Bionic Design and Optimization of a Rigid-Soft Hybrid Knee Exoskeleton

Knee exoskeletons have great potential in gait training and intervention for patients with knee osteoarthritis after surgery. Traditional single-axis rigid exoskeletons suffer from issues such as large joint volume, high inertia, and joint misalignment, significantly reducing wearer comfort. Soft exoskeletons can reduce the size and weight of artificial joints but lose the load support and mechanical limiting capabilities of rigid structures. This paper proposes a novel hybrid rigid-soft knee exoskeleton that employs a cross four-bar mechanism with loaded springs to mimic the multi-rotational center motion of the knee joint, avoiding joint misalignment and reducing the driving energy consumption of the exoskeleton. A genetic algorithm is used to optimize the structural parameters of the links and the stiffness coefficients of the springs by minimizing a weighted cost function composed of the deviation between the exoskeleton's rotation center and the user's knee joint, as well as the total potential energy of the system. Finally, a kinematic simulation is conducted to determine the linear relationship between the knee joint angle and the motor angle.