Acoustic detection method and device in unmanned driving environment

Abstract

The embodiment of the invention provides an acoustic detection method and device in an unmanned driving environment. The acoustic detection method comprises the steps that an acoustic signal of the environment where an unmanned vehicle is located at the current moment is acquired; performing multi-scale space decomposition on the acoustic signal to obtain multiple groups of decomposition coefficients of the acoustic signal in different scale spaces; performing acoustic feature extraction on each group of decomposition coefficients to obtain a plurality of feature vectors with the same group number as the decomposition coefficients; respectively and correspondingly inputting the plurality of feature vectors into a plurality of trained neural network models, and outputting classification results of sound production object recognition in the environment under different scale spaces; and determining whether the sounding object is a preset type of vehicle or not according to the classification results in all scale spaces. According to the method, the unmanned vehicle can identify specific sounding vehicles such as an ambulance and a police car by using acoustic signals, so that corresponding control operations such as avoidance and the like can be carried out.

Description

technical field [0001] The present application relates to the technical field of unmanned driving control, in particular to an acoustic detection method and device in an unmanned driving environment. Background technique [0002] In the unmanned driving environment, the environment is often more complex. The current unmanned vehicles' perception of surrounding objects is generally obtained through millimeter-wave radar, lidar, and cameras. These signals are based on the visual perception system. According to the "Road Traffic Safety Law", if you encounter special vehicles performing tasks such as police cars and ambulances, the driving vehicles need to give way. However, a single visual perception signal cannot enable unmanned vehicles to effectively perceive these special vehicle information. Acoustic signals, as another perception signal, can also provide rich environmental information. Under certain conditions, they can provide more accurate external environmental inform...

AI Technical Summary

Problems solved by technology

However, a single visual perception signal cannot enable unmanned vehicles to effectively perceive these special vehicle information

Images