A Compressed Sensing Approach for Multiple Obstacle Localisation Using Sonar Sensors in Air.

Eduardo Tondin Ferreira Dias,Hugo Vieira Neto,Fábio Kurt Schneider

doi:10.3390/s20195511

Eduardo Tondin Ferreira Dias, Hugo Vieira Neto + Show 1 more

Open Access

https://doi.org/10.3390/s20195511

Copy DOI

Abstract

Methods for autonomous navigation systems using sonars in air traditionally use the time-of-flight technique for obstacle detection and environment mapping. However, this technique suffers from constructive and destructive interference of ultrasonic reflections from multiple obstacles in the environment, requiring several acquisitions for proper mapping. This paper presents a novel approach for obstacle detection and localisation using inverse problems and compressed sensing concepts. Experiments were conducted with multiple obstacles present in a controlled environment using a hardware platform with four transducers, which was specially designed for sending, receiving and acquiring raw ultrasonic signals. A comparison between the performance of compressed sensing using Orthogonal Matching Pursuit and two traditional image reconstruction methods was conducted. The reconstructed 2D images representing the cross-section of the sensed environment were quantitatively assessed, showing promising results for robotic mapping tasks using compressed sensing.

Highlights

Navigation systems such as autonomous or remotely operated vehicles in general, including mobile robots, traditionally use lidar, radar or sonar sensors for object localisation and environment mapping
The level of novelty chosen for the stop criterion was 0.03 and 0.1 for Orthogonal Matching Pursuit (OMP)-R and OMP-C algorithms, respectively
The regularisation parameter used by the TIKH-R algorithm did not improve performance, with results close to ones obtained for PINV-R, and despite the suppressed artefacts by OMP-R, the algorithm still fails in most reconstructions

Summary

Introduction

Navigation systems such as autonomous or remotely operated vehicles in general, including mobile robots, traditionally use lidar, radar or sonar sensors for object localisation and environment mapping. The conventional detection method in sonar-based autonomous navigation uses the time-of-flight (TOF) concept, which consists of measuring the travel time between the signal burst sent to the environment and the detection of its reflections (echoes) from the existing obstacles [1,6]. This technique, when performed with a single ultrasonic sensor, results in small angular coverage due to the typical main lobe sensitivity characteristics of ultrasound transducers, as well as undesired artefacts due to side lobe characteristics. Adding more sonars to the detection system causes difficulties in discriminating true obstacle reflections due to crosstalk, multiple reflections in the environment, and constructive/destructive interference [3,4]

Results

Discussion

Conclusion