Adaptive Task Allocation for Heterogeneous Multi-Robot Teams with Evolving and Unknown Robot Capabilities

For multi-robot teams with heterogeneous capabilities, typical task\nallocation methods assign tasks to robots based on the suitability of the\nrobots to perform certain tasks as well as the requirements of the task itself.\nHowever, in real-world deployments of robot teams, the suitability of a robot\nmight be unknown prior to deployment, or might vary due to changing\nenvironmental conditions. This paper presents an adaptive task allocation and\ntask execution framework which allows individual robots to prioritize among\ntasks while explicitly taking into account their efficacy at performing the\ntasks---the parameters of which might be unknown before deployment and/or might\nvary over time. Such a \\emph{specialization} parameter---encoding the\neffectiveness of a given robot towards a task---is updated on-the-fly, allowing\nour algorithm to reassign tasks among robots with the aim of executing them.\nThe developed framework requires no explicit model of the changing environment\nor of the unknown robot capabilities---it only takes into account the progress\nmade by the robots at completing the tasks. Simulations and experiments\ndemonstrate the efficacy of the proposed approach during variations in\nenvironmental conditions and when robot capabilities are unknown before\ndeployment.\n