Unlock instant, AI-driven research and patent intelligence for your innovation.

A system and method for automatically identifying vehicle accidents

A recognition method and automatic recognition technology, which is applied in the field of automatic recognition of vehicle accident systems, achieves the effects of low cost, good scalability, and simple technical realization

Active Publication Date: 2022-04-15
湖南湘江智能科技创新中心有限公司
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to provide a system and method for automatically identifying vehicle accidents, which can automatically and intelligently identify accidents and report them automatically, and intercept the environment and vehicle state information within a period of time before and after the accident, and automatically store them as event information, which solves the existing problems. It is necessary to manually press the button to report the problem

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A system and method for automatically identifying vehicle accidents
  • A system and method for automatically identifying vehicle accidents
  • A system and method for automatically identifying vehicle accidents

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0040] Such as figure 1 A recognition method for automatically recognizing vehicle accidents shown, the recognition method includes the following steps:

[0041] S1: The control unit reads the vehicle data acquisition unit to collect vehicle data information such as brake throttle size, gear position, steering angle of the steering wheel, current vehicle speed, and vehicle light status;

[0042] S2: the control unit calculates the current running state of the vehicle based on the vehicle data information;

[0043] S3: the control unit reads the attitude estimation data collected by the measurement unit;

[0044] S4: The control unit judges whether an accident occurs through deep learning based on the current operating state data, attitude estimation data and pressure sensor data of the vehicle. If the accident does not occur, return to continue step S1, and if the accident occurs, enter step S5;

[0045] S5: Pack the running state data of the vehicle when the accident occurr...

Embodiment 2

[0052] Based on the recognition method for automatically identifying vehicle accidents in Embodiment 1, the vehicle operating state data uploaded in step S5 is combined with the actual accident of the vehicle to obtain the vehicle accident state and operating state model, and the vehicle accident state and operating state model have a significant impact on vehicle accidents. The judgment model is corrected and compensated, and the specific steps of correcting and compensating the vehicle accident judgment model are as follows:

[0053] A. Design the neural network structure and set the network structure parameters, including input layer-convolution layer-pooling layer-convolution layer-pooling layer-full connection layer-a network structure of 6 layers in total;

[0054] B. Input the two-dimensional spatial spectrogram into the convolutional neural network after setting the parameters to obtain the predicted category label, and compare it with the real category label of the sam...

Embodiment 3

[0058] A sort of figure 2 A recognition system for automatically identifying vehicle accidents is shown, the system includes a control unit, a pressure sensor, an inertial measurement unit, a vehicle data acquisition unit, and a data gateway; the vehicle data acquisition unit collects the size of the brake accelerator, gear position , the steering angle of the steering wheel, the current speed of the vehicle, the status of the vehicle lights and other vehicle data information; the inertial measurement unit is used to collect vehicle inertia data; Pressure sensors are installed on the bumper and a circle on the side of the car, surrounding the body. The pressure sensor is a sensor that converts the pressure into an electrical signal. When the position is changed by the pressure, the electrical signal of the corresponding part will change, so as to calculate the pressure value information; the control unit reads through the CAN bus system The current acceleration variation of ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a system and method for automatically identifying vehicle accidents, belonging to the field of unmanned intelligent transportation. It includes a control unit, a pressure sensor, an inertial measurement unit, a vehicle-machine data acquisition unit, and a data gateway; the vehicle-machine data collection unit collects vehicle data information; the inertial measurement unit is used to collect vehicle inertial data; the control unit passes The CAN bus system reads the acceleration variation of the current inertial measurement unit, the value of the pressure sensor, the state of the vehicle, the gear position and other comprehensive information, and recognizes the vehicle accident judgment model based on deep learning; the data gateway calculates the fused information of the control unit The information judged as the accident model is pushed to the cloud platform; the automatic intelligent identification of accidents and the automatic reporting of accident information and environmental and vehicle status information for a period of time before and after the accident are realized, which solves the existing problem of manual button operation for reporting.

Description

technical field [0001] The invention relates to the field of unmanned intelligent transportation, in particular to a system and method for automatically identifying vehicle accidents. Background technique [0002] With the breakthrough progress of deep learning in recent years, unmanned vehicles have also begun to be commercialized in these years, and with the sharp increase in car ownership in various countries, urban road conditions have become more severe and complicated, and urban traffic is facing unprecedented challenges. of great pressure. Coupled with human factors such as fatigue driving and drunk driving, the traffic accident rate in countries all over the world is increasing year by year, even more than the death toll in world wars. With the efficient integration of automotive technology, information communication technology and intelligent control technology, unmanned vehicles that integrate many technologies such as automatic control, artificial intelligence, a...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G08G1/01G07C5/00G07C5/08G06N3/04H04L67/12
CPCG08G1/0125G08G1/0137G07C5/008G07C5/085H04L67/12G06N3/045
Inventor 张福升刘高付程
Owner 湖南湘江智能科技创新中心有限公司