Introducing Various Novel Optimization Techniques for Task Allocation in Multi-Vehicles Systems

Several applications have been implemented since the rise of autonomous and intelligent vehicles. These applications include, but are not limited to, self-driving cars, warehouse management and automated factories. However, in the past few decades, research in this field was heavily directed towards the investigation of cooperative Multi-Vehicles Systems and their capabilities. Several challenges face such systems but the task allocation problem attained high attention from the research society due to its high importance and complexity. In this work, four different algorithms were utilized to solve the task allocation problem for a team of homogeneous vehicles and tasks. Two of which used optimization based approaches and the other two utilized the market-based approach. The algorithms were intensively tested on various scenarios to evaluate their performance in different situations. A comparative study between the algorithms is held at the end. Results show that the market-based approaches yields better results in large-scaled problems in terms of cost compared to optimization based approaches. Also, the market-based approaches outperform the optimization based approaches in terms of computation time in all cases. However, optimization-based approaches yield better costs in small-scaled scenarios.