Ant Colony Optimization Based Force-Position Control for Human Lower Limb Rehabilitation Robot

Abstract

The aim of human lower limb rehabilitation robot is to regain the ability of motion and to strengthen the weak muscles. This paper proposes the design of a force-position control for a four Degree Of Freedom (4-DOF) lower limb wearable rehabilitation robot. This robot consists of a hip, knee and ankle joints to enable the patient for motion and turn in both directions. The joints are actuated by Pneumatic Muscles Actuators (PMAs). The PMAs have very great potential in medical applications because the similarity to biological muscles. Force-Position control incorporating a Takagi-Sugeno-Kang- three- Proportional-Derivative like Fuzzy Logic (TSK-3-PD) Controllers for position control and three-Proportional (3-P) controllers for force control. They are designed and simulated to improve the desired joints position specifications such as minimum overshoot, minimum oscillation, minimum steady state error, and disturbance rejection during tracking the desired position medical trajectory. Ant Colony Optimization (ACO) is used to tune the gains of position and force parts of the Force-Position controllers to get the desired position trajectory according to the required specification. A comparison between the force-position controllers tuned manually and tuned by ACO shows an enhancement in the results of the second type as compared with the first one with an average of 39%.