Researches On The Algorithm Of Adaptive Driving Beam Based On Rapid Control Prototype

Automobile safety is the most important and eternal proposition of automobile development.With the rapid development of driverless cars and intelligent lighting,the performance requirements of headlights are becoming more and more stringent.After all,in an ideal self-driving world,whether drivers drive at night or pedestrians walk safely,there is no need to worry about accidents.The Adaptive Driving Beam as a kind of intelligent high Beam system,through the camera input signal to the car in front of the position and distance,the corresponding adjustment area,the illuminate of lamplight dim or closed across the traffic lights illuminate the area,to avoid glare to the car,for at the same time as much as possible to meet the demand of drivers' vision,will become the future automotive lighting security hot spots.This paper mainly studies the following aspects:(1)The Configuration Analysis of ADB System.This paper mainly introduces the closed-loop system of the whole adaptive hyperlight system,including three parts: sensor,controller and actuator.Then it mainly describes the mechanical ADB headlamp of the execution structure studied in this paper,disassembling it,showing the characteristics of each component in detail,and expounding the working principle of the headlamp,including the left and right upper and lower adjustment motion of the bottom stepper motor,and the movement of the top light source and light baffle driven by the stepper motor.Then,the relevant optical concepts of automotive lighting and the laws and regulations of national standards on automotive lamps are introduced,and the international legal and regulatory requirements for adaptive far-light are introduced,and the functional modes that ADB can realize include basic far-light and adaptive far-light.Finally,several typical light patterns,light intensity distribution and mode of the headlamp are analyzed by Lucidshape.(2)The Building of RCP platform for Mechanical ADB Headlamps.The main task is to find the appropriate LED drive and stepping motor drive for the existing mechanical ADB headlamp actuator,so as to control it through the dSPACE tool chain in the laboratory and MATLAB.Based on the analysis of the LED configuration of the headlamp and related optical parameters,the design of LED is studied,and the characteristics of the stepping motor are found.Then,based on dSPACE,the rapid control prototype platform of ADB system was built and HIL and DIL test schemes for ADB system were proposed.Finally,a simple scenario based on CarMaker modeling and optical simulation,for subsequent bench test or real vehicle test.(3)The Control Algorithm Research of ADB System.This paper mainly studies the parameters related to the stepping motor of the mechanical ADB headlamp actuator,including the left and right swing of the lighting main module of the stepper motor with ball screw at the bottom and the control of the drum or baffle of the stepper motor at the top with two-step deceleration mechanism.Then the paper studies the two kinds of test scenarios which are required by the law.Then we discuss the closed-loop control of headlights which is composed of sensing information such as sample learning and image recognition based on machine learning support vector regression,which lays the foundation for the following production-level complete system.(4)The Data Analysis of Bench Test.It mainly involves bench test.Firstly,the relationship between 3d projection and 2d plane projection of actual light is described,and then it can be quantified and compared effectively.Then we analyzed in detail the requirements of ADB test regulations,including vehicles coming from the opposite side and traveling in the same direction,and set corresponding distances.Based on our mechanical ADB headlamp with 5 LED light sources,the corresponding five regions are divided.Finally,three sets of data are obtained through measurement and calculation.According to the corresponding requirements of each regulation,the actual value is higher than the specified value,but the difference is not very big,and it is acceptable within the allowable range.