CN117494589B - Vehicle accident prediction method, device and storage medium based on vehicle body color - Google Patents
Vehicle accident prediction method, device and storage medium based on vehicle body color Download PDFInfo
- Publication number
- CN117494589B CN117494589B CN202410003747.6A CN202410003747A CN117494589B CN 117494589 B CN117494589 B CN 117494589B CN 202410003747 A CN202410003747 A CN 202410003747A CN 117494589 B CN117494589 B CN 117494589B
- Authority
- CN
- China
- Prior art keywords
- accident
- vehicle
- chain
- data
- chains
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 45
- 239000003086 colorant Substances 0.000 claims description 20
- 238000004458 analytical method Methods 0.000 claims description 15
- 238000003062 neural network model Methods 0.000 claims description 11
- 230000002452 interceptive effect Effects 0.000 claims description 10
- 238000012549 training Methods 0.000 claims description 10
- 230000006870 function Effects 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000007613 environmental effect Effects 0.000 claims description 4
- 238000012545 processing Methods 0.000 abstract description 5
- 230000000875 corresponding effect Effects 0.000 description 39
- 206010039203 Road traffic accident Diseases 0.000 description 8
- 238000004590 computer program Methods 0.000 description 6
- 230000001276 controlling effect Effects 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000009471 action Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000002265 prevention Effects 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 230000001808 coupling effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000008685 targeting Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Abstract
The invention relates to the technical field of data processing, and discloses a vehicle accident prediction method, device and storage medium based on vehicle body color. The method comprises the following steps: the method comprises the steps of obtaining current operation information composed of vehicle body color combination data, environment data and running control data of a vehicle to be predicted, searching a target accident chain matched with the current operation information in all pre-stored accident chains, further determining an accident prediction grade of the vehicle to be predicted according to an accident grade corresponding to the target accident chain, searching a plurality of candidate accident chains matched with the vehicle body color combination data and the environment data in the current operation information in all pre-stored accident chains if the accident grade is greater than the preset grade, further selecting a reference accident chain with the accident grade lower than the accident prediction grade from the pre-stored accident chains, and controlling the running of the vehicle according to the running control data in the reference accident chain, so that the accident prediction is more reliable and accurate, and the accident grade of the vehicle to be predicted can be reduced.
Description
Technical Field
The present invention relates to the field of data processing technologies, and in particular, to a vehicle accident prediction method, apparatus, and storage medium based on a vehicle body color.
Background
The traffic accident severity analysis refers to a comprehensive analysis method for evaluating the influence degree of traffic accidents on personal property and traffic flow. The severity analysis of the traffic accident has important significance in the traffic safety field, and can help traffic management departments and vehicle drivers to better understand the influence and damage degree of the accident, so that more effective strategies are provided for the prevention and the response of the traffic accident. The number and severity of traffic accidents are greatly affected by the global degree of urbanization and traffic congestion conditions, and thus prevention and treatment of traffic accidents become a common problem in various countries and regions at present.
Current traffic accident severity analysis research is mainly focused on various factors such as road width, road surface quality, vehicle age, vehicle braking system, etc. Analysis of accident severity has also been developed around these factors. However, 80% of information of a driver is obtained through vision in the driving process, the visual characteristics of the driver determine the sensitivity to different colors, and the visual significance and decision ability of different vehicle body colors for the driver also show a certain difference under a certain condition, so that the accident frequency and the accident severity of the different vehicle body colors show a certain difference. In the existing accident severity analysis method, the influence of the vehicle body color on the accident severity is not considered, so that the analysis result has deviation.
In view of this, the present invention has been made.
Disclosure of Invention
In order to solve the technical problems, the invention provides a vehicle accident prediction method, device and storage medium based on the colors of a vehicle body, which are used for realizing accident prediction considering the colors of the vehicle body and improving the reliability of the accident prediction.
The embodiment of the invention provides a vehicle accident prediction method based on vehicle body color, which comprises the following steps:
acquiring current running information of a vehicle to be predicted, wherein the current running information comprises vehicle body color combination data, environment data and running control data, and the vehicle body color combination data is used for describing the colors of the vehicle to be predicted and other interactive vehicles;
searching a target accident chain matched with the current operation information in all pre-stored accident chains, wherein the accident chain consists of vehicle body color combination data, environment data and driving control data;
determining an accident prediction grade of the vehicle to be predicted according to the accident grade corresponding to the target accident chain, and if the accident prediction grade is greater than a preset grade, searching a plurality of candidate accident chains matched with the vehicle body color combination data and the environment data in the current running information from all the pre-stored accident chains;
and selecting a reference accident chain with the accident level lower than the accident prediction level from all candidate accident chains, and controlling the vehicle to be predicted to run according to the running control data in the reference accident chain.
The embodiment of the invention provides electronic equipment, which comprises:
a processor and a memory;
the processor is configured to execute the steps of the vehicle accident prediction method based on the vehicle body color according to any embodiment by calling the program or the instructions stored in the memory.
