DSmT-Based Ultrasonic Detection Model for Estimating Indoor Environment Contour

Abstract

As for uncertainties of ranging and direction angle in the ultrasonic sensor measurement, an ultrasonic distance measurement model is first proposed for representing these uncertainties through analyzing the working principle of the ultrasonic sensor, which can be adopted to detect the contour of wall plan and cylinder. Moreover, the Dezert–Smarandache Theory (DSmT) method is employed to fuse the uncertainty data measured by using the ultrasonic sensor. Next, extended Hough transform (EHT) and least-squares method (LSM) are combined to identify the environmental contour. Then, the measurement uncertainty of the ultrasonic sensor is analyzed for setting the threshold Th used for distinguishing line and cylinder, and the detection range of the cylinder radius is estimated. Finally, we build an indoor environment and design the ultrasonic sensor hardware system to detect the indoor environment for experimental verification. The indoor environment contour obtained in the experimental results is consistent with the real environment, which illustrates the feasibility and effectiveness of the proposed method. The proposed method has certain reference value for research of simultaneous localization and mapping (SLAM) of the mobile robot.