A vehicle control method and system based on light intensity

Abstract

This invention discloses a vehicle control method and system based on light intensity. The method includes: acquiring time-geographic data of the vehicle's current driving position; calculating the solar altitude angle based on the time-geographic data; determining the road segment to be tested based on the vehicle's driving route, driving position, and speed; acquiring tree and building information of the road segment to be tested; calculating the shadow area ratio based on the solar altitude angle, tree information, building information, and the total road area of ​​the road segment to be tested; acquiring the light intensity and atmospheric transparency of the environment in which the vehicle is currently located; calculating a light intensity safety index value based on the light intensity, atmospheric transparency, and shadow area ratio; and alerting the driver when the light intensity safety index value exceeds a preset light intensity safety threshold. This invention can provide an early warning of light intensity to the driver when the light intensity of the planned driving segment ahead of the vehicle is strong, prompting the driver to avoid the effects of strong external light in advance, thereby ensuring driving safety.

Description

Technical Field

[0001] This invention relates to the field of vehicle technology, and in particular to a vehicle control method and system based on light intensity. Background Technology

[0002] As a common means of transportation in modern society, vehicles are closely related to people's daily lives. While driving, drivers need to constantly pay attention to surrounding traffic conditions to ensure driving safety. However, due to varying light intensity on different road sections and at different times of day, strong external light can impair a driver's vision, making it difficult to see clearly and maintain focus. If this occurs during complex traffic situations or lane changes, it can easily lead to traffic accidents. Therefore, some drivers choose to wear sunglasses while driving. However, wearing sunglasses darkens the field of vision, reduces visual clarity, and causes the pupils to dilate unconsciously. Over time, this increases eye fatigue, thus affecting driving safety. Summary of the Invention

[0003] This invention provides a vehicle control method and system based on light intensity, which can provide the driver with an early warning of light intensity when the light intensity of the planned driving section in front of the vehicle is strong, so as to prompt the driver to avoid the impact of strong external light in advance, thereby ensuring driving safety.

[0004] This invention provides a vehicle control method based on light intensity, comprising:

[0005] Obtain the time-geographic data of the vehicle's current driving location; wherein, the time-geographic data includes the geographical latitude, geographical longitude, Beijing time, and number of days of the driving location;

[0006] Calculate the solar altitude angle at the driving location based on the aforementioned time-geographic data;

[0007] The vehicle's current planned driving route and speed are obtained. Based on the driving route, driving position, and speed, the test road segment that the vehicle will enter after a preset time is determined.

[0008] Obtain tree information and building information for the road segment to be tested; wherein, the tree information includes the tree height and tree width of trees on both sides of the road segment to be tested, and the building information includes the building height and building width of buildings on both sides of the road segment to be tested;

[0009] Based on the solar altitude angle, the tree information, the building information, and the pre-acquired total road area of ​​the road segment to be tested, the proportion of the shaded area on the road segment to be tested is calculated;

[0010] Obtain the light intensity and atmospheric transparency of the current environment in which the vehicle is located;

[0011] The light intensity safety index value of the road section under test is calculated based on the light intensity, atmospheric transparency, and shadow area ratio.

[0012] When the light intensity safety index value exceeds the preset light intensity safety threshold, a light intensity warning is issued to the driver according to the pre-configured prompting strategy.

[0013] As an improvement to the above solution, the step of calculating the solar altitude angle of the driving location based on the time-geographic data includes:

[0014] Calculate the actual location time of the driving position based on the Beijing time and the geographical longitude;

[0015] Calculate the hour angle of the driving position based on the actual location and time;

[0016] Calculate the declination angle of the travel position based on the number of days;

[0017] Calculate the solar altitude angle of the driving position based on the geographical latitude, the declination angle, and the hour angle.

[0018] As an improvement to the above scheme, the step of calculating the proportion of shadow area on the road segment to be measured based on the solar altitude angle, the tree information, the building information, and the pre-acquired total road segment area includes:

[0019] Based on the solar altitude angle, the tree height and tree width on both sides of the road section to be measured, calculate the area of ​​the tree shadows cast by the trees on both sides of the road section to be measured on the road section to be measured.

[0020] Based on the solar altitude angle, the building height and building width of the buildings on both sides of the road section to be measured, calculate the area of ​​the building shadows cast by the buildings on both sides of the road section to be measured on the road section to be measured;

[0021] The total shadow area on the road segment under test is obtained by summing the shadow areas of the trees and the buildings.

[0022] The ratio of the total shaded area to the total road area of ​​the road segment to be tested is calculated to obtain the shaded area ratio of the road segment to be tested.

[0023] As an improvement to the above scheme, the step of calculating the area of ​​tree shadows cast by the trees on both sides of the road segment under test based on the solar altitude angle, the tree height, and the tree width on both sides of the road segment under test is specifically as follows:

[0024] The tree projection area of ​​each tree on both sides of the road segment under test is calculated according to the following formula, and the tree shadow area of ​​the road segment under test is obtained by summing all the tree projection areas:

[0025] S 树木 =w 树木 *(h 树木 / tanH S );

[0026] Among them, S 树木 w represents the projected area of ​​each tree on the road segment under test. 树木 h is the width of each tree. 树木 For the height of each tree, H S This is the solar altitude angle.

