Detecting Road Surface Wetness from Audio: A Deep Learning Approach
Authors:
Irman Abdić,
Lex Fridman,
Erik Marchi,
Daniel E Brown,
William Angell,
Bryan Reimer,
Björn Schuller
Abstract:
We introduce a recurrent neural network architecture for automated road surface wetness detection from audio of tire-surface interaction. The robustness of our approach is evaluated on 785,826 bins of audio that span an extensive range of vehicle speeds, noises from the environment, road surface types, and pavement conditions including international roughness index (IRI) values from 25 in/mi to 14…
▽ More
We introduce a recurrent neural network architecture for automated road surface wetness detection from audio of tire-surface interaction. The robustness of our approach is evaluated on 785,826 bins of audio that span an extensive range of vehicle speeds, noises from the environment, road surface types, and pavement conditions including international roughness index (IRI) values from 25 in/mi to 1400 in/mi. The training and evaluation of the model are performed on different roads to minimize the impact of environmental and other external factors on the accuracy of the classification. We achieve an unweighted average recall (UAR) of 93.2% across all vehicle speeds including 0 mph. The classifier still works at 0 mph because the discriminating signal is present in the sound of other vehicles driving by.
△ Less
Submitted 4 December, 2015; v1 submitted 22 November, 2015;
originally announced November 2015.
Automated Synchronization of Driving Data Using Vibration and Steering Events
Authors:
Lex Fridman,
Daniel E Brown,
William Angell,
Irman Abdić,
Bryan Reimer,
Hae Young Noh
Abstract:
We propose a method for automated synchronization of vehicle sensors useful for the study of multi-modal driver behavior and for the design of advanced driver assistance systems. Multi-sensor decision fusion relies on synchronized data streams in (1) the offline supervised learning context and (2) the online prediction context. In practice, such data streams are often out of sync due to the absenc…
▽ More
We propose a method for automated synchronization of vehicle sensors useful for the study of multi-modal driver behavior and for the design of advanced driver assistance systems. Multi-sensor decision fusion relies on synchronized data streams in (1) the offline supervised learning context and (2) the online prediction context. In practice, such data streams are often out of sync due to the absence of a real-time clock, use of multiple recording devices, or improper thread scheduling and data buffer management. Cross-correlation of accelerometer, telemetry, audio, and dense optical flow from three video sensors is used to achieve an average synchronization error of 13 milliseconds. The insight underlying the effectiveness of the proposed approach is that the described sensors capture overlap** aspects of vehicle vibrations and vehicle steering allowing the cross-correlation function to serve as a way to compute the delay shift in each sensor. Furthermore, we show the decrease in synchronization error as a function of the duration of the data stream.
△ Less
Submitted 1 March, 2016; v1 submitted 20 October, 2015;
originally announced October 2015.
