Obstacle detection system based on low quality factor ultrasonic transducers for medical devices

Abstract

The development of innovative ultrasonic devices can open up new perspectives for navigation in unstructured environments. Recent literature reports much investigation carried out with the aim of improving the performance of the navigational capabilities of robots and the development of medical aids for people with disabilities. The aim of the present work is focused on the development of a sonar-based aid for multiple obstacle detection. The device is a low-cost piezopolymer-based transducer that allows the reception/transmission of broadband signals. Three time-of-flight estimation approaches are herein compared using both real and simulated data. Specifically, the performance of the system in the case of two reflecting surfaces is evaluated in terms of both relative and absolute target distance measurement. The proposed approaches are based on signal envelope, cross-correlation and the least mean square adaptive filtering algorithm, respectively. Preliminary results show that although the envelope-based approach offers better performance in the case of a single target, the adaptive filter-based approach shows interesting properties in more complex scenarios.