Optimal Robust Hybrid Active Force Control of a Lower Limb Exoskeleton

In this paper, an optimal robust hybrid active force controller based on Harmony Search Algorithm is designed for a lower limb exoskeleton robot. Dynamic equations are derived using Lagrange method by considering three actuators on the hip, knee and ankle joints to track a specific trajectory. One of the major problems of exoskeleton robots is non-synchronization of movements and transfer of power between the robot and human body which affects the robot in form of disturbance. In order to mitigate the effect of disturbances and increase precision, combination of active force control (Corrective loop of control input) with position control is used as an effective and robust method. In the active force control, to elicit robust input control against disturbances, the moment of inertia of the links is estimated at each instant by minimizing the Criteria of ITAE and the control input rate, using the Harmony Search algorithm and the control input is modified. Also, two controllers are designed for the position control loop using sliding mode and feedback linearization methods. In order to validate the performance of the designed controllers, the robot is modeled in ADAMS and control inputs are applied to the Adams model. For appropriate comparison, all control parameters are optimized using the harmony search algorithm and then performance of position controllers are compared in hybrid and conventional (without the force control loop). Results indicate the outperforming of the hybrid sliding mode controller rather than to the other designed controllers.