Traditional residential firefighting methods have problems such as high labor intensity and harsh operating environment, which are difficult to meet the high standard requirements of modern residential firefighting system construction. The application of robotics in the field of firefighting provides a new way to solve these problems. Firefighting robots are equipped with functions such as autonomous navigation, fire source identification and fire extinguishing, which can perform firefighting tasks in dangerous environments and effectively reduce the safety risks of firefighters. In this paper, we propose a design scheme for a residential firefighting system based on robotics, which adopts SLAM technology that fuses LiDAR and RGB-D cameras to realize robot localization and mapping, and combines thermal imaging technology and artificial intelligence algorithms to complete fire source identification and autonomous fire extinguishing. In terms of methodology, the system fuses LiDAR and RGB-D camera to construct the environment map, applies rtabmap algorithm for data fusion processing, adopts the maximum interclass variance method to identify the fire source, and realizes the precise location of the fire source through the alignment of the point cloud and thermal imaging image. The experimental results show that the obstacle recognition rate of the fused SLAM composition technology reaches 95.9%, the absolute position error is reduced by 51.6%, the fire source recognition accuracy reaches 100%, and the maximum error of fire source orientation is 3°. The system was able to effectively recognize the fire source within a distance of 35 meters, and the recognized temperature range was between 262 and 349°C. The conclusion shows that the residential fire fighting system based on robotics technology performs well in terms of orientation accuracy, fire source recognition and autonomous fire fighting, and provides an effective technical solution for the construction of intelligent fire fighting system, which has good practical value and application prospects.
