Vehicle Siren Detection and Acoustic Source Localization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles lack effective methods for detecting and classifying siren signals and localizing their sources, which can lead to increased risks of accidents with emergency vehicles, as they rely primarily on visual and radar sensors, neglecting the importance of sound signals in navigation and collision avoidance.
Innovation Solution
The implementation of microphone arrays that capture sound signals, preprocess them using bandpass filters, and employ deep learning models like convolutional neural networks (CNNs) for acoustic scene classification, combined with triangulation algorithms to estimate the bearing angles and range of siren signal sources, enabling the autonomous vehicle to generate plans and avoid collisions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If autonomous vehicles rely primarily on visual and radar sensors, then the device complexity is reduced, but the reliability of detecting siren signals and localizing emergency vehicles deteriorates
Solution Approach 1:
The patent combines multiple sensor types (microphone arrays for acoustic signals, visual sensors, and radar) into an integrated sensor system. The microphone arrays detect siren signals while other sensors provide complementary information, creating a multi-modal perception system that improves reliability without excessive complexity increase.
Solution Approach 2:
The sensor system is designed to perform multiple functions: detecting siren signals acoustically, localizing emergency vehicles through triangulation, and providing overall environmental perception. This multi-functional approach allows a single integrated system to address various detection needs simultaneously.
2Reliability
If microphone arrays and deep learning models are added to detect and classify siren signals, then the reliability of emergency vehicle detection is improved, but the device complexity increases
Solution Approach 1:
The signal processing system is segmented into distinct functional modules: microphone arrays for signal capture, bandpass filters for frequency selection, deep learning models for classification, and triangulation algorithms for localization. This modular segmentation manages complexity by organizing functions into separate, manageable components.
Solution Approach 2:
Bandpass filters are applied preliminarily to isolate siren signal frequencies before the deep learning model processes the signals. This preliminary frequency filtering reduces the complexity of subsequent classification by pre-processing the input data to focus only on relevant frequency ranges.
3Measurement precision
If triangulation algorithms are used to localize siren signal sources, then the measurement precision of emergency vehicle location is improved, but the loss of time for computation increases
Solution Approach 1:
The system computes bearing angles from multiple microphone arrays to achieve triangulation, using more computational resources than a single-array approach would require. This partial redundancy across multiple arrays improves measurement precision through cross-validation and error reduction.
Solution Approach 2:
The patent replaces complex mechanical localization systems with acoustic signal processing and computational triangulation. By using sound wave propagation characteristics and digital signal processing instead of mechanical positioning systems, the approach achieves high precision while reducing mechanical complexity.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This solution enhances the autonomous vehicle's ability to detect and classify siren signals, accurately localize their sources, and generate routes to avoid emergency vehicles, thereby reducing the risk of collisions and improving safety in emergency scenarios.
Implementation Method 1
The sound signals are pre-processed using a bandpass filter configured with a passband to pass a known range of frequencies of siren signals and to attenuate all other signals
Implementation Method 2
The sound source localizer computes time delay of arrival estimates using a maximum likelihood criterion obtained by implementing a generalized cross correlation method
Implementation Method 3
One or more microphone arrays capture sound signals from an environment in which an autonomous vehicle is operating
Data Source
AI summary
In an embodiment, a method comprises: capturing, by one or more microphone arrays of a vehicle, sound signals in an environment; extracting frequency spectrum features from the sound signals; predicting, using an acoustic scene classifier and the frequency spectrum features, one or more siren signal classifications; converting the one or more siren signal classifications into one or more siren signal event detections; computing time delay of arrival estimates for the one or more detected siren signals; estimating one or more bearing angles to one or more sources of the one or more detected siren signals using the time delay of arrival estimates and a known geometry of the microphone array; and tracking, using a Bayesian filter, the one or more bearing angles. If a siren is detected, actions are performed by the vehicle depending on the location of the emergency vehicle and whether the emergency vehicle is active or inactive.