[0027] As an improvement to the above scheme, the calculation of the building shadow area of ​​the buildings on both sides of the road segment under test based on the solar altitude angle, the building height and building width of the buildings on both sides of the road segment under test is specifically as follows:

[0028] The projected area of ​​each building on both sides of the road segment under test is calculated using the following formula, and the projected areas of all buildings are summed to obtain the shaded area of ​​the buildings on the road segment under test:

[0029] S 建筑物 =w 建筑物 *(h 建筑物 / tanH S );

[0030] Among them, S 建筑物 w represents the projected area of ​​each building on the road segment under test. 建筑物 h is the width of each building. 建筑物 H represents the building height of each building. S This is the solar altitude angle.

[0031] As an improvement to the above scheme, the step of calculating the light intensity safety index value of the road section to be tested based on the light intensity, the atmospheric transparency, and the shadow area ratio includes:

[0032] The feature vector to be measured is constructed from the shadow area ratio, the light intensity, and the atmospheric transparency.

[0033] The feature vector to be tested is input into a pre-trained convolutional neural network to obtain the light intensity safety index value of the road section to be tested; wherein, the convolutional neural network is trained based on several feature vectors used for training and the corresponding light intensity safety index values; the feature vectors used for training are constructed from different shadow area ratios, light intensities and atmospheric transparency.

[0034] As an improvement to the above scheme, the convolutional neural network is composed of an embedding layer, a convolutional layer, a pooling layer, a first fully connected layer, a second fully connected layer, and a third fully connected layer connected in sequence; wherein, the first fully connected layer uses the ReLU activation function, the second fully connected layer uses the Softmax function, and the third fully connected layer uses the argmax function.

[0035] As an improvement to the above solution, when the light intensity safety index value exceeds the preset light intensity safety threshold, a light intensity warning is issued to the driver according to a pre-configured alert strategy, specifically as follows:

[0036] When the light intensity safety index value exceeds the preset light intensity safety threshold, a light intensity warning will be issued to the driver via voice broadcast, and the driver will be prompted to change the driving route.

[0037] As an improvement to the above solution, the vehicle control method further includes:

[0038] After the vehicle finishes its trip, if the driver inputs a light intensity safety evaluation value for any test segment during the trip, the light intensity safety evaluation value is input into the convolutional neural network. The error between the light intensity safety evaluation value and the light intensity safety index value of the test segment is calculated, and the error value is backpropagated. The parameters of the convolutional neural network are then adjusted using a gradient descent algorithm.

[0039] Accordingly, another embodiment of the present invention provides a vehicle control method system based on light intensity, comprising:

[0040] The location information acquisition module is used to acquire the time-geographic data of the vehicle's current driving location; wherein, the time-geographic data includes the geographical latitude, geographical longitude, Beijing time and number of days of the driving location;

[0041] The solar altitude angle calculation module is used to calculate the solar altitude angle of the driving position based on the time-geographic data.

[0042] The test road segment determination module is used to obtain the vehicle's currently planned driving route and speed, and based on the driving route, driving position and speed, determine the test road segment that the vehicle will enter after a preset time.

[0043] The road segment information acquisition module acquires tree information and building information of the road segment to be tested; wherein, the tree information includes the tree height and tree width of the trees on both sides of the road segment to be tested, and the building information includes the building height and building width of the buildings on both sides of the road segment to be tested;

[0044] The shadow area calculation module is used to calculate the shadow area ratio on the road segment to be measured based on the solar altitude angle, the tree information, the building information, and the total road segment area of ​​the road segment to be measured obtained in advance.

[0045] The light intensity information acquisition module is used to acquire the light intensity and atmospheric transparency of the current environment in which the vehicle is located;

[0046] The safety index calculation module is used to calculate the light intensity safety index value of the road section to be tested based on the light intensity, the atmospheric transparency and the shadow area ratio.

[0047] The light intensity warning module is used to issue a light intensity warning to the driver according to a pre-configured prompting strategy when the light intensity safety index value exceeds the preset light intensity safety threshold.

[0048] Compared with existing technologies, the vehicle control method and system based on light intensity disclosed in this invention first acquires the time-geographic data of the vehicle's current driving position; wherein, the time-geographic data includes the geographical latitude, geographical longitude, Beijing time, and number of days of the driving position; and calculates the solar altitude angle of the driving position based on the time-geographic data; secondly, acquires the vehicle's currently planned driving route and speed, and determines the test road segment that the vehicle will enter after a preset time period based on the driving route, the driving position, and the speed; and acquires tree information and building information of the test road segment; wherein, the tree information includes the tree height and tree width of trees on both sides of the test road segment, and the building information includes buildings on both sides of the test road segment. The system calculates the building height and width; then, based on the solar altitude angle, tree information, building information, and the pre-acquired total road area of ​​the road segment to be tested, it calculates the shadow area ratio on the road segment to be tested; and acquires the light intensity and atmospheric transparency of the current environment in which the vehicle is located; finally, based on the light intensity, atmospheric transparency, and shadow area ratio, it calculates the light intensity safety index value of the road segment to be tested, thereby inferring the degree of influence of the light intensity of the road segment that the vehicle is about to enter on the driver, and when the light intensity safety index value exceeds the preset light intensity safety threshold, it provides a light intensity warning to the driver according to a pre-configured prompting strategy to prompt the driver to avoid the impact of strong external light in advance, thereby ensuring driving safety. Attached Figure Description

[0049] Figure 1 This is a schematic flowchart of a vehicle control method based on light intensity provided in an embodiment of the present invention;

[0050] Figure 2 This is a schematic diagram of the structure of a convolutional neural network provided in an embodiment of the present invention;

[0051] Figure 3 This is a schematic diagram of the projected area of ​​a tree or building provided in an embodiment of the present invention.

