A Novel Acoustic Localization for Fire Rescue Robots in the NLOS Indoor Corridor

Fire environment rescue faces significant challenges to the effectiveness of current localization technologies for rescue robots, characterized by thick smoke and reduced visibility, necessitate remote robot operation in Non-Line of Sight (NLOS) conditions and hinder pre-deployment of localization infrastructure, complicating efforts. This paper introduces a novel acoustic indoor localization approach tailored for indoor corridor rescue robots operating in NLOS environments. This method employs a Frequency Division Multiple Access (FDMA) system for the localization signal. An advanced echo-ranging algorithm is employed to accurately demonstrate the distance between the rescue robot and the corridor walls. Furthermore, the path length from the initial point around the corridor corner to the robot is determined using acoustic methods. The precise localization of the rescue robot is subsequently achieved by integrating these distance measurements through novel path matching and data fusion techniques. Experiments demonstrate the capability to achieve localization errors as low as around 0.25 meters in NLOS curved corridors. The proposed method offers significant advantages for operation in challenging environments: it facilitates accurate localization of rescue robots in conditions of reduced visibility and NLOS, enables rescue personnel to control the robots from covered positions safely and is cost-effective as well as straightforward to deploy. The data has been uploaded to GitHub and can be accessed at: https://github.com/ChirsJia/CCC.