Multi-robot Task Allocation and Rescue for Mowing

Abstract

Multi-mowing robots have an advantage over a single robot for a large mowing task. The task allocation of the multi-mowing robots' system needs to solve several problems, such as minimizing mowing time, minimizing duplicate areas, etc. In particular, when one or more mowing robots break down in the multi-mowing robots' system, the uncompleted tasks need to be allocated again. There are two algorithms to accomplish the multi-robot mowing in this paper. One is the multi-mowing robots task allocation(MMRTA) algorithm based on the modified genetic algorithm, the other is the filling rescue algorithm. The task allocation was simulated with 4–32 robots to mow the grass field divided into 64 task blocks. The simulation result shows that the tasks and paths of each mowing robot are approximately equal, and proves that the task allocation algorithm proposed in this paper is feasible and effective. The filling rescue algorithm is simulated with 4 or 8 robots that surround the breakdown robot to complete the unfinished tasks of the breakdown robot. The simulation result testifies that the remaining tasks of the breakdown robot are reasonably and efficiently allocated.