[0052] Figure 4 This is a schematic diagram of a vehicle control method system based on light intensity provided in an embodiment of the present invention. Detailed Implementation

[0053] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0054] See Figure 1 , Figure 1 This is a schematic flowchart of a vehicle control method based on light intensity provided in an embodiment of the present invention.

[0055] The vehicle control method based on light intensity provided in this embodiment of the invention includes the following steps:

[0056] S11. Obtain the time-geographic data of the vehicle's current driving location; wherein, the time-geographic data includes the geographical latitude, geographical longitude, Beijing time, and number of days of the driving location;

[0057] S12. Calculate the solar altitude angle of the driving location based on the time-geographic data;

[0058] S13. Obtain the vehicle's currently planned driving route and speed, and based on the driving route, driving position and speed, determine the test section that the vehicle will enter after a preset time.

[0059] S14. Obtain tree information and building information of the road section to be tested; wherein, the tree information includes the tree height and tree width of the trees on both sides of the road section to be tested, and the building information includes the building height and building width of the buildings on both sides of the road section to be tested;

[0060] S15. Based on the solar altitude angle, the tree information, the building information, and the pre-acquired total road area of ​​the road segment to be tested, calculate the proportion of the shadow area on the road segment to be tested;

[0061] S16. Obtain the light intensity and atmospheric transparency of the current environment in which the vehicle is located;

[0062] S17. Calculate the light intensity safety index value of the road section to be tested based on the light intensity, the atmospheric transparency and the shadow area ratio;

[0063] S18. When the light intensity safety index value exceeds the preset light intensity safety threshold, a light intensity warning is given to the driver according to the pre-configured prompting strategy.

[0064] It should be noted that the light intensity safety threshold can be a pre-set threshold by the driver or other passengers, or a threshold calculated by the vehicle based on big data statistics, with the maximum acceptable light intensity safety index value that does not affect the driver's safe driving as a reference. In practice, after the vehicle is started, the driver can access the neural network parameters previously stored in the database through the vehicle network, and can also view the previously preset light intensity safety thresholds in the database. If the light intensity safety threshold set by the driver or other passengers cannot be obtained, the driver or other users can be prompted to enter the set light intensity safety threshold on the corresponding input page of the in-vehicle screen.

[0065] Specifically, calculating the solar altitude angle of the driving location based on the time-geographic data includes:

[0066] Calculate the actual location time of the driving position based on the Beijing time and the geographical longitude;

[0067] Calculate the hour angle of the driving position based on the actual location and time;

[0068] Calculate the declination angle of the travel position based on the number of days;

[0069] Calculate the solar altitude angle of the driving position based on the geographical latitude, the declination angle, and the hour angle.

[0070] It should be noted that the number of days mentioned refers to the cumulative number of days starting from January 1st of this year. For example, February 3rd, 2022 corresponds to 34 days.

[0071] Specifically, the step of calculating the actual location time of the driving position based on the Beijing time and the geographical longitude is as follows:

[0072] The actual location time at the stated driving position is calculated using the following formula:

[0073]

[0074] Where t′ represents the actual location and time of the driving position, t represents Beijing time, and α represents the geographical longitude of the driving position.

[0075] It should be noted that time differs across latitudes. The actual time at the travel location, converted to Beijing time, corresponds to a 1-hour time difference for every 1° difference in longitude. Therefore, the actual time at the travel location can be calculated based on the Beijing time and the geographical longitude.

[0076] Specifically, calculating the hour angle of the driving position based on the actual location and time involves:

[0077] The hour angle of the driving position is calculated using the following formula:

[0078] ω = (t′-12)*15°;

[0079] Where ω is the hour angle of the driving position.

[0080] It is understandable that the hour angle is calculated starting from 12 noon as 0°, and the hour angle increases by 15° every hour.

[0081] Specifically, the calculation of the declination angle of the travel position based on the number of days is as follows:

[0082] The declination angle of the driving position is calculated using the following formula:

[0083]

[0084] Where δ is the declination angle of the driving position, and N is the number of days.

[0085] Specifically, the step of calculating the solar altitude angle of the driving position based on the geographical latitude, the declination angle, and the hour angle is as follows:

[0086] The solar altitude angle at the driving position is calculated using the following formula:

[0087]

[0088] Among them, H S The solar altitude angle, The geographical latitude of the driving location.

