EKF-Based 6D SLAM for Air-Duct Cleaning Robot Using Inertial Sensor and Stereo Vision

The air-duct cleaning robot (ADCR) is an effective and safe tool for accomplishing terrible cleaning work in an air-duct network. In this paper, a six-dimensional simultaneous localizing and mapping algorithm based on multi-sensor information fusion is proposed to improve the ADCR's autonomous ability in unknown duct environment. By combing an inertial measurement unit (IMU) with a stereo camera and fusing the measurements of sensors with the extended Kalman filter, the uncertainty of measurement caused by noises can be bounded in a low level. Furthermore, the performance of the proposed scheme was proved to be accurate and robust with experiments in an experimental ventilation-duct environment.