Obstacle Detection with Ultrasonic Sensors and Signal Analysis Metrics

GIBBS, Gerard, JIA, Huamin and MADANI, Irfan (2018). Obstacle Detection with Ultrasonic Sensors and Signal Analysis Metrics. Transportation Research Procedia, 28, 173-182. [Article]

Documents
24880:540626
[thumbnail of Madani_ObstacleDetection(VoR).pdf]
Preview
PDF
Madani_ObstacleDetection(VoR).pdf - Published Version
Available under License Creative Commons Attribution Non-commercial No Derivatives.

Download (960kB) | Preview
Abstract
One of the basic tasks for autonomous flight with aerial vehicles (drones) is the detection of obstacles within its flight environment. As the technology develops and becomes more robust, drones will become part of the toolkit to aid maintenance repair and operation (MRO) and ground personnel at airports. Currently laser technology is the primary means for obstacle detection as it provides high resolution and long range. The high intensity laser beam can result in temporary blindness for pilots when the beam targets the windscreen of aircraft on the ground or on final approach within the vicinity of the airport. An alternative is ultrasonic sensor technology, but this suffers from poor angular resolution. In this paper we present a solution using time-of-flight (TOF) data from ultrasonic sensors. This system uses a single commercial 40 kHz combined transmitter/ receiver which returns the distance to the nearest obstacle in its field of view, +/- 30 degrees given the speed of sound in air at ambient temperature. Two sonar receivers located either side of the transmitter / receiver are mounted on a horizontal rotating shaft. Rotation of this shaft allows for separate sonar observations at regular intervals which cover the field of view of the transmitter / receiver. To reduce the sampling frequency an envelope detector is used prior to the analogue-digital-conversion for each of the sonar channels. A scalar Kalman filter for each channel reduces the effects of signal noise by providing real time filtering (Drongelen, 2017a). Four signal metrics are used to determine the location of the obstacle in the sensors field of view: 1. Maximum (Peak) frequency 2. Cross correlation of raw data and PSD 3. Power Spectral Density 4. Energy Spectral Density Results obtained in an actual indoor environment are presented to support the validity of the proposed algorithm.
More Information
Statistics

Downloads

Downloads per month over past year

Metrics

Altmetric Badge

Dimensions Badge

Share
Add to AnyAdd to TwitterAdd to FacebookAdd to LinkedinAdd to PinterestAdd to Email

Actions (login required)

View Item View Item