I. Introduction
Multi-Robot Task Allocation (MRTA) is typically required in a number of applications such as search and rescue and precision agriculture [1]. Many of those applications involve a variety of concurrent tasks, which in turn requires the robots to have varying capabilities. One such example is disaster relief, where the team of robots must both seek and rescue the victims [2]. In such scenarios, where the team is heterogeneous, it is crucial to consider the specialization of each robot with respect to the different available tasks [3] to perform an effective task allocation.