Hybrid Algorithm Based on Bidirectional A* and Floodfill Applied to Coverage Path Planning for AGVs

• Lucas L. M. Carvalho CEFET-RJ
• Diogo G. Marques CEFET-RJ
• Gabriel G.R. Castro CEFET-RJ
• Thiago M. Tuxi CEFET-RJ
• Milena F. Pinto CEFET-RJ

Resumo

This paper presents a hybrid Coverage Path Plan-ning (CPP) strategy that integrates the Floodfill algorithm with bidirectional A * (BA *) to optimize navigation for Autonomous Ground Vehicles (AGVs) tasked with inspecting photovoltaic farms. The proposed approach leverages the global exploration capabilities of Floodfill with the heuristic efficiency of bidirectional A *, enabling fast and complete area coverage while min-imizing redundant traversal and computational overhead. The framework is validated in a Software-In-the-Loop (SITL) simulation using Gazebo and ROS, and its performance is benchmarked against conventional Floodfill, Floodfill with A *, and Wavefront-based methods. Results demonstrate significant improvements in coverage efficiency, path smoothness, and execution time, highlighting the method's potential for real-world large-scale CPP applications. The proposed method achieved overlap and direction change rates below 1 %, significantly improving path efficiency.

Palavras-chave: Photovoltaic systems, Navigation, Conferences, Education, Inspection, Benchmark testing, Path planning, Land vehicles, Computational efficiency, Robots, Autonomous Ground Vehicle, Robotic Path Planning, Hybrid Algorithm, Coverage Path Planning, Photo-voltaic Inspection