An embodiment of the present invention provides a computer-readable storage medium storing a program or instructions that cause a computer to execute the steps of the vehicle body color-based vehicle accident prediction method according to any of the embodiments.
The embodiment of the invention has the following technical effects:
the method comprises the steps of acquiring current operation information consisting of vehicle body color combination data, environment data and running control data of a vehicle to be predicted, searching a target accident chain matched with the current operation information in all pre-stored accident chains, further determining the accident prediction grade of the vehicle to be predicted according to the accident grade corresponding to the target accident chain, if the accident grade is greater than a preset grade, searching a plurality of candidate accident chains matched with the vehicle body color combination data and the environment data in the current operation information in all pre-stored accident chains, further selecting a reference accident chain with the accident grade lower than the accident prediction grade from the candidate accident chains, and controlling the running of the vehicle according to the running control data in the reference accident chain.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the description of the embodiments or the prior art will be briefly described, and it is obvious that the drawings in the description below are some embodiments of the present invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a vehicle accident prediction method based on vehicle body color provided by an embodiment of the invention;
FIG. 2 is a schematic diagram of a combination of colors of a vehicle body according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below. It will be apparent that the described embodiments are only some, but not all, embodiments of the invention. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the invention, are within the scope of the invention.
The vehicle accident prediction method based on the vehicle body color is mainly suitable for predicting whether the vehicle collides with other interactive vehicles or not and the severity of the collision condition based on the vehicle body color of the vehicle, the vehicle body color of other interactive vehicles, and the running environment data and control data of the vehicle. The vehicle accident prediction method based on the vehicle body color provided by the embodiment of the invention can be executed by electronic equipment in an integrated vehicle, such as a whole vehicle controller.
Fig. 1 is a flowchart of a vehicle accident prediction method based on a vehicle body color according to an embodiment of the present invention. Referring to fig. 1, the vehicle accident prediction method based on the vehicle body color specifically includes:
s110, acquiring current operation information of the vehicle to be predicted, wherein the current operation information comprises vehicle body color combination data, environment data and driving control data, and the vehicle body color combination data is used for describing colors of the vehicle to be predicted and other interactive vehicles.
In the embodiment of the invention, the vehicle body color combination data of the vehicle to be predicted can be obtained by combining the colors of the vehicle to be predicted and the colors of other interactive vehicles. The other interactive vehicles can be other vehicles which are positioned in the perception range of the vehicle to be predicted and have an overlapping part with the running track of the vehicle to be predicted.
Specifically, the color of other interactive vehicles can be obtained through a sensor on the vehicle to be predicted, such as a front camera, and the color of the vehicle to be predicted can also be obtained from pre-stored vehicle information.
Further, the color of the vehicle to be predicted and the color of other interactive vehicles can be combined, for example, the color of the vehicle to be predicted can be converted into an RGB channel value, the color of the other interactive vehicles can be also converted into an RGB channel value, and the converted two RGB channel values are combined in a weighted manner to obtain the vehicle body color combination data.
In the embodiment of the present invention, the environmental data of the vehicle to be predicted may be data for describing a road, weather, an in-vehicle environment, and an out-of-vehicle environment. By way of example, the environmental data may include road type, lane location, altitude, weather type, in-vehicle temperature, in-vehicle humidity, number of in-vehicle personnel, out-of-vehicle temperature, out-of-vehicle humidity, and the like.
For example, information such as the temperature in the vehicle, the humidity in the vehicle, the number of people in the vehicle and the like of the vehicle to be predicted can be obtained through a sensor on the vehicle to be predicted; or, a request can be sent to the cloud through a T-Box on the vehicle to be predicted, and the weather type, the outside temperature and the outside humidity fed back by the cloud are obtained; or, the current position of the vehicle can be obtained through a positioning device on the vehicle to be predicted, the current position is sent to the cloud end through a T-Box on the vehicle to be predicted, and the road type and the altitude fed back by the cloud end are obtained.
In the embodiment of the invention, the running control data may be data describing a control state during running of the vehicle to be predicted. By way of example, the travel control data may include data of speed, longitudinal acceleration, longitudinal deceleration, yaw rate, or steering angle, etc.
For example, the overall vehicle controller of the vehicle to be predicted may communicate with other controllers on the vehicle to be predicted, such as an engine controller, to obtain travel control data.
S120, searching a target accident chain matched with the current operation information in all pre-stored accident chains, wherein the accident chain consists of vehicle body color combination data, environment data and driving control data.
In the embodiment of the invention, the accident chain is used for describing the values of various parameters in association with the accident, namely the vehicle body color combination data, the environment data and the driving control data.
Each pre-stored accident chain has a corresponding accident level. The accident level may describe the severity of the accident, the higher the accident level, the higher the severity of the accident, e.g., the accident level may be 0 (representing no accident), 1 (representing a light collision), 2 (representing a moderate collision), 3 (representing a severe collision), etc. The embodiment of the invention does not limit the division mode of the accident level.
