Robot RBF fuzzy PID motion control based on Sine Chaos-improved particle swarm optimization

Abstract

Pipeline robots play a crucial role in ensuring public safety by routine inspection and cleaning operations. However, their performance is frequently hindered by complex working environments and unexpected collisions. To overcome these challenges, we propose a control algorithm that integrates an enhanced particle swarm optimization (PSO) method with sine chaotic mapping and a radial basis function (RBF)-based fuzzy Proportional–Integral–Derivative (PID) control (IP-PID). Specifically, sine chaotic mapping is utilized within the PSO framework to increase population diversity. Furthermore, adaptive inertia weights and compression factors are introduced to strengthen the PSO’s global search capability and convergence stability. The optimized PSO is then employed to tune the proportional and quantization parameters of the fuzzy PID controller, thereby enhancing its adaptability and control accuracy. To validate the effectiveness of the proposed approach, we conduct comprehensive comparisons with the Back Propagation—Proportional-Integral-Differential (BP-PID) controller, Model Predictive Control (MPC) controller, cascade feed-forward neuro-fuzzy PID controller (CFF-NFPID), and optimal hybrid interval type-2 fuzzy PID + I logic controller (OH-IT2FPID + I) using Simulink simulations and physical experiments. The simulation results demonstrate that, compared to MPC, IP-PID reduces settling time by 83.9% and decreases overshoot by 85%. In both velocity regulation and trajectory tracking tasks, the proposed approach achieves substantially improved reference-tracking accuracy. Experimental results further corroborate the superiority of IP-PID, demonstrating higher control precision, enhanced tracking performance, and greater robustness against external disturbances.