Social-welfare based task allocation for multi-robot systems with resource constraints

This paper aims to propose a distributed task allocation algorithm for a team of robots that have constraints on energy resources and operate in an unknown dynamic environment. The objective of the allocation is to maximize task completion ratio while minimizing resource usage. The approach we propose is inspired by the social welfare in economics that helps extend the combined operational lifetime of the team by balancing resource consumptions among robots. This social welfare based task allocation method positions a robot team appropriately in preparedness for dynamic future events and enables to achieve the objectives of the system flexibly depending on the application context. Our simulation-based experiments show that the proposed algorithm outperforms a typical market-based approach in various scenarios.

► We present a multi-robot task allocation problem with resource constraints. ► We argue that the preparedness can be realized by conserving resources in a balanced way. ► We define a metric “resource welfare” based on the notion of social welfare in social sciences. ► We propose the resource welfare based task allocation framework.