Specifically, a plurality of accident chains and corresponding accident levels can be obtained by analyzing each pre-collected historical accident data, or accident chains can be randomly generated, and the corresponding accident levels can be predicted by an accident prediction model.
In a specific implementation manner, the method provided by the embodiment of the invention further comprises the following steps:
step 11, acquiring each historical accident data, constructing a corresponding accident chain according to each historical accident data, and storing all accident chains and corresponding accident grades;
step 12, training to obtain an accident prediction model based on all accident chains and corresponding accident grades;
and 13, generating a plurality of new accident chains, determining the accident level corresponding to each new accident chain according to the accident prediction model, and storing all the new accident chains and the corresponding accident levels.
The historical accident data may be related data of accident vehicles collected in advance, for example, may include data of colors, environments, running control and the like of the accident vehicles.
For example, effective record data including accident vehicles, persons, roads, environments may be extracted from the accident information as historical accident data, and accident occurrence place, time, collision type, and the like may be extracted from the accident information to determine an accident level based thereon.
In consideration of the extracted historical accident data, abnormal values may exist, so that all the historical accident data can be preprocessed to remove missing values and abnormal values.
After obtaining a plurality of historical accident data, for each historical accident data, the vehicle body color combination data in the historical accident data can be determined, and then the vehicle body color combination data is coupled with other data to form a corresponding accident chain.
For the step 11, optionally, constructing a corresponding accident chain according to each historical accident data, including the following steps:
step 111, extracting the car body color chart numbers of the first collision car and the second collision car, the environment data and the running control data of the first collision car and the second collision car from the historical accident data for each historical accident data;
step 112, converting the car body color card numbers of the first collision car and the second collision car into RGB channel values to obtain a first car body color of the first collision car and a second car body color of the second collision car;
step 113, determining vehicle body color combination data according to the first vehicle body color and the second vehicle body color;
and 114, correlating the vehicle body color combination data, the environment data and the driving control data to obtain a corresponding accident chain.
In the embodiment of the invention, considering that certain differences may exist between vehicles with the same color, for example, the same vehicle color has a shade and a bright and dark score, and the historical accident data record is the RAL color card number of the accident vehicle, therefore, the vehicle can be quantified, and color numerical values are given to facilitate establishing connection with other data.
Specifically, the vehicle body color chart numbers (i.e., RAL color chart numbers) of the first collision vehicle and the second collision vehicle can be extracted from the historical accident data, wherein the first collision vehicle and the second collision vehicle are accident vehicles, and can be an automobile-automobile or an automobile-two-wheel vehicle.
Further, the body color chart numbers of the first collision vehicle and the second collision vehicle can be converted into RGB channel values, wherein the RGB channel values can comprise values in R, G, B channels, the value range of the values in each channel is 0-255, and the first body color of the first collision vehicle and the second body color of the second collision vehicle are obtained.
Further, the first body color and the second body color may be combined in a weighted manner. For example, the first body color and the second body color may be combined to obtain the body color combination data by the following formula:
；
in the method, in the process of the invention,for the first body color->For the second body colour->、/>、/>And the weight factors corresponding to the R, G, B color channels are respectively, and Y is the vehicle body color combination data.
For example, fig. 2 is a schematic view of a combination of colors of a vehicle body according to an embodiment of the present invention, as shown in fig. 2, a participant 1 of an accident (i.e. a first crashed vehicle) may be multiple colors, a participant 2 of an accident (i.e. a second crashed vehicle) may be multiple colors, and the colors of the participant 1 and the participant 2 may be combined.
Further, the vehicle body color combination data may be correlated with the environmental data and the travel control data extracted from the historical accident data, to obtain a corresponding accident chain. In addition, accident numbers may be associated together. It should be noted that each of the historical accident data may generate a corresponding accident chain.
Through the steps 111-114, the colors of the vehicles of the accident sides are coupled with other data to form accident chains, deviation of results caused by independent analysis of certain data without consideration of coupling action of other data is avoided, the colors of the vehicle bodies after the association and combination are coupled with other data in the historical accident data, the accident numbers are from the colors of the vehicle bodies of the accident sides, and then the vehicle bodies are coupled with other influencing factors to form an accident chain with comprehensive action of multiple factors, and the historical accident data of each accident can form an independent accident chain. The accident chains have the same factors, so that the chains can be crossed, a plurality of accident chains are crossed to form an accident network, the analysis of the severity of the accident under different color combinations can be facilitated, the data support is provided for the color distribution of vehicles and the analysis of traffic accidents of future roads, the design of the colors of the vehicles can be guided, the important significance is provided for reducing the severity of the accident, the traffic safety level can be further improved, and the method has application value in practical engineering.
