Improving ultrasound-based indoor localization systems for quality assurance in manual assembly

Ultrasound-based indoor localization systems can be used for quality control in manual assembly processes by localizing objects such as tools or human hands. For this purpose, distances between objects attached to transmitters and room fixed receivers are measured by time-of-flight measurements followed by multilateration to receive 3D positions. The time-of-flight values themselves are calculated in the receiver by circular correlation with model-based signal replicas. Many prototypes of these systems have been published in research. However, when implementing an economically scalable indoor positioning system based on off-the-shelf standard hardware components, some issues arise, out of which three major are focused on in this publication to optimize the localization performance: Firstly, an adaption circuit is presented to increase the bandwidth of the ultrasound transmission characteristic. This maintains the distance measurement accuracy high and keeps the variance low. Secondly, spectral leakage effects caused by the analog to digital converter's sampling frequency are mitigated with two approaches to decrease the measurement variance. Thirdly, an efficient circular correlation method is presented that allows the usage of field-programmable gate arrays by adapting the signals' lengths. Distance measurement experiments are presented to highlight the performance of these methods.