[0089] As one specific embodiment, the calculation of the shadow area ratio on the road segment to be measured based on the solar altitude angle, the tree information, the building information, and the pre-acquired total road segment area includes:

[0090] Based on the solar altitude angle, the tree height and tree width on both sides of the road section to be measured, calculate the area of ​​the tree shadows cast by the trees on both sides of the road section to be measured on the road section to be measured.

[0091] Based on the solar altitude angle, the building height and building width of the buildings on both sides of the road section to be measured, calculate the area of ​​the building shadows cast by the buildings on both sides of the road section to be measured on the road section to be measured;

[0092] The total shadow area on the road segment under test is obtained by summing the shadow areas of the trees and the buildings.

[0093] The ratio of the total shaded area to the total road area of ​​the road segment to be tested is calculated to obtain the shaded area ratio of the road segment to be tested.

[0094] Preferably, the preset duration is 5 minutes.

[0095] It is worth noting that, taking a preset duration of 5 minutes as an example, when the vehicle estimates that it will enter the next road segment in 5 minutes based on the currently planned driving route, speed and current driving position, the next road segment is taken as the test segment. The standard map and high-precision map of the current location are obtained through the vehicle's Internet of Vehicles. The building information on both sides of the test segment is obtained from the standard map, and the tree information is obtained from the high-precision map.

[0096] It is worth noting that due to the solar azimuth angle Therefore, see Figure 2 It can be derived that the projected area of ​​trees or buildings on a road segment is S = w * (h / tanH). S ); where w is the width of the tree or building, and h is the height of the tree or building.

[0097] Furthermore, the step of calculating the area of ​​tree shadows cast by the trees on both sides of the road segment under test based on the solar altitude angle, the tree height, and the tree width on both sides of the road segment under test is specifically as follows:

[0098] The tree projection area of ​​each tree on both sides of the road segment under test is calculated according to the following formula, and the tree shadow area of ​​the road segment under test is obtained by summing all the tree projection areas:

[0099] S 树木 =w 树木 *(h 树木 / tanH S );

[0100] Among them, S 树木 w represents the projected area of ​​each tree on the road segment under test.树木 h is the width of each tree. 树木 For the height of each tree, H S This is the solar altitude angle.

[0101] Furthermore, the step of calculating the building shadow area of ​​the buildings on both sides of the road segment under test based on the solar altitude angle, the building height and building width of the buildings on both sides of the road segment under test is specifically as follows:

[0102] The projected area of ​​each building on both sides of the road segment under test is calculated using the following formula, and the projected areas of all buildings are summed to obtain the shaded area of ​​the buildings on the road segment under test:

[0103] S 建筑物 =w 建筑物 *(h 建筑物 / tanH S );

[0104] Among them, S 建筑物 w represents the projected area of ​​each building on the road segment under test. 建筑物 h is the width of each building. 建筑物 H represents the building height of each building. S This is the solar altitude angle.

[0105] It should be noted that the total area of ​​the road segment to be tested can be calculated from the road segment length and width contained in the standard map (such as Baidu Map) or high-precision map obtained by the vehicle.

[0106] In some preferred embodiments, calculating the light intensity safety index value of the road segment under test based on the light intensity, the atmospheric transparency, and the shadow area ratio includes:

[0107] The feature vector to be measured is constructed from the shadow area ratio, the light intensity, and the atmospheric transparency.

[0108] The feature vector to be tested is input into a pre-trained convolutional neural network to obtain the light intensity safety index value of the road section to be tested; wherein, the convolutional neural network is trained based on several feature vectors used for training and the corresponding light intensity safety index values; the feature vectors used for training are constructed from different shadow area ratios, light intensities and atmospheric transparency.

[0109] In this embodiment, the light intensity is measured in real time by a light intensity sensor installed on the vehicle, and the atmospheric transparency is obtained by the vehicle from the network via the vehicle-to-everything (V2X) network.

[0110] See Figure 3 In one specific implementation, the convolutional neural network is composed of an embedding layer, a convolutional layer, a pooling layer, a first fully connected layer, a second fully connected layer, and a third fully connected layer connected in sequence; wherein, the first fully connected layer uses the ReLU activation function, the second fully connected layer uses the Softmax function, and the third fully connected layer uses the argmax function.

[0111] Specifically, the first fully connected layer has 64 neurons, and the second fully connected layer has 10 neurons.

[0112] Understandably, in this embodiment, the shadow area ratio, the light intensity, and the atmospheric transparency are used as reference indicators to calculate the light intensity safety index value for the next road segment (i.e., the road segment to be tested). If the light intensity safety index value exceeds a preset light intensity safety threshold, a warning is issued to the driver, suggesting that the driver change the route. It should be noted that calculating the light intensity safety index value using a convolutional neural network is only one preferred implementation method. In actual operation, the light intensity safety index value can also be obtained by summing the shadow area ratio, the light intensity, and the atmospheric transparency according to a preset ratio, etc., and this is not limited here. Of course, in subsequent processes, reference indicators can be added or reduced based on feedback from drivers and other users and actual conditions.

[0113] Specifically, the feature vector to be measured is X = [A, B, C]; where A is the shadow area ratio, B is the light intensity, and C is the atmospheric transparency.