In the embodiment of the invention, in order to further ensure the coverage comprehensiveness of the accident chain, besides extracting data from the historical accident data to form the accident chain and obtaining the corresponding accident grade, the accident prediction model can be trained according to the historical accident data, so as to generate a plurality of new accident chains, and the accident grade of the new accident chains is predicted by the model.
It should be noted that, considering that there may be data with very small correlation with the accident level in each data in the accident chain, that is, data with little influence on the accident level, such as in-vehicle humidity, etc., such data may be removed before training the model, so as to improve the training efficiency of the model and avoid that such data affects the accuracy of the model.
Optionally, before training to obtain the accident prediction model based on all accident chains and the corresponding accident levels, the method further comprises:
taking the accident level as an output parameter, and determining each input parameter according to each data in an accident chain; based on all accident chains and corresponding accident levels, analyzing the correlation and collinearity between each input parameter and each output parameter, and determining each associated parameter in all the input parameters based on the analysis result; and eliminating other input parameters except the associated parameters from all accident chains.
Specifically, the accident level may be used as an output parameter, a parameter corresponding to each data in the accident chain is used as an input parameter, the correlation and the collinearity between each input parameter and the output parameter are analyzed according to all the accident chains and the corresponding accident level, for example, a correlation coefficient between each input parameter and the output parameter is calculated, the correlation number is used as a correlation analysis result, a variance expansion factor between each input parameter and the output parameter is calculated, and the variance expansion factor is used as a collinearity analysis result.
Further, according to the correlation and the collinearity analysis result, the correlation parameters are selected from all the input parameters, for example, the input parameters of the first N of the correlation sequences are selected as the correlation parameters, and the input parameters of the collinearity sequences within the first M of the correlation parameters are removed from all the selected correlation parameters, so that repeated input parameters in the selected correlation parameters are removed, and the prediction precision of the model is ensured.
After all the associated parameters are determined, other input parameters except the associated parameters can be removed from all the accident chains, namely, corresponding data of the other input parameters except the associated parameters in each accident chain are removed.
According to the embodiment, the accident level can be used as the dependent variable, the parameters corresponding to the data in the accident chain are used as the independent variable, the correlation test and the co-linearity analysis between the dependent variable and the independent variable are carried out, the data corresponding to the parameters with small influence on the accident level in the accident chain are removed, the training efficiency of the model is improved, and the training precision of the model is prevented from being influenced by the data.
Furthermore, accident prediction models can be trained by adopting all accident chains and accident grades corresponding to the accident chains.
For the step 12, optionally, training to obtain an accident prediction model based on all accident chains and corresponding accident levels includes:
constructing a neural network model, and inputting all accident chains into the neural network model; and calculating the loss function according to the result output by the neural network model and the corresponding accident level, and adjusting network parameters in the neural network model according to the calculation result until the calculation result converges to obtain an accident prediction model.
Specifically, the neural network model may predict the accident level corresponding to each input accident chain, output the predicted result, and perform loss calculation according to the result output by the neural network, that is, the predicted level and the accident level corresponding to the accident chain, for example, a cross entropy loss function, a mean square error loss function, and the like, reversely adjust the network parameters in the neural network model according to the calculated loss value, and repeat the steps until the calculation result of the loss function converges, where the trained neural network model may be used as the accident prediction model.
Further, a plurality of new accident chains can be generated, for example, part of data in the accident chains generated according to the historical accident data can be adjusted to obtain the new accident chains; and inputting the new accident chain into the accident prediction model to obtain the accident grade output by the accident prediction model.
Through the steps 11-13, part of accident chains and corresponding accident grades can be obtained based on the historical accident data, part of accident chains and corresponding accident grades can be obtained based on the accident prediction model, all the accident chains and the corresponding accident grades are stored, the comprehensiveness of the stored accident chains is guaranteed, the situation that the corresponding accident chains cannot be matched in the subsequent accident prediction is avoided, and the reliability of the accident prediction is further guaranteed.
In the embodiment of the invention, the current running information of the vehicle to be predicted can be matched with each pre-stored accident chain, and then the accident chain matched with the current running information is found out as a target accident chain.
S130, determining an accident prediction grade of the vehicle to be predicted according to the accident grade corresponding to the target accident chain, and if the accident prediction grade is greater than a preset grade, searching a plurality of candidate accident chains matched with the vehicle body color combination data and the environment data in the current running information from all the pre-stored accident chains.
Specifically, after the target accident chain is matched, the corresponding accident level can be used as the accident prediction level of the vehicle to be predicted, so as to judge whether the accident prediction level is greater than a preset level. The preset level may be a preset critical accident level, such as 1 (representing a slight collision), which needs to reduce the severity of the accident.
Further, if the accident prediction grade is greater than the preset grade, the fact that the vehicle to be predicted has collision risk and high severity is indicated, and at the moment, the purpose of reducing the accident grade can be achieved by controlling the vehicle, so that the running safety of a user is ensured as much as possible.
