The article presents a system that allows dynamic task allocation within heterogeneous robot teams. The system leverages Behavior Trees (BTs) to specify missions, which can then be dynamically allocated to the robots in the team. The system also models an individual robot’s capabilities using the rosbtpyBT framework and offers an auction system to allocate tasks dynamically to the most capable robot. The system aims to improve mission execution quality and prevent faulty assignments. It was tested on a simulated find-and-decontaminate mission with three robots, showing improved team effectiveness and flexibility in team composition.
Publication date: 6 Feb 2024
Project Page: Not provided
Paper: https://arxiv.org/pdf/2402.02833
