A vehicle driving control method, system and storage medium based on intelligent networking

Abstract

The present invention discloses a method, system, and storage medium for controlling vehicle driving based on intelligent networking, relating to the field of driving control technology. The method comprises the following specific steps: Step 1: Acquire data, historical data, and accident outcome data of multiple vehicles involved in traffic accidents at the time of the accident; Step 2: Construct a safety prediction model based on the data, historical data, and accident outcome data of multiple vehicles involved in traffic accidents at the time of the accident; Step 3: Acquire historical data and driving data of the vehicle to be controlled, and obtain safe driving parameters using a risk factor dataset and the safety prediction model; Step 4: Perform driving control on the vehicle to be controlled based on the safe driving parameters. The present invention utilizes traffic accident data to screen risk factors and uses the risk factors to determine safety parameters, providing safe and reliable reference data for driving control, thereby improving the driving safety of intelligent vehicles.

Description

Technical Field

[0001] The present invention relates to the field of driving control technology, and more particularly to a method, system and storage medium for controlling automobile driving based on intelligent networking. Background Art

[0002] Intelligent connected vehicles refer to the implementation of network technology and modern communication technology in the control, sensing and execution mechanisms of vehicles, thereby enabling the vehicle to share information resources with the background and people, and ultimately forming a modern new-energy vehicle with environmental perception, intelligent control, intelligent execution and automated decision-making. However, safety accidents are emerging in an endless stream in existing intelligent connected vehicles. Therefore, how to ensure the safety of vehicle driving control is an urgent problem that technicians in this field need to solve. Summary of the Invention

[0003] In view of this, the present invention provides a vehicle driving control method, system and storage medium based on intelligent networking, which overcome the above-mentioned defects.

[0004] In order to achieve the above object, the present invention provides the following technical solutions:

[0005] A method for controlling vehicle driving based on intelligent networking, comprising the following specific steps:

[0006] Step 1: Obtain data on the time of occurrence, historical data, and accident result data of multiple traffic accident vehicles;

[0007] Step 2: Build a safety prediction model based on the data of multiple traffic accident vehicles at the time of the accident, historical data, and accident result data;

[0008] Step 3: Obtain historical data and driving data of the vehicle to be controlled, and use the risk factor dataset and safety prediction model to obtain safe driving parameters;

[0009] Step 4: Control the vehicle to be controlled according to the safe driving parameters.

[0010] Optionally, the steps for obtaining the safety assessment model are:

[0011] Step 21: extracting characteristic parameters based on the data of the multiple traffic accident vehicles at the time of the accident, historical data, and accident result data, and constructing a characteristic parameter set;

[0012] Step 22: Use the risk matrix method and Borda ordinal value method to select risk factors from the characteristic parameter set to construct a risk factor data set;

[0013] Step 23: Input the risk factors in the risk factor data set into the neural network for iterative training to generate a safety prediction model.

[0014] Optionally, the characteristic parameters include driving data, vehicle data and environmental data, which come from the vehicle driving log data of the vehicle involved in the traffic accident.

[0015] Optionally, the risk matrix method is constructed based on risk loss level and risk probability level.

[0016] Optionally, the expression of Borda ordinal value method is:

[0017] ;

[0018] Where, N is the number of risk factors; i For the i risk factors; k For the preset rules, n The number of preset rules.

[0019] Optionally, the steps for obtaining safe driving parameters are:

[0020] Step 31: Obtain historical data and driving data of the vehicle to be controlled;

[0021] Step 32: Based on the historical data and driving data of the vehicle to be controlled and expert experience, the risk factor dataset is used to estimate the safe driving parameters of the vehicle to be controlled in different scenarios;

[0022] Step 33: Correct the estimated safe driving parameters using a safety estimation model;

[0023] Step 34: Set the safe driving parameters of the vehicle to be controlled according to the corrected estimated safe driving parameters.

[0024] Optionally, the specific steps of driving control are:

[0025] Step 41: The controlled vehicle drives based on the safe driving parameters and obtains information about adjacent vehicles in the protection identification zone of the driving vehicle and driving status information of the left and right lanes in real time;

[0026] Step 42: Determine whether to change lanes, slow down, or stop to avoid an obstacle based on the real-time data obtained.