Specifically, the current running information of the vehicle to be predicted may be divided into controllable information and uncontrollable information, wherein the controllable information is running control data, and the uncontrollable information is vehicle body color combination data and environment data, that is, the vehicle body color combination data and the environment data are fixed. And matching the accident chain according to the uncontrollable information in the accident chain, and obtaining a plurality of candidate accident chains matched with the vehicle body color combination data and the environment data.
And S140, selecting a reference accident chain with the accident level lower than the accident prediction level from all candidate accident chains, and controlling the vehicle to be predicted to run according to the running control data in the reference accident chain.
Specifically, after matching a plurality of candidate accident chains, each candidate accident chain is consistent with the vehicle body color combination data and the environment data in the target accident chain.
In order to reduce the accident level, an accident chain with the accident level lower than the accident prediction level can be selected from the candidate accident chains and used as a reference accident chain, and further, a control instruction is generated based on the running control data in the reference accident chain so as to control the running of the vehicle to be predicted based on the control instruction, so that the state of the vehicle to be predicted is adjusted to be in accordance with the running control data in the reference accident chain, and the aim of reducing the accident level of the vehicle is fulfilled.
In a specific embodiment, selecting a reference accident chain with an accident level lower than the accident prediction level from all candidate accident chains comprises:
selecting an accident chain to be selected from all candidate accident chains by taking the accident grade equal to the lowest grade as a target; if the number of the accident chains to be selected is at least one, selecting a reference accident chain from all the accident chains to be selected, otherwise, selecting the accident chains to be selected from all the candidate accident chains again with the accident level higher than the lowest level and lower than the accident prediction level as the target.
That is, the candidate accident chain with the lowest accident level may be selected as the candidate accident chain.
It should be noted that, in consideration of the possibility that the accident level of the accident chain to be selected is lower than the accident prediction level, the vehicle to be predicted cannot be adjusted by control to the control travel data in the accident chain to be selected, for example, cannot be decelerated to 30km/h, or the steering angle cannot be adjusted to 140 °. Therefore, in order to ensure the effectiveness of the driving control, the selected accident chain can be screened to remove the candidate accident chain which cannot be realized.
Optionally, after selecting the candidate accident chain from all the candidate accident chains, the method further includes:
and judging whether the running control data in the current running information can be adjusted to the running control data in the accident chain to be selected in the preset time aiming at each accident chain to be selected, and if not, eliminating the accident chain to be selected.
The preset time may be a preset critical time for controlling and adjusting the vehicle under the condition of accident risk. If the running control data in the current running information can be adjusted to the running control data in the accident chain to be selected within the preset time, the running control data in the accident chain to be selected can be realized within the preset time, and then the running control data can be reserved, otherwise, the running control data is removed.
Through the optional implementation manner, after the accident chain to be selected is screened out, the accident chain which cannot be realized by the running control data is removed, the effectiveness of the running control of the vehicle is ensured, and the running control according to the running control data which cannot be realized is avoided to cause larger safety risk.
Further, if the number of the accident chains to be selected is zero, the candidate accident chains with the accident level higher than the lowest level and lower than the accident prediction level can be selected again as the accident chains to be selected. Of course, after the selected accident chain is reselected, whether the selected accident chain which cannot be realized by the driving control data exists or not can be continuously judged so as to be eliminated.
If the number of accident chains to be selected is not zero, further, one accident chain can be selected as a reference accident chain. For example, the reference accident chain may be selected in combination with a gap between the travel control data in the reference accident chain and the travel control data in the current operation information.
Optionally, selecting a reference accident chain from all the accident chains to be selected, including:
determining an adjustment amplitude between the driving control data in each accident chain to be selected and the driving control data in the current operation information, and determining the accident chain to be selected with the minimum adjustment amplitude as a reference accident chain; or,
and determining the quantity of adjustment parameters between the driving control data in each accident chain to be selected and the driving control data in the current operation information, and determining the accident chain to be selected with the minimum quantity of the adjustment parameters as a reference accident chain.
The adjustment amplitude may be a difference value between the to-be-selected accident chain and data corresponding to the same parameter in the running control data of the current running information, such as a speed difference, an angle difference, and the like. The number of adjustment parameters may be the number of parameters to be adjusted in the driving control data of the accident chain to be selected and the current operation information, that is, the number of parameters having differences between the corresponding data.
Specifically, the smallest adjustment range can be targeted, and the selected accident chain with the smallest adjustment range can be selected from all the accident chains to be selected as the reference accident chain. In addition, the minimum number of adjustment parameters may be used as a target, and the selected accident chain with the minimum number of adjustment parameters may be selected from all the selected accident chains as a reference accident chain.
Through the mode, the reference accident chain can be selected by adjusting the amplitude or the number of the adjustment parameters, so that a user can quickly adjust the running control data to the running control data in the reference accident chain, the accident level is reduced, the situation that the adjustment amplitude is too large or the number of the adjustment parameters is too large and the user cannot finish in a short time is avoided, the reliability of the reduction of the accident level is further ensured, and the running safety of the user is ensured as much as possible.
