Research On Design Method Of Acoustic Vehicle Alerting System

Due to the silent nature of electric/hybrid vehicles powertrain,Pedestrians,bicyclists and other road users fail to hear approaching of electric/hybrid vehicles,which is a hidden danger to their safety.With the trend of automobile electrification,the sales of electric vehicles are rising steadily.But the problem of the threat to pedestrian safety,which is caused by low noise level of electric vehicles,is becoming more and more serious.Several survey reports show that when electric and hybrid vehicles are running at low speed,the traffic safety accidents caused by low noise level of them can not be underestimated.In order to tackle this problem,many countries and regions have formulated relevant laws and regulations,forcing electric cars to install an electronic warning system.The relevant mandatory laws and regulations of our country are also about to be issued.So it is necessary to design a set of Acoustic Vehicle Alerting System before the promulgation of the laws and regulations to seize the market in time and to improve the share of the domestic parts of the vehicle.In this study,the design process of AVAS is analyzed in detail.Based on the current EU regulations,the key algorithm of AVAS is comprehensively disassembled and implemented.The specific contents of this study are as follows:1)According to the relevant statistical survey data from abroad,the necessity and feasibility of the design of the external warning sound of the electric vehicle are obtained.Based on the analysis of the technical requirements of foreign related laws and regulations,especially,the EU regulations,the algorithm design requirements and hardware design requirements are obtained,which provides reliable design input for the design of the system.2)In the process of preprocessing,the necessity of sound source processing is discussed.Based on the spectrum waterfall graph,the concept of order is introduced,and then the relationship between the energy intensity of every order and our impression of the engine's sound,and connect the Musical Harmonic Theory with the Active Sound Design is discussed.In order to obtain the target sound,all aspects of the sound preprocessing is introduced in detail,including engine spectrum analysis,order tracking and extraction,order signal filtering and order enhancement or attenuation,and their related algorithms implementation in MATLAB and Simulink3)On the basis of EU regulations,the algorithm of each part of the Acoustic Vehicle Alerting System and the theoretical basis of its application are analyzed in detail.After analyzing the specific algorithm used in every part of the system,we use Simulink to model and simulate the algorithm,and then carry out a rapid prototyping experiment to verify the feasibility of the algorithm.4)The C code of algorithm is generated by the Simulink model,and integrated with the well written driver,launch code and diagnostic program,then download to hardware,finally,we completed debugging of the software.On this basis,the related PC programs are designed to change the algorithm parameters and then the functional integrity and performance of the hardware and software are verified.In this study,the algorithm of AVAS and the implementation of its hardware and software are analyzed in detail.It mainly includes sound preprocessing and target sound design,algorithm analysis,modeling and simulation,hardware design and software design,debugging and calibration of hardware and software and so on.The whole process of the AVAS is demonstrated in detail and clearly.The purpose of this study is to discuss the algorithm principle and general design process of the AVAS,to make up the related blank.There are two core algorithms in this study: one is the Void-Kalman filter algorithm used for the preprocessing of the sound,and the other is the pitch shifting algorithm of the sound.In this study,the two algorithms are derived in detail,and the details of the numerical realization of the algorithm are described.