eduzhai > Applied Sciences > Engineering >

Hearing What You Cannot See Acoustic Vehicle Detection Around Corners

  • king
  • (0) Download
  • 20210506
  • Save

... pages left unread,continue reading

Document pages: 8 pages

Abstract: This work proposes to use passive acoustic perception as an additionalsensing modality for intelligent vehicles. We demonstrate that approachingvehicles behind blind corners can be detected by sound before such vehiclesenter in line-of-sight. We have equipped a research vehicle with a roof-mountedmicrophone array, and show on data collected with this sensor setup that wallreflections provide information on the presence and direction of occludedapproaching vehicles. A novel method is presented to classify if and from whatdirection a vehicle is approaching before it is visible, using as inputDirection-of-Arrival features that can be efficiently computed from thestreaming microphone array data. Since the local geometry around theego-vehicle affects the perceived patterns, we systematically study severalenvironment types, and investigate generalization across these environments.With a static ego-vehicle, an accuracy of 0.92 is achieved on the hiddenvehicle classification task. Compared to a state-of-the-art visual detector,Faster R-CNN, our pipeline achieves the same accuracy more than one secondahead, providing crucial reaction time for the situations we study. While theego-vehicle is driving, we demonstrate positive results on acoustic detection,still achieving an accuracy of 0.84 within one environment type. We furtherstudy failure cases across environments to identify future research directions.

Please select stars to rate!


0 comments Sign in to leave a comment.

    Data loading, please wait...