And the candidate accident chain with the lowest accident level is preferentially selected as the candidate accident chain so as to be convenient for selecting the reference accident chain from the candidate accident chain, so that the accident level of the user can be reduced as much as possible, and the running safety of the user is further ensured.
After the reference accident chain is selected, further, the actions of related parts in the vehicle to be predicted can be controlled based on the running control data in the reference accident chain so as to control the running of the vehicle to be predicted.
For example, a control command may be generated based on the travel control data in the reference accident chain and sent to the vehicle system, so that the vehicle system controls the travel of the vehicle to be predicted based on the control command and targeting the travel control data in the reference accident chain.
Before the vehicle-mounted system controls the vehicle to run based on the received control instruction, the vehicle-mounted reminding can display the take-over prompt information through a human-computer interaction interface or a central control screen of the vehicle to be predicted, and meanwhile, the take-over prompt information can be played through a voice module of the vehicle to be predicted, so that a user can know that the vehicle is taken over at the moment, and the user does not need to manually control the vehicle any more.
Of course, after the running of the vehicle to be predicted is controlled based on the running control data, the current running information of the vehicle to be predicted can be obtained again to redetermine the accident prediction grade of the vehicle to be predicted, if the newly determined accident prediction grade is zero, the taking over of the cancelling information can be displayed through a human-computer interaction interface or a central control screen of the vehicle to be predicted, the running control of the vehicle to be predicted is stopped, and then the user can manually control the running of the vehicle.
The invention has the following technical effects: the method comprises the steps of acquiring current operation information consisting of vehicle body color combination data, environment data and running control data of a vehicle to be predicted, searching a target accident chain matched with the current operation information in all pre-stored accident chains, further determining the accident prediction grade of the vehicle to be predicted according to the accident grade corresponding to the target accident chain, if the accident grade is greater than a preset grade, searching a plurality of candidate accident chains matched with the vehicle body color combination data and the environment data in the current operation information in all pre-stored accident chains, further selecting a reference accident chain with the accident grade lower than the accident prediction grade from the candidate accident chains, and running control according to the running control data in the reference accident chain.
Fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present invention. As shown in fig. 3, electronic device 400 includes one or more processors 401 and memory 402.
The processor 401 may be a Central Processing Unit (CPU) or other form of processing unit having data processing capabilities and/or instruction execution capabilities and may control other components in the electronic device 400 to perform desired functions.
Memory 402 may include one or more computer program products that may include various forms of computer-readable storage media, such as volatile memory and/or non-volatile memory. The volatile memory may include, for example, random Access Memory (RAM) and/or cache memory (cache), and the like. The non-volatile memory may include, for example, read Only Memory (ROM), hard disk, flash memory, and the like. One or more computer program instructions may be stored on the computer readable storage medium that may be executed by the processor 401 to implement the vehicle body color based vehicle accident prediction method and/or other desired functions of any of the embodiments of the present invention described above. Various content such as initial arguments, thresholds, etc. may also be stored in the computer readable storage medium.
In one example, the electronic device 400 may further include: an input device 403 and an output device 404, which are interconnected by a bus system and/or other forms of connection mechanisms (not shown). The input device 403 may include, for example, a keyboard, a mouse, and the like. The output device 404 may output various information to the outside, including early warning prompt information, braking force, etc. The output device 404 may include, for example, a display, speakers, a printer, and a communication network and remote output devices connected thereto, etc.
Of course, only some of the components of the electronic device 400 that are relevant to the present invention are shown in fig. 3 for simplicity, components such as buses, input/output interfaces, etc. are omitted. In addition, electronic device 400 may include any other suitable components depending on the particular application.
In addition to the methods and apparatus described above, embodiments of the present invention may also be a computer program product comprising computer program instructions which, when executed by a processor, cause the processor to perform the steps of the vehicle body color-based vehicle accident prediction method provided by any of the embodiments of the present invention.
The computer program product may write program code for performing operations of embodiments of the present invention in any combination of one or more programming languages, including an object oriented programming language such as Java, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device, partly on a remote computing device, or entirely on the remote computing device or server.
Furthermore, embodiments of the present invention may also be a computer-readable storage medium, on which computer program instructions are stored, which, when being executed by a processor, cause the processor to perform the steps of the vehicle body color-based vehicle accident prediction method provided by any of the embodiments of the present invention.
The computer readable storage medium may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may include, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium would include the following: an electrical connection having one or more wires, a portable disk, a hard disk, random Access Memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of the present application. As used in this specification, the terms "a," "an," "the," and/or "the" are not intended to be limiting, but rather are to be construed as covering the singular and the plural, unless the context clearly dictates otherwise. The terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method or apparatus comprising such elements.
