Scheduling and coordination of multi-robot systems

In the last decades, Multi-Robot Systems (MRS) have received significant attention due to their intrinsic advantages over Single-Robot Systems (SRS). MRS, composed of multiple robots capable of cooperative task execution, exhibit redundancy, better scalability, simplified robot design and enhanced performance. MRS applications span exploration, surveillance, search and rescue and many more. One of the primary challenges in MRS is optimizing task allocation, known as the Multi-Robot Task Allocation (MRTA) problem, which is notoriously NP-hard. This thesis, conducted in collaboration with Leonardo company, employs the Tabu Search (TS) metaheuristic algorithm to minimize the total distance covered by a team of robots to complete the assigned tasks, constituting a complex operation. The application scenario involves a team of heterogeneous rovers extinguishing forest fires and aiding fire victims. Precedence constraints between the tasks can be specified. A comprehensive literature review of the prominent approaches and algorithms to solve the MRTA problem is conducted, motivating the choice of the TS algorithm. The TS algorithm is implemented in Visual Studio Code and outputs GANTT charts indicating the optimal task scheduling. Results, obtained by varying rovers and tasks configurations and tuning TS parameters, confirm the suitability of the approach.