An Integrated Design Framework for Path Planning and Trajectory Tracking of Mobile Robots with Application in Tank Inspection

  • Jiameng Xue

Student thesis: Phd

Abstract

Wheeled Mobile Robots (WMR) are widely used in varieties of fields. For instance, cargo transportation, space exploration, pipeline and storage tank inspection etc. One of the significant issues for WMRs is navigation, which is defined as a process of planning and directing a path in the environment where a mobile robot must drive safely from one position to another without getting lost or colliding with obstacles. Generally, there are three problems in navigation: positioning, path planning and motion control. In this thesis, an integrated design framework is proposed for path planning methods and trajectory tracking problems of mobile robots with application in tank inspection. This work will address the problem of the conversion process from path planning to trajectory tracking. For real application, the path planning and the tracking control for wall-climbing WMRs on a cylindrical tank will be discussed. First, for mobile robot path planning, a modified D* Lite path planning method are presented, which consists of the safe distance consideration, the adjacent node expansion and the path smoother design. The purpose of the modified method is to make the path smoother to some certain degree. Hence, the trajectory tracking for a WMR will be more feasible. Second, a quintic-based trajectory planning algorithm is presented to determine the position and velocity profile. It is based on quintic G2 splines under the consideration of velocity and acceleration constraints. As a result, the continuity in velocity and acceleration could be ensured throughout the whole curve length. Third, the Differential Wheeled Mobile Robot (DWMR) configuration is taken into account as it is suitable for real application. A wall-climbing DWMR is designed especially for tank inspection tasks. Fourth, Sliding Mode Control (SMC) is introduced for tracking control of mobile robots as its robustness to system uncertainties and external disturbances, easy implementation and fast response. Finally, in order to apply the path planning and trajectory tracking method for the designed robot, a cylindrical tank experimental platform was built. A multi-sensor data fusion positioning technique is presented. In a cylindrical coordinate system, a hybrid control strategy is proposed for trajectory tracking of WMRs based on the Backstepping Sliding Mode Control (BSMC). The simulation and prototype experiments results indicate that the framework has a good performance for real applications.
Date of Award1 Aug 2023
Original languageEnglish
Awarding Institution
  • The University of Manchester
SupervisorAlexandru Stancu (Supervisor) & Long Zhang (Supervisor)

Keywords

  • backstepping control
  • sliding mode control
  • trajectory tracking
  • velocity planning
  • trajectory planning
  • D* Lite
  • tank inspection
  • climbing robot
  • mobile robot
  • path planning

Cite this

'