Ultrasound based Localization and Tracking

Summarise with VIRAT
Authors: Azra Talic
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Object localization and tracking in today's world is everywhere around us. The majority of studies in this domain focus on computer vision, enabling systems to identify and follow objects in motion using sequential image frames, particularly in applications such as video surveillance and human activity recognition. Beyond classical computer vision applications, object localization and tracking is greatly important for autonomous systems, namely self-driving cars, robotic systems, and drones. In this context, applications are based on different sensor systems and their integration to obtain more accurate and reliable performance. Different types of sensors, for example cameras, LiDAR, RADAR, GPS, and ultrasonic sensors, are based on different physical concepts; for instance, ultrasonic sensors operate with high frequency sound waves, while RADAR and LiDAR systems rely on radio frequency waves and light, respectively. Ultrasonic sensors offer advantages in cost efficiency and ease of signal processing, albeit at the expense of reduced precision and range. This work explores the utilization of ultrasonic sensing concepts using a newly developed transceiver platform, termed the Sensedu Shield. This platform leverages multiple Micro-Electro-Mechanical System (MEMS) microphones having resonant frequencies within the ultrasonic range. The system is designed for Arduino GIGA development board, providing a compact form factor. Distance measurements are derived from Time-of-Flight (ToF) measurements of ultrasonic waves traveling between a transmitter (the speaker), an object, and receivers (microphones). These measurements are then utilized in the Extended Kalman Filter (EKF) algorithm to accurately reconstruct the 3D trajectory of an object in close proximity to the system. The approach is demonstrated through a combination of simulations and experimental validations, showing potential for the cost-effective, short-range object localization and tracking in practical scenarios.

Fields of Study & Subjects

Indoor and Outdoor Localization Technologies Target Tracking and Data Fusion in Sensor Networks Arduino and IoT Applications

Publication details

Journal / source
University of Klagenfurt
Publication date
1 January 2026
Language
EN
Document type
Thesis