Fire-Fighting Task Simulation System of Heterogeneous Multi-UAVs Based on ROS

Abstract

Aiming at the application of UAVs in fire-fighting, a fire-fighting task simulation system of heterogeneous multi-UAVs is proposed in this paper, which simulates the process of fire-fighting tasks in the city and the forest based on ROS. Firstly, the task architecture is based on the OODA loop (Observe, Orient, Decide, Act). Secondly, the construction of two typical scenarios of city and forest were completed based on Gazebo and Blender and different strategies to perform fire-fighting tasks were adopted in the two scenarios. When performing the task, according to the perception of real-time situation, the fire-fighting strategy which depends on the situation assessment and trajectory planning, is taken. The whole process of simulation was accomplished by heterogeneous multi-UAVs with good visualization effect. It has formed a relatively complete rescue system and solved the key problems such as perception and trajectory planning. The simulation of the above strategy in the system has successfully verified its feasibility.