Tuning of PID Controller Parameters of a Biped Robot using IWO Algorithm

Abstract

This paper presents a recently established stochastic optimization approach that has been stimulated from the behavior of weed colonization for biped robot applications. The aim of using the algorithm is to tune the gains (i.e. Kp, Kd and Ki) of the PID controller used by a biped robot while walking over a flat terrain. The dynamics of the biped robotic mechanism is derived after using Lagrange-Euler formulation. These dynamic equations are further used to design the PID controller for each joint of the biped robot. Initially, the performance of the Invasive Weed Optimization (IWO)-tuned PID controller is compared in terms of error and the torque required at various joints. Further, the IWO tuned PID controller is tested on a real 18-DOF biped robot and found that it has successfully negotiated a flat terrain with the help of a dynamically balanced gait generated by the controller.