Research On Intelligent Control Of Vehicle Lights Based On Micro Controller

With the continuous increase in the number of household cars,the narrowing of the field of view during nighttime driving has gradually become a major factor affecting driving safety,leading to frequent nighttime accidents.In order to address this challenge,intelligent light technology is seen as an effective means to improve nighttime driving safety,providing better lighting conditions for drivers in specific road conditions and environments.This study designs an intelligent control system for automotive lighting,ADB(Adaptive Driving Beam),based on a microcontroller.The system can intelligently switch between high and low lights based on external lighting and the distance of the opposing vehicle.At the same time,the system can also adjust the angle of the headlights based on the vehicle’s driving direction and terrain,providing the driver with the best view in various driving environments.The main work completed in this article is as follows:1 This article proposes an intelligent control system based on a microcontroller,which achieves free switching of high and low beam lights when vehicles meet or enter a bend.Firstly,the overall structure of the system is determined to include an environment detection module,a control module,and a driver module;By combining sensors with the improved YOLOv5 algorithm,data is collected and processed from the external environment of the car lights.Finally,a set of program frameworks are designed and corresponding physical verification is conducted to verify the effectiveness of the method proposed in this paper.2 Propose a system design plan for the functions of each module;Using STC89C51 as the core processing unit,select and design peripheral circuits,including control modules,driving environment detection modules,execution modules,etc;Design software for each functional module using programming software such as Keil5.In automatic mode,use sensors and YOLO algorithm to achieve intelligent switching between high and low beams;Send the attitude fusion data transmitted by the sensor to the microcontroller for processing,driving the steering gear to drive the headlight system to follow the rotation and execute corresponding control;Draw a flowchart.3 Construct a YOLOv5 deep learning network to predict the trajectory of target light sources in the external environment of headlights.Divide the target light area and input the ADB car headlight external environment image into the YOLOv5 network.Use convolutional layers to extract image features.After being detected by the YOLOv5 network,the ADB car headlight external environment multiple moving target light sources are grouped into a set.Use the extended Kalman filtering algorithm to predict the trajectory of each element in the set,Improve detection accuracy.Finally,the accuracy of this method was verified through experiments.Under different experimental conditions,the average error between the detection results of the target light source position and the actual results does not exceed 3 * 3 pixels.4 Using Proteus to debug the system hardware,Keil successfully debugs the hardware circuit and software program,and then inputs them into Proteus to debug the entire system function.After repeated testing in the construction of a simulation platform,it is verified that the system proposed in this article is efficient and stable.