It should also be noted that the positional or positional relationship indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. are based on the positional or positional relationship shown in the drawings, are merely for convenience of describing the present invention and simplifying the description, and do not indicate or imply that the apparatus or element in question must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the present invention. Unless specifically stated or limited otherwise, the terms "mounted," "connected," and the like are to be construed broadly and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present invention will be understood in specific cases by those of ordinary skill in the art.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present invention.
Claims (10)
1. A vehicle accident prediction method based on a vehicle body color, characterized by comprising:
acquiring current running information of a vehicle to be predicted, wherein the current running information comprises vehicle body color combination data, environment data and running control data, and the vehicle body color combination data is used for describing the colors of the vehicle to be predicted and other interactive vehicles;
searching a target accident chain matched with the current operation information in all pre-stored accident chains, wherein the accident chain consists of vehicle body color combination data, environment data and driving control data;
determining an accident prediction grade of the vehicle to be predicted according to the accident grade corresponding to the target accident chain, and if the accident prediction grade is greater than a preset grade, searching a plurality of candidate accident chains matched with the vehicle body color combination data and the environment data in the current running information from all the pre-stored accident chains;
and selecting a reference accident chain with the accident level lower than the accident prediction level from all candidate accident chains, and controlling the vehicle to be predicted to run according to the running control data in the reference accident chain.
2. The method according to claim 1, wherein selecting a reference accident chain having an accident level lower than the accident prediction level from all candidate accident chains comprises:
selecting an accident chain to be selected from all candidate accident chains by taking the accident grade equal to the lowest grade as a target;
if the number of the accident chains to be selected is at least one, selecting a reference accident chain from all the accident chains to be selected, otherwise, selecting the accident chains to be selected from all the candidate accident chains again with the accident level higher than the lowest level and lower than the accident prediction level as the target.
3. The method of claim 2, further comprising, after selecting the candidate accident chain from among all candidate accident chains:
and judging whether the running control data in the current running information can be adjusted to the running control data in the accident chain to be selected in preset time aiming at each accident chain to be selected, and if not, rejecting the accident chain to be selected.
4. The method of claim 2, wherein selecting a reference accident chain from all candidate accident chains comprises:
determining the adjustment amplitude between the driving control data in each accident chain to be selected and the driving control data in the current operation information, and determining the accident chain to be selected with the minimum adjustment amplitude as a reference accident chain; or,
and determining the quantity of adjustment parameters between the driving control data in each accident chain to be selected and the driving control data in the current operation information, and determining the accident chain to be selected with the minimum quantity of the adjustment parameters as a reference accident chain.
5. The method according to claim 1, wherein the method further comprises:
acquiring each historical accident data, constructing a corresponding accident chain according to each historical accident data, and storing all accident chains and corresponding accident grades;
training to obtain an accident prediction model based on all accident chains and corresponding accident grades;
generating a plurality of new accident chains, determining the accident level corresponding to each new accident chain according to the accident prediction model, and storing all the new accident chains and the corresponding accident levels.
6. The method of claim 5, further comprising, prior to said training to derive an accident prediction model based on all accident chains and corresponding accident levels:
taking the accident level as an output parameter, and determining each input parameter according to each data in an accident chain;
based on all accident chains and corresponding accident levels, analyzing the correlation and collinearity between each input parameter and each output parameter, and determining each associated parameter in all the input parameters based on the analysis result;
and eliminating other input parameters except the associated parameters from all accident chains.
7. The method of claim 5, wherein the training to obtain the accident prediction model based on all accident chains and corresponding accident levels comprises:
constructing a neural network model, and inputting all accident chains into the neural network model;
and calculating a loss function according to the result output by the neural network model and the corresponding accident level, and adjusting network parameters in the neural network model according to the calculation result until the calculation result converges to obtain the accident prediction model.
8. The method of claim 5, wherein constructing a corresponding accident chain from each historical accident data comprises:
extracting, for each of the historical accident data, a body color chart number of the first and second collision vehicles, environmental data, and travel control data of the first and second collision vehicles from the historical accident data;
converting the vehicle body color card numbers of the first collision vehicle and the second collision vehicle into RGB channel values to obtain a first vehicle body color of the first collision vehicle and a second vehicle body color of the second collision vehicle;
determining body color combination data according to the first body color and the second body color;
and correlating the vehicle body color combination data, the environment data and the driving control data to obtain a corresponding accident chain.
9. An electronic device, the electronic device comprising:
a processor and a memory;
the processor is configured to execute the steps of the vehicle accident prediction method based on the vehicle body color according to any one of claims 1 to 8 by calling a program or instructions stored in the memory.