[0114] It is worth noting that the embedding layer performs a dimensionality transformation preprocessing on the one-dimensional feature vector to be tested using word2vecc or one-hot encoding. In this embodiment, one-hot encoding is chosen to transform the dimensionality of the feature vector to be tested. This can solve the problem of classifiers having difficulty processing discrete data and can expand the features to a certain extent. However, the features obtained by one-hot encoding are discrete and sparse. Therefore, the embedding layer can represent each class variable with fewer dimensions and can also show the relationship between different class variables to a certain extent. After converting the feature vector to be tested into a two-dimensional feature matrix, the two-dimensional feature matrix is ​​put into convolutional and pooling layers for feature extraction; finally, the light intensity safety index value S of the road segment to be tested is obtained in the fully connected layer.

[0115] Taking the light intensity safety index value as an integer ranging from [1-10] as an example, the original output values ​​are processed by the Softmax function of the second fully connected layer. The interrelationship between the probabilities of the different light intensity safety index values ​​obtained conforms to the probability distribution, as shown in the following formula:

[0116]

[0117] Among them, Z j Z is the output value of the j-th neuron in the second fully connected layer. k The output of the forward propagation is the output of one of the neurons in the second fully connected layer, where K is the number of neurons in the second fully connected layer, representing the category of the light intensity safety index value. For Z k The summation of the exponential function e.

[0118] Therefore, the light intensity safety index value x output by the convolutional neural network from the feature vector to be tested is calculated according to the following formula:

[0119]

[0120] Where, f(x) = Z j Let x be the light intensity safety index value, and y be any value within the range of all values ​​of the light intensity safety index value x. Let y = f(x), and f(x) and f(y) be a mapping relationship.

[0121] In some preferred embodiments, when the light intensity safety index value exceeds a preset light intensity safety threshold, a light intensity warning is issued to the driver according to a pre-configured alert strategy, specifically as follows:

[0122] When the light intensity safety index value exceeds the preset light intensity safety threshold, a light intensity warning will be issued to the driver via voice broadcast, and the driver will be prompted to change the driving route.

[0123] It is understood that when the light intensity safety index value exceeds the preset light intensity safety threshold, a light intensity warning is issued to the driver via voice broadcast to prompt the driver to change the driving route. If the driver changes the driving route again, the light intensity safety index value of the next road segment to be tested is determined based on the changed driving route. It should be noted that the pre-configured prompt strategy, in addition to voice broadcast reminders, can also assist in warning the driver through methods such as lights and vibration. When the vehicle detects that the light intensity safety index value exceeds the preset light intensity safety threshold, it can also replan a driving route with less impact from light intensity based on the vehicle's current driving position and the destination in the driving route, for the driver and other users to choose from.

[0124] It is worth noting that if the light transmittance and color of the vehicle's windshield are adjustable, such as if the windshield is made of electrochromic material or equipped with an electrochromic film or other sunshade film, the color and / or light transmittance of the windshield can be adjusted when the light intensity safety index value exceeds the preset light intensity safety threshold, so as to reduce the impact of external light on the driver.

[0125] Furthermore, the vehicle control method further includes:

[0126] After the vehicle finishes its trip, if the driver inputs a light intensity safety evaluation value for any test segment during the trip, the light intensity safety evaluation value is input into the convolutional neural network. The error between the light intensity safety evaluation value and the light intensity safety index value of the test segment is calculated, and the error value is backpropagated. The parameters of the convolutional neural network are then adjusted using a gradient descent algorithm.

[0127] Specifically, the convolutional neural network calculates the error value L between the light intensity safety evaluation value and the light intensity safety index value of the road segment under test according to the following formula:

[0128]

[0129] Where C represents the total number of scoring intervals for the light intensity safety index value of the road section under test. For symbolic functions, The output of the second fully connected layer represents the predicted probability that the feature vector X to be tested belongs to the m-th scoring interval, where N is the number of feature vectors input in this round of error analysis, and L... X The loss function is the loss function of the feature vector X to be tested, which is the output of the neuron in the second fully connected layer.

[0130] It should be noted that if the second fully connected layer has 10 neurons, then L X Let L be the loss function of the measured feature vector X output by each neuron in the second fully connected layer. X The number is 10.

[0131] Specifically, the light intensity safety index value of the road section to be tested is pre-divided into several scoring intervals. For example, the upper and lower limits of the light intensity safety index value are [1-10], and the segment difference value of each scoring interval is set to 1, then C equals 10. At this point, there are 10 different scoring intervals, i.e., m = 1, 2, 3, 4, 5, 6, 7, 8, 9, 10. If the light intensity safety evaluation value (i.e., the true interval of the feature vector X to be tested) is equal to the m-th scoring interval, then... If the light intensity safety evaluation value is not equal to the m-th scoring interval, then... It equals 0.

