Improved Coverage Path Planning for Indoor Robots Based on BIM and Robotic Configurations

This paper proposes an improved CPP system based on building information modeling (BIM) and robotic configurations for indoor robots. Firstly, BIM is semantically enriched for CPP tasks, based on which semantic trapezoidal grid maps are constructed with accurate information. Secondly, a new concept called “Coverage Bonus” is integrated into the coverage pattern analysis, which can be applied in the scenario that robot’s coverage size is different from its body size. Finally, the coverage sequence is optimized by solving Cluster Generalized Traveling Salesman Problem (CGTSP), and the generated paths can observe the travel rules and obtain high efficiency. The CPP system is validated in a representative university building floor and a typical anti-epidemic robot. It is demonstrated that the BIM-based TGM generation method can improve mapping quality compared to Classical methods. In addition, the improved CPP algorithm and one of its ablation studies achieve the best and similar coverage performance (e.g., 97% coverage ratio). However, the improved CPP algorithm outperforms this ablation study in satisfying the coverage rules strictly.