Efficient Multi-Robot Coverage of an Unknown Environment

Abstract

This paper discusses the complete area coverage of multi-robots in unknown environments. Finding the optimal solution for the problem of area coverage with multiple robots has been proven to be NP-hard. This paper proposes the Multi-agent Node Backtracking Area Coverage (MANBAC) algorithm, which uses improved BCD local area coverage method, and the node backtracking method that is based on the A* algorithm to achieve complete coverage of the entire unknown area. Compared with the area coverage of a single robot, use multi robots can greatly improve the efficiency of area coverage. Compared with the existing coverage algorithms in known area, this study can be applied to a wider range of area scenarios which are not known.