CN-121989848-A - Vehicle light adjusting method and system, computer equipment and vehicle

CN121989848ACN 121989848 ACN121989848 ACN 121989848ACN-121989848-A

Abstract

The application provides a vehicle light adjusting method and system, computer equipment and a vehicle, wherein the method comprises the steps of obtaining driving parameters of the vehicle, determining at least two matrixes with different dimensions according to the driving parameters, carrying out weighted calculation on thresholds corresponding to all the matrixes to obtain a light adjusting weighted threshold, and carrying out light adjustment on the vehicle under the condition that the light adjusting weighted threshold is greater than or equal to a preset threshold so as to enable the driving environment of the vehicle after light adjustment to be better than that before light adjustment. According to the application, the matrix with different dimensions is formed by driving parameters, future illumination demands and traffic conditions can be predicted from different dimensions, so that the self-adaptive light system of the vehicle can carry out light adjustment relative to the advance part time, the problem of delay of response of the sensor and lag of the control strategy in the self-adaptive light system is solved or alleviated, the flexibility and reliability of the self-adaptive light system are improved, the operation frequency of adjusting light manually by a driver is reduced, and the driving comfort of the driver is improved.

Inventors

LI NAN
WANG JUNLIN
TAN CHUNYAN
YE SONGLIN

Assignees

重庆赛力斯凤凰智创科技有限公司

Dates

Publication Date: 20260508
Application Date: 20260309

Claims (10)

1. A method of adjusting vehicle lighting, the method comprising the steps of: Acquiring driving parameters of a vehicle, wherein the driving parameters comprise a plurality of parameter factors, and the plurality of parameter factors are used for representing at least one of a driving state of the vehicle, an external environment of the vehicle and an external target state of the vehicle; determining at least two matrixes with different dimensions according to the plurality of parameter factors, and carrying out weighted calculation on thresholds corresponding to all the matrixes to obtain light adjustment weighted thresholds; And under the condition that the light adjustment weighting threshold value is larger than or equal to a preset threshold value, light adjustment is carried out on the vehicle, so that the driving environment of the vehicle after light adjustment is better than the driving environment of the vehicle before light adjustment.
2. The vehicle light adjustment method according to claim 1, wherein determining the matrix of two different dimensions from the plurality of parameter factors comprises: Each parameter factor is respectively used as a row index and a column index of one matrix, the weight value of all parameter factors after being combined in pairs is determined based on the weight value of all parameter factors, the weight value of all parameter factors after being combined in pairs is filled into the corresponding index position to form a matrix for representing the data feature fusion dimension and is recorded as a data fusion matrix, and And filling the data value and the weight value corresponding to each parameter factor into the corresponding index position to form a matrix for predicting the dimension of the risk event, and recording the matrix as a risk judgment matrix, wherein the risk event comprises a vehicle collision event.
3. The vehicle light adjustment method according to claim 1, wherein determining the matrix of two different dimensions from the plurality of parameter factors comprises: Each parameter factor is respectively used as a row index and a column index of one matrix, the weight value of all parameter factors after being combined in pairs is determined based on the weight value of all parameter factors, the weight value of all parameter factors after being combined in pairs is filled into the corresponding index position to form a matrix for representing the data feature fusion dimension and is recorded as a data fusion matrix, and And respectively taking each parameter factor as a row index and a column index of another matrix, determining the relative importance degree of all the parameter factors after being combined in pairs based on the importance degree of all the parameter factors, filling the relative importance degree of all the parameter factors after being combined in pairs to corresponding index positions to form a matrix for predicting the dimension of a risk event, and recording the matrix as a risk judgment matrix, wherein the risk event comprises a vehicle collision event, and the importance degree of each parameter factor is obtained based on the weight value of the corresponding parameter factor.
4. A vehicle light adjustment method according to claim 2 or 3, wherein the process of weighting the thresholds corresponding to all the matrices to obtain the light adjustment weighted threshold comprises: Inputting the data fusion matrix into a neural network model established in advance or in real time for pooling and convolution, and determining a matrix with row dimension and column dimension smaller than the data fusion matrix and marking the matrix as a target matrix; and determining a threshold value of the target matrix based on the target matrix and the transposed matrix of the target matrix, and carrying out weighted calculation on the threshold value of the target matrix and the threshold value of the risk judgment matrix to obtain the light adjustment weighted threshold value.
5. The vehicle light regulating method of claim 4, wherein inputting the data fusion matrix into a neural network model established in advance or in real time for pooling and convolution comprises: inputting the data fusion matrix into a first pooling layer of the neural network model for pooling, determining a matrix with row dimension and column dimension smaller than the data fusion matrix, and marking the matrix as a first pooling matrix; Inputting the data fusion matrix and the first pooling matrix into a first convolution layer of the neural network model for convolution, and determining a matrix with the same row dimension and column dimension as the first pooling matrix, and marking the matrix as a first convolution matrix; Inputting the first convolution matrix into a second pooling layer of the neural network model for pooling, determining a matrix with row dimension and column dimension smaller than the first convolution matrix, and marking the matrix as a second pooling matrix; And inputting the first convolution matrix and the second pooling matrix into a second convolution layer of the neural network model for convolution, and determining a matrix with the same row dimension and column dimension as the second pooling matrix as the target matrix.
6. The method according to claim 1, wherein the process of adjusting the light of the vehicle includes adjusting the low beam of the vehicle so that the driving environment of the vehicle after the low beam adjustment is better than the driving environment before the low beam adjustment, and/or adjusting the high beam of the vehicle so that the driving environment of the vehicle after the high beam adjustment is better than the driving environment before the high beam adjustment, wherein the irradiation range of the high beam is larger than the irradiation range of the low beam or the irradiation brightness of the high beam is larger than the irradiation brightness of the low beam.
7. The vehicle light adjustment method according to claim 1, wherein the plurality of parameter factors for characterizing the driving state of the vehicle in the driving parameters include a real-time vehicle speed, a steering wheel angle, a current lane position, and a vehicle body pitch angle, the plurality of parameter factors for characterizing the external environment of the vehicle in the driving parameters include an ambient light intensity, a road type, a road surface gradient, an ambient weather, and a road curvature radius, and the plurality of parameter factors for characterizing the external target state of the vehicle in the driving parameters include a target category, a target relative vehicle distance, a target relative vehicle speed, and a target lateral position.
8. A vehicle light regulation system, the system comprising: A driving parameter module for obtaining driving parameters of the vehicle, the driving parameters including a plurality of parameter factors for characterizing at least one of a driving state of the vehicle, an external environment of the vehicle, and an external target state of the vehicle; The weighting calculation module is used for determining at least two matrixes with different dimensions according to the plurality of parameter factors, and carrying out weighting calculation on the thresholds corresponding to all the matrixes to obtain a light adjustment weighting threshold; and the light adjusting module is used for adjusting the light of the vehicle under the condition that the light adjusting weighted threshold value is larger than or equal to a preset threshold value, so that the driving environment of the vehicle after light adjustment is better than the driving environment before light adjustment.
9. A computer device comprising a memory, a processor and a computer program stored on the memory, the processor executing the computer program to perform the steps of the vehicle light adjustment method of any one of claims 1 to 7.
10. A vehicle comprising a vehicle light regulating system according to claim 8 or comprising a computer device according to claim 9.

