Publications / 2025 Proceedings of the 42nd ISARC, Montreal, Canada
Mobile robots are increasingly used to explore and inspect confined environments, often in situations that are hazardous or inaccessible to humans, highlighting the need for advanced mapping and sensing capabilities. This paper presents a novel approach for constructing a multi-dimensional map by integrating sensor data from a hexapod robot with Simultaneous Localization and Mapping (SLAM). The robot, equipped with a DHT22 sensor and powered by a Raspberry Pi 4 Model B, was deployed in a mechanical room at the University of Alberta to collect humidity and temperature data while simultaneously mapping the environment. The Cartographer SLAM algorithm was used for mapping, and the sensor data was fused with the generated map by the Inertial Measurement Unit (IMU) using a bilinear interpolation algorithm. This pilot experiment serves as a demonstration of the proposed robot system, which results in a multidimensional map that combines 2D geographical map with sensory maps such as humidity and temperature. The map provides a visualization of the spatial distribution of environmental variables within the confined area. This approach has potential applications in various scenarios, including quick mapping of hazardous areas and routine inspections of areas with limited access. Future work will focus on incorporating 3D SLAM and exploring the use of machine learning techniques for automated anomaly detection within the multi-dimensional map, while addressing the current limitations related to real-time processing and visualization.