[0132] For example, after the driver's trip, if the driver finds the light intensity safety protection experience satisfactory and gives a default positive review, it means the driver accepts the system-generated light intensity safety index value, and the convolutional neural network's error analysis process will not be initiated. If the passenger is dissatisfied with the light intensity safety protection, they can rate a specific segment of the trip. The driver's light intensity safety evaluation value S0 will be compared with the system-generated light intensity safety index value S, and the data will be uploaded to a remote server. The remote server's convolutional neural network will then perform error analysis. During error analysis, the error value L is first calculated, and then the convolutional neural network backpropagates the error value L, automatically adjusting the corresponding parameters at each level using the gradient descent algorithm. After multiple rounds of these adjustments, the light intensity safety index value predicted by the convolutional neural network will better align with the passenger's subjective rating.

[0133] It should be noted that, in addition to using convolutional neural networks to calculate the light intensity safety index value, other neural networks and deep learning models with classification functions can also be selected, and there are no restrictions on them here.

[0134] It's worth noting that after a driver completes the rating, the vehicle uploads the relevant parameters from each layer of the convolutional neural network to the database for future use when passing the same road segment. Simultaneously, a remote server can also be configured with a neural network to train the coefficients of the rating formula using massive amounts of data uploaded to the database by different drivers. This generates a universal rating formula for road segment light intensity safety indicators, sufficient for the initial use of most new drivers. Subsequently, customized rating formulas for specific road segment light intensity safety indicators can be trained based on driver ratings in actual use.

[0135] See Figure 4 This is a schematic diagram of a vehicle control method system based on light intensity provided in an embodiment of the present invention.

[0136] The vehicle control method system based on light intensity provided in this embodiment of the invention includes:

[0137] The location information acquisition module 21 is used to acquire the time-geographic data of the vehicle's current driving location; wherein, the time-geographic data includes the geographical latitude, geographical longitude, Beijing time and number of days of the driving location;

[0138] The solar altitude angle calculation module 22 is used to calculate the solar altitude angle of the driving position based on the time-geographic data.

[0139] The test road segment determination module 23 is used to obtain the vehicle's currently planned driving route and speed, and based on the driving route, driving position and speed, determine the test road segment that the vehicle will enter after a preset time.

[0140] The road segment information acquisition module 24 acquires tree information and building information of the road segment to be tested; wherein, the tree information includes the tree height and tree width of the trees on both sides of the road segment to be tested, and the building information includes the building height and building width of the buildings on both sides of the road segment to be tested;

[0141] The shadow area calculation module 25 is used to calculate the shadow area ratio on the road segment to be measured based on the solar altitude angle, the tree information, the building information and the total road segment area of ​​the road segment to be measured obtained in advance;

[0142] The light intensity information acquisition module 26 is used to acquire the light intensity and atmospheric transparency of the current environment of the vehicle.

[0143] The safety index calculation module 27 is used to calculate the light intensity safety index value of the road section to be tested based on the light intensity, the atmospheric transparency and the shadow area ratio.

[0144] The light intensity warning module 28 is used to provide a light intensity warning to the driver according to a pre-configured prompting strategy when the light intensity safety index value exceeds the preset light intensity safety threshold.

[0145] As an improvement to the above solution, the solar altitude angle calculation module 22 includes:

[0146] The time calculation unit is used to calculate the actual location time of the driving position based on the Beijing time and the geographical longitude.

[0147] The hour angle calculation unit is used to calculate the hour angle of the driving position based on the actual location time;

[0148] The declination angle calculation unit is used to calculate the declination angle of the driving position based on the number of days.

[0149] The solar altitude angle calculation unit is used to calculate the solar altitude angle of the driving position based on the geographical latitude, the declination angle, and the hour angle.

[0150] Specifically, the shadow area calculation module 25 includes:

[0151] The tree shadow calculation unit is used to calculate the area of ​​the tree shadows cast by the trees on both sides of the road section under test based on the solar altitude angle, the tree height and tree width of the trees on both sides of the road section under test.

[0152] The building shadow calculation unit is used to calculate the building shadow area of ​​the buildings on both sides of the road segment under test based on the solar altitude angle, the building height and building width of the buildings on both sides of the road segment under test;

[0153] The road segment shadow calculation unit is used to calculate the sum of the shadow areas of the trees and the shadow areas of the buildings to obtain the total shadow area on the road segment to be measured;

[0154] The shadow ratio calculation unit is used to calculate the ratio of the total shadow area to the total road area of ​​the road segment to be tested, thereby obtaining the shadow area ratio on the road segment to be tested.

[0155] Furthermore, the tree shadow calculation unit is specifically used for:

[0156] The tree projection area of ​​each tree on both sides of the road segment under test is calculated according to the following formula, and the tree shadow area of ​​the road segment under test is obtained by summing all the tree projection areas:

[0157] S 树木 =w 树木 *(h 树木 / tanH S );

[0158] Among them, S 树木 w represents the projected area of ​​each tree on the road segment under test. 树木 h is the width of each tree. 树木 For the height of each tree, H S This is the solar altitude angle.

[0159] Furthermore, the building shadow calculation unit is specifically used for:

[0160] The projected area of ​​each building on both sides of the road segment under test is calculated using the following formula, and the projected areas of all buildings are summed to obtain the shaded area of ​​the buildings on the road segment under test:

[0161] S 建筑物 =w 建筑物 *(h 建筑物 / tanH S );

[0162] Among them, S 建筑物 w represents the projected area of ​​each building on the road segment under test. 建筑物 h is the width of each building. 建筑物 H represents the building height of each building. S This is the solar altitude angle.

