Balancing Task Allocation in Multi-robot Systems Using adpK-Means Clustering Algorithm

Multi-robot systems are becoming more and more significant in industrial, where allocating tasks for every robot in a reasonable way is a tedious process. The current research mainly focused on reducing the distance between robots and tasks while ignoring the balance of workloads between robots. To address the aforementioned issues, this paper proposes an adaptive K-means clustering algorithm (adpK-means) in order to control a team of robots to accomplish all tasks with a good balance and at a minimal cost. Compared with the K-means clustering algorithm, our proposed algorithm has better performance, where through adaptive dynamic scaling of the clustering space in the iterative process, multiple robots can complete missions with well-distributed workloads. The experimental results show that the algorithm effectively reduces the total energy consumption of the entire robot system and ensures that the tasks of robots are comparative.