Chen Wang

Topic

Severe Faults Accommodation with Formation Reconfiguration in Multi-robot Transportation System

Department of Mechanical Engineering

Date & location

• Thursday, September 11, 2025

• 10:30 A.M.

• Virtual Defence

Reviewers

Supervisory Committee

• Dr. Daniela Constantinescu, Department of Mechanical Engineering, UVic (Supervisor)

• Dr. Yang Shi, Department of Mechanical Engineering, UVic (Member) 

External Examiner

• Dr. Panajotis Agathoklis, Department of Electrical and Computer Engineering, UVic 

Chair of Oral Examination

• Dr. Marc Klimstra, School of Exercise Science, Physical and Health Education, UVic

   

Abstract

This thesis focuses on handling severe faults in multi-robot transportation systems. It proposes a fault-tolerant framework that integrates optimization models to generate an optimal formation for robots to carry the object for transportation, the Improved Artificial Potential Field (IAPF) method for path planning and motion control, and a proposed strategy of robot cooperation to keep the object balanced all the time.

Multi-Robot Systems (MRSs) offer significant advantages in terms of collaborative intelligence, task expandability, and system reliability. However, addressing severe faults such as motor burnout or structural damage still presents challenges in generating new formations for the remaining healthy robots, as well as reallocating tasks and planning paths during reconfiguration. For MRSs whose mobile robots cooperatively transport an object by supporting it from below, this thesis proposes a path planning framework that ensures that the object reaches the desired destination even when several of the robots fail during transportation. The proposed framework distributes the robots so as to maximize a proposed measure of transportation stability. Specifically, the objective of the constrained optimization that yields the spatial distribution of the robots depends on the distances from the centroid of the object to each robot and to the boundary of the robot formation, with constraints including distribution range, collision avoidance, and the centroid of the object remaining in the convex hull of the robot formation. When faults occur, faulty robots become obstacles, and the same optimization framework generates a new formation for the remaining healthy robots. The Hungarian algorithm serves to construct a distance cost matrix for task reassignment, to minimize the total distance over which the robots move and to reduce energy consumption during reconfiguration.

For path planning and motion control, an IAPF method integrates attractive forces, obstacle and boundary repulsive forces, and a temporary target point strategy to escape local minima. A collaborative strategy based on half-plane determination and triangular region analysis verifies that the robot formation always includes the centroid of the object in its interior, ensuring that the robots continue to balance the object during reconfiguration. Considering the transportation system a coupled rigid body, IAPF guides its centroid to avoid obstacles and to resume the task. MATLAB simulations validate the framework across various scenarios in which MRSs with various numbers of robots transport different convex polygonal objects, which, when severe faults happen, need to reconfigure themselves and to resume the transportation while avoiding the failed robots and other obstacles in the environment. Sequential Quadratic Programming (SQP) is applied to solve the optimization problems. Results illustrate that the proposed framework effectively handles severe faults, ensuring continuous and stable transportation tasks. Future work will focus on real time reconfiguration without task suspension in dynamic environments, constructing a coupled model considering robot and object dynamics, introducing force-feedback collaborative control strategies, and extending to non-holonomic robot scenarios.