Description

Vehicle light adjusting method and system, computer equipment and vehicle Technical Field The present application relates to the field of vehicle technologies, and in particular, to a vehicle light adjustment method and system, a computer device, and a vehicle. Background Most vehicles can be configured with the self-adaptive light system when leaving the factory, but under a complex driving scene, the self-adaptive light system of part of the vehicles mainly depends on fixed sensors and preset rules to run, and the problems of sensor response delay and control strategy hysteresis can occur, so that the accuracy is low when the part of vehicles perform light self-adaptive control, a driver is required to perform manual light adjustment operation in time, and the flexibility and reliability of the self-adaptive light system are reduced. Disclosure of Invention The application provides a vehicle light adjusting method and system, computer equipment and a vehicle, so as to solve or alleviate the problems described above. The application provides a vehicle light adjusting method, which comprises the following steps of obtaining driving parameters of a vehicle, wherein the driving parameters comprise a plurality of parameter factors, the parameter factors are used for representing at least one of a vehicle driving state, a vehicle external environment and a vehicle external target state, determining matrixes with at least two different dimensions according to the parameter factors, carrying out weighted calculation on thresholds corresponding to all the matrixes to obtain a light adjusting weighted threshold, and carrying out light adjustment on the vehicle under the condition that the light adjusting weighted threshold is larger than or equal to a preset threshold so that the driving environment of the vehicle after light adjustment is better than that before light adjustment. Compared with the threshold value obtained by relying on a fixed sensor and a preset rule, the light adjusting weighted threshold value obtained by weighting calculation has the advantages that different dimensions are considered, so that the accuracy is higher, and more complex driving scenes can be adapted when the light is adjusted. Meanwhile, the matrix with different dimensions is formed through driving parameters, future illumination demands and traffic conditions can be predicted from different dimensions, the self-adaptive light system of the vehicle can carry out light adjustment in advance of part of time relatively, the problem that the response of the sensor in the self-adaptive light system is delayed and the control strategy is lagged is solved or relieved, the flexibility and the reliability of the self-adaptive light system are improved, the operation frequency of adjusting light manually by a driver is reduced, and the driving comfort of the driver is improved. In a first possible implementation manner of the first aspect, the process of determining two matrices with different dimensions according to the plurality of parameter factors includes taking each parameter factor as a row index and a column index of one matrix respectively, determining weight values of all parameter factors after combining two by two based on the weight values of all parameter factors, filling the weight values of all parameter factors after combining two by two to corresponding index positions to form a matrix for representing a data feature fusion dimension, and marking the matrix as a data fusion matrix, and taking each parameter factor as a column index of another matrix, taking a risk level of a risk event as a row index of the other matrix, and filling a data value and a weight value corresponding to each parameter factor to corresponding index positions to form a matrix for a risk event prediction dimension, and marking the matrix as a risk judgment matrix, wherein the risk event comprises a vehicle collision event. In a first possible implementation manner of the first aspect, a matrix for representing a data feature fusion dimension and a matrix for a risk event prediction dimension may be established according to driving parameters, so that when a weighting calculation is performed to obtain a light adjustment weighting threshold value, vehicle light adjustment can be performed from the two dimensions of the data feature fusion and the risk event prediction. In a second possible implementation manner of the first aspect, the process of determining the two matrices with different dimensions according to the plurality of parameter factors includes that each parameter factor is used as a row index and a column index of one matrix respectively, weight values of all parameter factors after being combined in pairs are determined based on the weight values of all parameter factors, the weight values of all parameter factors after being combined in pairs are filled to corresponding index positions to form a matrix used for representing a data f