Adaptive Fuzzy-PD Tracking Controller for Optimal Visual-Servoing of Wheeled Mobile Robots

Journal of Control Engineering and Applied Informatics

Omer Saleem Bhatti¹ 

[1] FAST - National University of Computer and Emerging Sciences (NUCES), Lahore, Pakistan.

关键词: Wheeled mobile robot;    visual-servoing;    self-tuned proportional-integral-derivative controller;    adaptive fuzzy-logic controller;    least-square method;    particle swarm optimization.;   

DOI  :  

学科分类：农业科学（综合）

来源: Societatea Romana de Automatica si Informatica Tehnica

【 摘 要 】

This paper presents a framework for the development of a motion controller to optimize the tracking control of position and orientation of a differentially-driven wheeled mobile robot (WMR) in indoor environments using the visual feedback provided by an overhead camera. The visual information is processed to generate a collision-free trajectory for the WMR towards its destination. While traversing the planned trajectory, the extent of WMRs deviation in actual and desired posture is continuously checked via the visual feedback and motor-encoders. Two-layered control architecture is proposed for optimal visual-servoing of the WMR. The Proportional-Integral (PI) controllers are used as low-level motor speed controllers. An Adaptive-Fuzzy-Tuned-Proportional-Derivative (AFT-PD) control scheme is implemented in the high-level controller to remove the tracking errors. The Fuzzy-Logic (FL) controller serves to auto-tune the PD coefficients. The center(s) of the output membership functions of these FL controllers are adaptively updated via the least-squares method. The tracking performance of the proposed AFT-PD controller is compared with the PD controllers tuned by Particle-Swarm-Optimization (PSO) algorithm. The experimental results of the AFT-PD and PSO-PD control schemes are presented to validate the efficiency and robustness of the proposed scheme.

【 授权许可】

CC BY