Genetic algorithm-based coverage path planning for autonomous aircraft cabin cleaning by reconfigurable robot

Abstract

Designing an optimal Coverage Path Planning (CPP) framework for autonomous aircraft cabin cleaning is a critical challenge due to the time-sensitive nature of aircraft turnaround operations. Conventional domestic cleaning robots struggle to adapt to the confined and irregular cabin layouts of commercial aircraft. To address this, the paper proposes a two-stage CPP approach utilizing the reconfigurable robot. In the first stage, the robot operates in its full-size configuration to efficiently clean open regions such as aisles and galleys, skipping hard-to-access seat rows to minimize total cleaning time. In the second stage, a Genetic Algorithm (GA)-based Traveling Salesman Problem (TSP) optimization process determines the optimal visiting sequence for the skipped areas, while simultaneously accounting for the robot’s reconfiguration energy model. This integrated framework explicitly models the trade-off between coverage efficiency, energy consumption, and reconfiguration cost, ensuring that the robot autonomously selects the most energy-optimal path under operational constraints. The experiments incorporating airline procedures and cabin geometry demonstrate that the proposed approach significantly outperforms conventional CPP strategies in both coverage time and energy usage. The results validate the feasibility of deploying reconfigurable robotic systems for real-world autonomous aircraft cabin cleaning during turnaround operations.