[0163] In one preferred embodiment, the security indicator calculation module 27 is specifically used for:

[0164] The feature vector to be measured is constructed from the shadow area ratio, the light intensity, and the atmospheric transparency.

[0165] The feature vector to be tested is input into a pre-trained convolutional neural network to obtain the light intensity safety index value of the road section to be tested; wherein, the convolutional neural network is trained based on several feature vectors used for training and the corresponding light intensity safety index values; the feature vectors used for training are constructed from different shadow area ratios, light intensities and atmospheric transparency.

[0166] Specifically, the convolutional neural network in the security index calculation module 27 is composed of an embedding layer, a convolutional layer, a pooling layer, a first fully connected layer, a second fully connected layer, and a third fully connected layer connected in sequence; wherein, the first fully connected layer uses the ReLU activation function, the second fully connected layer uses the Softmax function, and the third fully connected layer uses the argmax function.

[0167] As one optional implementation, the light intensity warning module 28 is specifically used for:

[0168] When the light intensity safety index value exceeds the preset light intensity safety threshold, a light intensity warning will be issued to the driver via voice broadcast, and the driver will be prompted to change the driving route.

[0169] Furthermore, the vehicle control system also includes:

[0170] The network parameter correction module is used to, after the vehicle finishes its current trip, if it receives the light intensity safety evaluation value of any test road segment during the current trip input by the driver, input the light intensity safety evaluation value into the convolutional neural network, calculate the error value between the light intensity safety evaluation value of the test road segment and the light intensity safety index value, and backpropagate the error value to adjust the parameters of the convolutional neural network through the gradient descent algorithm.

[0171] It should be noted that the specific descriptions and beneficial effects of the various embodiments of the vehicle control method system based on light intensity in this embodiment can be referred to the specific descriptions and beneficial effects of the various embodiments of the vehicle control method based on light intensity described above, and will not be repeated here.

[0172] It should be noted that the system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the system embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0173] In summary, the vehicle control method and system based on light intensity provided by this invention first acquires the time-geographic data of the vehicle's current driving location; wherein the time-geographic data includes the geographical latitude, geographical longitude, Beijing time, and number of days of the driving location; and calculates the solar altitude angle of the driving location based on the time-geographic data; secondly, acquires the vehicle's currently planned driving route and speed, and determines the test road segment that the vehicle will enter after a preset time period based on the driving route, the driving location, and the speed; and acquires tree information and building information of the test road segment; wherein the tree information includes the tree height and tree width of trees on both sides of the test road segment, and the building information includes the buildings on both sides of the test road segment. The system calculates the building height and width; then, based on the solar altitude angle, tree information, building information, and the pre-acquired total road area of ​​the road segment to be tested, it calculates the shadow area ratio on the road segment to be tested; and acquires the light intensity and atmospheric transparency of the current environment in which the vehicle is located; finally, based on the light intensity, atmospheric transparency, and shadow area ratio, it calculates the light intensity safety index value of the road segment to be tested, thereby inferring the degree of influence of the light intensity of the road segment that the vehicle is about to enter on the driver, and when the light intensity safety index value exceeds the preset light intensity safety threshold, it provides a light intensity warning to the driver according to a pre-configured prompting strategy, so as to prompt the driver to avoid the impact of strong external light in advance, thereby ensuring driving safety.

[0174] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A vehicle control method based on light intensity, characterized in that, include: Obtain the time-geographic data of the vehicle's current driving location; wherein, the time-geographic data includes the geographical latitude, geographical longitude, Beijing time, and number of days of the driving location; Calculate the solar altitude angle at the driving location based on the aforementioned time-geographic data; The vehicle's current planned driving route and speed are obtained. Based on the driving route, driving position, and speed, the test road segment that the vehicle will enter after a preset time is determined. Obtain tree information and building information for the road segment to be tested; wherein, the tree information includes the tree height and tree width of trees on both sides of the road segment to be tested, and the building information includes the building height and building width of buildings on both sides of the road segment to be tested; Based on the solar altitude angle, the tree information, the building information, and the pre-acquired total road area of ​​the road segment to be tested, the proportion of the shaded area on the road segment to be tested is calculated; Obtain the light intensity and atmospheric transparency of the current environment in which the vehicle is located; The light intensity safety index value of the road section under test is calculated based on the light intensity, atmospheric transparency, and shadow area ratio. When the light intensity safety index value exceeds the preset light intensity safety threshold, a light intensity warning is issued to the driver according to the pre-configured prompting strategy.

2. The vehicle control method based on light intensity as described in claim 1, characterized in that, The step of calculating the solar altitude angle at the driving location based on the time-geographic data includes: Calculate the actual location time of the driving position based on the Beijing time and the geographical longitude; Calculate the hour angle of the driving position based on the actual location and time; Calculate the declination angle of the travel position based on the number of days; Calculate the solar altitude angle of the driving position based on the geographical latitude, the declination angle, and the hour angle.

3. The vehicle control method based on light intensity as described in claim 1, characterized in that, The calculation of the shadow area ratio on the road segment under test, based on the solar altitude angle, tree information, building information, and the pre-acquired total road segment area, includes: Based on the solar altitude angle, the tree height and tree width on both sides of the road section to be measured, calculate the area of ​​the tree shadows cast by the trees on both sides of the road section to be measured on the road section to be measured. Based on the solar altitude angle, the building height and building width of the buildings on both sides of the road section to be measured, calculate the area of ​​the building shadows cast by the buildings on both sides of the road section to be measured on the road section to be measured; The total shadow area on the road segment under test is obtained by summing the shadow areas of the trees and the buildings. The ratio of the total shaded area to the total road area of ​​the road segment to be tested is calculated to obtain the shaded area ratio of the road segment to be tested.

4. The vehicle control method based on light intensity as described in claim 3, characterized in that, The calculation of the area of ​​tree shadows cast on the road segment by the trees on both sides of the road segment under test, based on the solar altitude angle, tree height, and tree width, is specifically as follows: The tree projection area of ​​each tree on both sides of the road segment under test is calculated according to the following formula, and the tree shadow area of ​​the road segment under test is obtained by summing all the tree projection areas: ； in, Let the projected area of ​​each tree on the road segment to be measured be denoted as . For the width of each tree, For the height of each tree, This is the solar altitude angle.

5. The vehicle control method based on light intensity as described in claim 3, characterized in that, The calculation of the building shadow area on the road segment under test, based on the solar altitude angle, the building height and width of the buildings on both sides of the road segment under test, is specifically as follows: The projected area of ​​each building on both sides of the road segment under test is calculated using the following formula, and the projected areas of all buildings are summed to obtain the shaded area of ​​the buildings on the road segment under test: ； in, The projected area of ​​each building on the road segment to be measured is given. For the width of each building, For the building height of each building, This is the solar altitude angle.

6. The vehicle control method based on light intensity as described in claim 1, characterized in that, The step of calculating the light intensity safety index value of the road section under test based on the light intensity, atmospheric transparency, and shadow area ratio includes: The feature vector to be measured is constructed from the shadow area ratio, the light intensity, and the atmospheric transparency. The feature vector to be tested is input into a pre-trained convolutional neural network to obtain the light intensity safety index value of the road section to be tested; wherein, the convolutional neural network is trained based on several feature vectors used for training and the corresponding light intensity safety index values; the feature vectors used for training are constructed from different shadow area ratios, light intensities and atmospheric transparency.

7. The vehicle control method based on light intensity as described in claim 6, characterized in that, The convolutional neural network is composed of an embedding layer, a convolutional layer, a pooling layer, a first fully connected layer, a second fully connected layer, and a third fully connected layer connected in sequence; wherein, the first fully connected layer uses the ReLU activation function, the second fully connected layer uses the Softmax function, and the third fully connected layer uses the argmax function.

8. The vehicle control method based on light intensity as described in claim 1, characterized in that, When the light intensity safety index value exceeds the preset light intensity safety threshold, a light intensity warning is issued to the driver according to a pre-configured alert strategy, specifically: When the light intensity safety index value exceeds the preset light intensity safety threshold, a light intensity warning will be issued to the driver via voice broadcast, and the driver will be prompted to change the driving route.

9. The vehicle control method based on light intensity as described in claim 6, characterized in that, The vehicle control method further includes: After the vehicle finishes its trip, if the driver inputs a light intensity safety evaluation value for any test segment during the trip, the light intensity safety evaluation value is input into the convolutional neural network. The error between the light intensity safety evaluation value and the light intensity safety index value of the test segment is calculated, and the error value is backpropagated. The parameters of the convolutional neural network are then adjusted using a gradient descent algorithm.

10. A vehicle control system based on light intensity, characterized in that, include: The location information acquisition module is used to acquire the time-geographic data of the vehicle's current driving location; wherein, the time-geographic data includes the geographical latitude, geographical longitude, Beijing time and number of days of the driving location; The solar altitude angle calculation module is used to calculate the solar altitude angle of the driving position based on the time-geographic data. The test road segment determination module is used to obtain the vehicle's currently planned driving route and speed, and based on the driving route, driving position and speed, determine the test road segment that the vehicle will enter after a preset time. The road segment information acquisition module acquires tree information and building information of the road segment to be tested; wherein, the tree information includes the tree height and tree width of the trees on both sides of the road segment to be tested, and the building information includes the building height and building width of the buildings on both sides of the road segment to be tested; The shadow area calculation module is used to calculate the shadow area ratio on the road segment to be measured based on the solar altitude angle, the tree information, the building information, and the total road segment area of ​​the road segment to be measured obtained in advance. The light intensity information acquisition module is used to acquire the light intensity and atmospheric transparency of the current environment in which the vehicle is located; The safety index calculation module is used to calculate the light intensity safety index value of the road section to be tested based on the light intensity, the atmospheric transparency and the shadow area ratio. The light intensity warning module is used to issue a light intensity warning to the driver according to a pre-configured prompting strategy when the light intensity safety index value exceeds the preset light intensity safety threshold.

Images