Path Planning for Robots in Fire Scenarios based on Dijkstra Algorithm and Genetic Algorithm

Abstract

Robots have emerged as a pivotal element in enhancing firefighting operations, offering a blend of efficiency and safety to human responders. This paper delves into the development of a path planning strategy for robots navigating through fire scenarios. Particularly we fused the Dijkstra Algorithm and Genetic Algorithm (GA). Our methodology commences with a simplified yet comprehensive definition of the fire environment, incorporating factors such as obstacle height, surface roughness, and fire sources. The environment and surroundings are represented by a 2.5-dimensional grid map. On the other hand, the robot’s traversal and ascent capabilities modeled to reflect varying velocities across different terrains. The Dijkstra Algorithm is subsequently utilized to identify the optimal path from the starting point to the destination, ensuring a balance between minimal traversal time and reduced thermal exposure. Our results, demonstrated through MATLAB simulations, reveal a marked improvement in path planning when GA optimization is applied. The comparative analysis across three scenarios underscores the versatility and effectiveness of our approach, showcasing a significant reduction in both traversal time and thermal exposure. Index Terms—Robots, Firefighting, Path planning, Dijkstra, GA