The development of the vehicle sound source localization system

Jang, Yoonho; Kim, Jae Kwang; Kim, Jin-Hak

doi:10.1109/apsipa.2015.7415472

Cited by 12 publications

(18 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, the output of Prony HD and Steiglitz-McBride HD is denoted by 5 coefficients as a 1, a 2 , b 0 , b 1 , and b 2 according to the ARMA model shown in Equation (10). Also, the output of Yule-Walker HD is indicated by two coefficients as a 1 and a 2 along with the AR model demonstrated in Equation (8). The given length of output will be served as the featured information for complete SSL system based on the machine learning (or deep learning) in future works.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Parametric Estimations Based on Homomorphic Deconvolution for Time of Flight in Sound Source Localization System

Park

Choi

Kim

2020

Sensors

View full text Add to dashboard Cite

Vehicle-mounted sound source localization systems provide comprehensive information to improve driving conditions by monitoring the surroundings. The three-dimensional structure of vehicles hinders the omnidirectional sound localization system because of the long and uneven propagation. In the received signal, the flight times between microphones delivers the essential information to locate the sound source. This paper proposes a novel method to design a sound localization system based on the single analog microphone network. This article involves the flight time estimation for two microphones with non-parametric homomorphic deconvolution. The parametric methods are also suggested with Yule-walker, Prony, and Steiglitz-McBride algorithm to derive the coefficient values of the propagation model for flight time estimation. The non-parametric and Steiglitz-McBride method demonstrated significantly low bias and variance for 20 or higher ensemble average length. The Yule-walker and Prony algorithms showed gradually improved statistical performance for increased ensemble average length. Hence, the non-parametric and parametric homomorphic deconvolution well represent the flight time information. The derived non-parametric and parametric output with distinct length will serve as the featured information for a complete localization system based on machine learning or deep learning in future works.Sensors 2020, 20, 925 2 of 32 deliver the situation over the non-line-of-sight (NLOS) locations. Presently, the human and vehicle are cooperated to drive the transport sorely based on the vision information. Hence, indirect imminent endangerment cannot be realized until the human has the visual contact. For example, a car with emergency braking cannot be perceived by the indirect position observers. The squeal sound provides the situation. However, the system cannot recognize the direction to activate the pre-emptive safety devices which reduce or remove the impact of the secondary collisions. The sound information can be used for improving the safety of future autonomous transport system.The driver barely obtains the acoustic information since the vehicle structure debilitates the propagation by airtight cabin. The acoustic perception on sound source and arrival direction are both required to understand the situation. The SSL system mounted on a vehicle could provide the comprehensive information to improve the driving conditions by monitoring the surroundings including the NLOS observation. The conventional SSL approaches recently employed for transport are as follows. The moving vehicle presences are identified by sound localization based on the arrival time difference between the microphones [8]. A sensing technique to localize an approaching vehicle is proposed by an acoustic cue from the spatial-temporal gradient method [9]. For the sequential movement events of vehicles, robust direction-of-arrival estimation is realized by the incoherent signal-subspace method based on a small microphone array [10].The 3D structure of ve...

show abstract

Section: Resultsmentioning

confidence: 99%

“…The conventional SSL approaches recently employed for transport are as follows. The moving vehicle presences are identified by sound localization based on the arrival time difference between the microphones [8]. A sensing technique to localize an approaching vehicle is proposed by an acoustic cue from the spatial-temporal gradient method [9].…”

mentioning

confidence: 99%

Parametric Estimations Based on Homomorphic Deconvolution for Time of Flight in Sound Source Localization System

Park

Choi

Kim

2020

Sensors

View full text Add to dashboard Cite

show abstract

“…[6]. While mobile robotic platforms in outdoor environments may suffer from vibrations and wind, various works have demonstrated detection and localization of salient sounds on moving drones [7] and wheeled platforms [8], [9].…”

Section: Related Workmentioning

confidence: 99%

Hearing What You Cannot See: Acoustic Vehicle Detection Around Corners

Schulz¹,

Mattar²,

Hehn³

et al. 2021

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

This work proposes to use passive acoustic perception as an additional sensing modality for intelligent vehicles. We demonstrate that approaching vehicles behind blind corners can be detected by sound before such vehicles enter in line-ofsight. We have equipped a research vehicle with a roof-mounted microphone array, and show on data collected with this sensor setup that wall reflections provide information on the presence and direction of occluded approaching vehicles. A novel method is presented to classify if and from what direction a vehicle is approaching before it is visible, using as input Direction-of-Arrival features that can be efficiently computed from the streaming microphone array data. Since the local geometry around the ego-vehicle affects the perceived patterns, we systematically study several environment types, and investigate generalization across these environments. With a static ego-vehicle, an accuracy of 0.92 is achieved on the hidden vehicle classification task. Compared to a state-of-the-art visual detector, Faster R-CNN, our pipeline achieves the same accuracy more than one second ahead, providing crucial reaction time for the situations we study. While the ego-vehicle is driving, we demonstrate positive results on acoustic detection, still achieving an accuracy of 0.84 within one environment type. We further study failure cases across environments to identify future research directions.

show abstract

“…Fortunately, microphone arrays bring new potential solutions to deal with these problems. Microphone arrays are widely used in civilian and military fields, such as noise reduction and signal enhancement [ 1 , 2 , 3 , 4 ], hearing aid [ 5 , 6 ], outdoor wind noise measurement [ 7 , 8 ], acoustic source localization and tracking [ 9 , 10 , 11 , 12 ], weapon classification and shooter localization [ 13 , 14 ], and so on. In the last few years, due to the development of the micro-electro-mechanical system (MEMS) technology, small aperture arrays have shown their potential superiority in terms of portability and easy deployment, especially in the field.…”

Section: Introductionmentioning

confidence: 99%

A Robust Real Time Direction-of-Arrival Estimation Method for Sequential Movement Events of Vehicles

Liu

Zhou

et al. 2018

Sensors

View full text Add to dashboard Cite

Parameters estimation of sequential movement events of vehicles is facing the challenges of noise interferences and the demands of portable implementation. In this paper, we propose a robust direction-of-arrival (DOA) estimation method for the sequential movement events of vehicles based on a small Micro-Electro-Mechanical System (MEMS) microphone array system. Inspired by the incoherent signal-subspace method (ISM), the method that is proposed in this work employs multiple sub-bands, which are selected from the wideband signals with high magnitude-squared coherence to track moving vehicles in the presence of wind noise. The field test results demonstrate that the proposed method has a better performance in emulating the DOA of a moving vehicle even in the case of severe wind interference than the narrowband multiple signal classification (MUSIC) method, the sub-band DOA estimation method, and the classical two-sided correlation transformation (TCT) method.

show abstract

The development of the vehicle sound source localization system

Cited by 12 publications

References 7 publications

Parametric Estimations Based on Homomorphic Deconvolution for Time of Flight in Sound Source Localization System

Parametric Estimations Based on Homomorphic Deconvolution for Time of Flight in Sound Source Localization System

Hearing What You Cannot See: Acoustic Vehicle Detection Around Corners

A Robust Real Time Direction-of-Arrival Estimation Method for Sequential Movement Events of Vehicles

Contact Info

Product

Resources

About