Indoor Environment RGB-DT Mapping for Security Mobile Robots

Many robot applications, such as environmental monitoring, security and surveillance help people to do tasks in day-to-day scenarios. However, the growing security demand for environment perception is a key issue of mapping or frequent updating in the long term, such as fire detection in early stage. A hybrid mapping method is proposed based on fusing RGB, depth and thermal (DT) information from Kinect and infrared sensors equipped in the mobile robot. Firstly, the proposed pipeline will estimate the robot’s pose by extracting and matching ORB features in RGB images successively. Then Poses corresponding to each depth and thermal- Infrared image are estimated through a combination of timestamp synchronization and the result of the extrinsic calibration of the system, and the map with both appearance and the temperature of environment is generated by the combination of The RGB and temperature information. Finally, the depth information is used to project the pixel points to the world coordinate system to generate the RGB-DT map. Extensive results verify the effectiveness of the proposed RGB-DT mapping for environments perception.