Priority-Aware Multi-Robot Coverage Path Planning

TL;DR

This work introduces Priority-Aware MCPP (PA-MCPP), a lexicographic-track multi-robot coverage formulation that first minimizes the priority-weighted latency to cover high-importance zones and then minimizes the overall makespan. The authors propose a two-phase framework: phase one uses greedy zone assignment and zone-wise spanning-tree planning to rapidly cover prioritized areas, while phase two employs Steiner-tree–guided residual coverage with workload balancing to ensure complete area coverage. Empirical results show substantial reductions in priority-weighted latency (around 62.5% on average) with only modest makespan overhead (about 9.7%), and sensitivity analyses demonstrate scalable performance with more robots and tunable zone behavior via weights. The approach integrates classical STC/MCPP techniques with a priority-aware objective, offering a practical pathway for time-critical, large-scale coverage tasks in heterogeneous environments.

Abstract

Multi-robot systems are widely used for coverage tasks that require efficient coordination across large environments. In Multi-Robot Coverage Path Planning (MCPP), the objective is typically to minimize the makespan by generating non-overlapping paths for full-area coverage. However, most existing methods assume uniform importance across regions, limiting their effectiveness in scenarios where some zones require faster attention. We introduce the Priority-Aware MCPP (PA-MCPP) problem, where a subset of the environment is designated as prioritized zones with associated weights. The goal is to minimize, in lexicographic order, the total priority-weighted latency of zone coverage and the overall makespan. To address this, we propose a scalable two-phase framework combining (1) greedy zone assignment with local search, spanning-tree-based path planning, and (2) Steiner-tree-guided residual coverage. Experiments across diverse scenarios demonstrate that our method significantly reduces priority-weighted latency compared to standard MCPP baselines, while maintaining competitive makespan. Sensitivity analyses further show that the method scales well with the number of robots and that zone coverage behavior can be effectively controlled by adjusting priority weights.

Figures (10)

• Figure 1: Comparison of MCPP and Priority-Aware MCPP. While standard MCPP minimizes overall makespan by distributing coverage among robots, Priority-Aware MCPP additionally accounts for spatial priority by encouraging early completion of high-priority zones. Prioritized zones are indicated by green, yellow, and purple shaded areas in the right figure.
• Figure 2: Overview of the single-robot path planning strategy. Red and blue shaded areas indicate the first and second assigned zones $(Z_1,Z_2)$, respectively, along with their corresponding spanning trees $(T_1,T_2)$.
• Figure 3: Overview of the PA-MCPP algorithm. The instance consists of maps with prioritized zones and associated weights. Stage 1 assigns prioritized zones to robots using a greedy allocation followed by local search to refine assignments, then plans traversal sequences within each zone. Stage 2 covers remaining areas to ensure complete coverage, balancing workloads based on previous assignments to minimize makespan.
• Figure 4: PA-MCPP map instances. Black, white, and colored squares represent obstacles, normal terrain, and prioritized zones, respectively. Different colors indicate different zones.
• Figure 5: Visualization of zone assignments and robot coverage paths. Left: Matched priority zones with corresponding spanning trees and assigned paths. Right: Full coverage paths for Robot 1 (red), Robot 3 (green), and Robot 15 (blue), showing prioritized zone coverage followed by residual area coverage.