Robust Tracking Controller Design for an Omnidirectional Mobile Vehicle

by Ya-chao Yang

Institution: NSYSU
Department: Mechanical and Electro-Mechanical Engineering
Degree: PhD
Year: 2015
Keywords: Omnidirectional wheel; Sliding mode control; Robust adaptive control; Parameter estimation; Mobile Robot
Record ID: 1387823
Full text PDF: http://etd.lib.nsysu.edu.tw/ETD-db/ETD-search/view_etd?URN=etd-0028115-162950


This dissertation presents a robust trajectory tracking control scheme for the omnidirectional mobile vehicle (OMV) to provide autonomous tracking capability. The OMV is equipped with three omnidirectional wheels, which allow the vehicle freely changes its moving direction regardless of the steering constraint. All the tracking control algorithms are developed on the basis of a robust adaptive sliding mode controller (ASMC) framework. The trajectory tracking performance of OMV is therefore ensured by utilizing the ASMC approach, which copes with system uncertainties and unknown external disturbances. Besides, in order to let the mobile vehicle own a complete sight around its environment, the vehicle is also supplied with an omnidirectional camera system. For the purpose of object tracking missions, the corresponding image processing techniques for the omnidirectional images are also established to extract the characteristic features of the target object. Through analysis of the image information, the position of the target object can be located in three-dimensional space. Based on this information, the proposed ASMC method is able to calculate suitable control efforts for those three omnidirectional wheels to drive the mobile vehicle towards the target object within a predetermined distance while overcoming possible uncertainties and disturbances. Actual tracking experiments including the trajectory tracking and the object tracking are conducted on an experimental omnidirectional mobile vehicle. Three different paths, a square path, an eight shape path, and a circular path, are investigated for trajectory tracking tasks. Comparison of tracking performance using the conventional SMC and the presented ASMC is also made. Experimental results demonstrate the feasibility and effectiveness of the proposed tracking control scheme. Furthermore, the superiority of the presented control method than the SMC is also verified.