10. A computer-readable storage medium storing a program or instructions that cause a computer to execute the steps of the vehicle body color-based vehicle accident prediction method according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410003747.6A CN117494589B (en) | 2024-01-03 | 2024-01-03 | Vehicle accident prediction method, device and storage medium based on vehicle body color |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410003747.6A CN117494589B (en) | 2024-01-03 | 2024-01-03 | Vehicle accident prediction method, device and storage medium based on vehicle body color |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117494589A CN117494589A (en) | 2024-02-02 |
| CN117494589B true CN117494589B (en) | 2024-04-09 |
Family
ID=89683379
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410003747.6A Active CN117494589B (en) | 2024-01-03 | 2024-01-03 | Vehicle accident prediction method, device and storage medium based on vehicle body color |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117494589B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106203727A (en) * | 2016-07-19 | 2016-12-07 | 江苏大学 | A kind of road traffic accident chain risk real time evaluating method |
| CN107430006A (en) * | 2014-12-02 | 2017-12-01 | 凯文·孙林·王 | Avoid the method and system of accident |
| CN108154681A (en) * | 2016-12-06 | 2018-06-12 | 杭州海康威视数字技术股份有限公司 | Risk Forecast Method, the apparatus and system of traffic accident occurs |
| CN114312763A (en) * | 2022-01-29 | 2022-04-12 | 中国第一汽车股份有限公司 | Vehicle control method, system, equipment and storage medium |
| CN117022323A (en) * | 2023-08-22 | 2023-11-10 | 南京项尚车联网技术有限公司 | Intelligent driving vehicle behavior analysis and prediction system and method |
| CN117271625A (en) * | 2023-09-28 | 2023-12-22 | 杭州海樊科技有限公司 | Risk data display method and system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR102279754B1 (en) * | 2020-02-10 | 2021-07-20 | 주식회사 서울로보틱스 | Method, apparatus, server, and computer program for preventing crash accident |
-
2024
- 2024-01-03 CN CN202410003747.6A patent/CN117494589B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107430006A (en) * | 2014-12-02 | 2017-12-01 | 凯文·孙林·王 | Avoid the method and system of accident |
| CN106203727A (en) * | 2016-07-19 | 2016-12-07 | 江苏大学 | A kind of road traffic accident chain risk real time evaluating method |
| CN108154681A (en) * | 2016-12-06 | 2018-06-12 | 杭州海康威视数字技术股份有限公司 | Risk Forecast Method, the apparatus and system of traffic accident occurs |
| CN114312763A (en) * | 2022-01-29 | 2022-04-12 | 中国第一汽车股份有限公司 | Vehicle control method, system, equipment and storage medium |
| CN117022323A (en) * | 2023-08-22 | 2023-11-10 | 南京项尚车联网技术有限公司 | Intelligent driving vehicle behavior analysis and prediction system and method |
| CN117271625A (en) * | 2023-09-28 | 2023-12-22 | 杭州海樊科技有限公司 | Risk data display method and system |
Non-Patent Citations (1)
| Title |
|---|
| 车身颜色与事故风险度相关性模型研究;许卉莹和裘晨璐;《中国公共安全(学术版)》;20150630(第2期);第59-63页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117494589A (en) | 2024-02-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10606276B2 (en) | User data-based autonomous vehicle system | |
| US11900738B2 (en) | Systems and methods to obtain feedback in response to autonomous vehicle failure events | |
| US10229461B2 (en) | Continuous identity monitoring for classifying driving data for driving performance analysis | |
| JP2020524632A (en) | System and method for obtaining occupant feedback in response to an autonomous vehicle driving event | |
| CN111989725B (en) | Information processing method and information processing system | |
| CN104859662A (en) | Fault handling in an autonomous vehicle | |
| US20200134729A1 (en) | Information processing device, information processing system, information processing method, and program | |
| CN112041201B (en) | Method, system, and medium for controlling access to vehicle features | |
| CN112937520A (en) | Emergency braking method and device for vehicle, commercial vehicle and storage medium | |
| CN111613060B (en) | Data processing method and equipment | |
| CN114049677A (en) | Vehicle ADAS control method and system based on emotion index of driver | |
| CN113353083B (en) | Vehicle behavior recognition method | |
| CN112319505B (en) | Automatic driving lane change judgment method, device, equipment and storage medium | |
| CN112172829B (en) | Lane departure warning method and device, electronic equipment and storage medium | |
| CN117494589B (en) | Vehicle accident prediction method, device and storage medium based on vehicle body color | |
| CN116665155B (en) | Underground trackless man-vehicle getting-off early warning safety management system based on image processing technology | |
| CN115720555A (en) | Method and system for improving user alertness in an autonomous vehicle | |
| CN115169996B (en) | Road risk determination method, apparatus and storage medium | |
| CN113408499B (en) | Joint evaluation method and device of dual-network model and storage medium | |
| CN117104255B (en) | Intelligent perception human-vehicle interaction system and method for intelligent driving vehicle | |
| CN111341153B (en) | Early warning method, device, server and medium for travel driving | |
| US20230382389A1 (en) | Information collection control device and method for vehicle | |
| CN116957344B (en) | Safety evaluation method, device, equipment and medium for automatic emergency braking system | |
| CN113844456B (en) | ADAS automatic opening method and device | |
| CN116564038A (en) | Early warning method and device based on voice and image recognition, and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |