CoG during walking, thereby reducing the energy consumption of the human body. Using the Isaac Gym platform and the PPO reinforcement learning algorithm, simulation experiments were conducted with the optimization goals of reducing CoG fluctuations and minimizing energy consumption at each joint. According to the simulation results, this could result in an approximate 20% to 41% reduction in energy metabolism.