Shape detection of object behind thin medium using ultrasonic sensors. (26th March 2018)

• Record Type: Journal Article
• Authors: Win, Yin Thu; Afzulpurkar, Nitin; Punyasai, Chumnarn
• Is Part Of: International journal of advanced robotic systems. Volume 15:Number 2(2018:Mar./Apr.)
• Publication Date: 2018-03-26
• Subjects: Ultrasonic sensors -- object shape detection -- surveillance autonomous ground vehicle -- mathematical model -- numerical simulation; Robotics -- Periodicals; Robotics; Periodicals; 629.892
• DOI: 10.1177/1729881418764638
• Languages: English
• ISSNs: 1729-8806

Abstract

This research aims to develop a novel method for recognizing the shape of an object located behind thin medium utilizing combination of ultrasonic sensor signal and edge detection technique. This system is extensively designed so that it can be easily mounted on a surveillance autonomous ground vehicle and is robust for operating in a hazardous environment. Once the autonomous ground vehicle reaches near the hidden objects, the proposed ultrasonic sensor system will identify shapes of the objects. Generally, the number of different sensors mounted on the robot and the number of tasks related to navigation, exploration and detection operation make the design of the autonomous ground vehicle control system very complex. The proposed ultrasonic sensing system for an autonomous ground vehicle provides two-dimensional shape information with low computational cost and will not add significant computational burden to the autonomous ground vehicle's overall control system. At first, a mathematical model for the ultrasonic sensor is created to find the relationships between the attenuation coefficient, the transmission coefficient, the thickness of the medium and ultrasonic wavelength. The mathematical model is followed by numerical simulation to establish several design parameters. Finally, a prototype for the proposed shape detection system is developed using the kHz range off-the-shelf ultrasonic sensors and edge detection technique. Experimental results confirmed that the accuracy of the proposed mathematical model and shape of an object located behind a thin medium could be estimated successfully. It was concluded from the results that the developed novel method was useful for identifying objects located behind thin media and this will help in conducting the further development of research for surveillance and rescue operation using the autonomous ground vehicle.