Trajectory Control of an Active and Passive Hybrid Hydraulic Ankle Prosthesis Using an Improved PSO-PID Controller

The development of intelligent prostheses has provided convenience and confidence to amputees’ lives. For the majority of patients that only use a passive ankle prosthesis structure without its corresponding function, the development of intelligent ankle prostheses is very important. In this paper, a controllable active and passive hybrid hydraulic ankle prosthesis (APHHAP) with active drive, precise damping regulation, and energy recovery functions is presented. Also, an embedded control board capable of communicating with motors, a computer, and a mobile phone terminal as well as connecting various types of sensors for real-time monitoring of an ankle prosthesis motion state is designed. This board is based on STM32F429ZI. For a better control performance, generalized opposition-based learning, variable parameters based on the sigmoid function, and adaptive elite mutation are introduced to improve the traditional particle swarm optimization (PSO) algorithm. The new algorithm is called improved particle swarm optimization (IGOPSO) algorithm. The IGOPSO algorithm achieves better optimization and faster convergence than traditional PSO algorithms. Using simulation, which adopts a mathematical model based on a piecewise function, the parameter range values are obtained, and the number of invalid running times is reduced. Experiments on a physical prototype are conducted to validate the control algorithm performance. The obtained experimental results demonstrate that by combining IGOPSO with the proportion integration differentiation (PID) algorithm (IGOPSO-PID control algorithm), efficient track tracking control of the APHHAP dorsiflexion and plantarflexion as well as significant improvement in its control accuracy can be achieved.