[0027] Optionally, after making a driving strategy based on real-time data in step 42, the driving strategy is sent to a safety prediction model for risk assessment, and the driving strategy is corrected based on the assessment results.

[0028] An automobile driving control system based on intelligent networking, comprising:

[0029] Training data acquisition module: used to obtain data on the time of occurrence of multiple traffic accidents, historical data and accident result data;

[0030] A safety prediction model building module is used to build a safety prediction model based on the data of multiple traffic accident vehicles at the time of the accident, historical data, and accident result data;

[0031] A safe driving parameter estimation module is used to obtain historical data and driving data of the vehicle to be controlled, and obtain safe driving parameters based on the historical data and driving data of the vehicle to be controlled using a risk factor dataset and a safety estimation model;

[0032] The driving control module is used to control the driving of the vehicle to be controlled according to safe driving parameters.

[0033] A storage medium stores a computer program, wherein the computer program is configured to execute the above-mentioned vehicle driving control method based on intelligent networking when running.

[0034] It can be seen from the above technical solution that compared with the existing technology, the present invention provides a vehicle driving control method, system and storage medium based on intelligent networking, which uses traffic accident data to screen risk factors and uses risk factors to determine safety parameters, providing safe and reliable reference data for driving control, thereby improving the driving safety of smart cars. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0036] Figure 1 Schematic diagram of the method flow of the present invention;

[0037] Figure 2 A schematic diagram of a method flow chart of a safety prediction model of the present invention;

[0038] Figure 3 This is a schematic diagram of the driving control process of the present invention;

[0039] Figure 4 A schematic diagram of the position of the protection identification area of ​​the present invention;

[0040] Figure 5 Schematic diagram of the system structure of the present invention. DETAILED DESCRIPTION

[0041] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0042] The embodiment of the present invention discloses a vehicle driving control method based on intelligent networking, such as Figure 1 As shown, the specific steps are:

[0043] Step 1: Obtain data on the time of occurrence of multiple traffic accidents, historical data, and accident result data; wherein the data on the time of occurrence of the accident includes the vehicle's motion information, posture information, and driving data; the historical data is the vehicle driving log data within a preset time period, and in this embodiment, the vehicle driving log data of the previous quarter is used; the vehicle driving log data is a data representation that records all movement processes from starting to stopping during the driving process of the vehicle, including the lateral / longitudinal movement of the main vehicle, the working status of each vehicle system, the position and angle of other vehicles from the vehicle, etc.

[0044] Step 2: Build a safety prediction model based on the data of multiple traffic accident vehicles at the time of the accident, historical data, and accident result data, such as Figure 2 As shown, the specific steps are:

[0045] Step 21: extracting characteristic parameters based on the data of the multiple traffic accident vehicles at the time of the accident, historical data, and accident result data, and constructing a characteristic parameter set;

[0046] Step 22: Use the risk matrix method and Borda ordinal value method to select risk factors from the characteristic parameter set to construct a risk factor data set;

[0047] Step 23: Input the risk factors in the risk factor data set into the neural network for iterative training to generate a safety prediction model.

[0048] In this embodiment, in step 21, the traffic accident data is cleaned before being extracted to remove duplicate and incomplete data. The extracted feature parameters include driving data, vehicle data, and environmental data.

[0049] Among them, driving data includes driver driving data and unmanned driving data; among them, driver driving data includes the driver's health status, driving experience, driving habits, safety awareness, etc.; unmanned driving data includes historical driving data; vehicle data includes driving data and vehicle condition data, among which driving data includes speed, acceleration, position, traffic flow, etc.; vehicle condition data includes model, health status; environmental data includes fixed environmental data and changing environmental data, among which fixed environmental data includes road condition data, speed limit data, etc., and changing environmental data includes environmental data outside the vehicle, such as temperature data, weather data, traffic flow data and neighboring vehicle data, and environmental data inside the vehicle, such as temperature, humidity and sound inside the vehicle, etc.

[0050] The extracted characteristic parameters in this embodiment also include the cause of the accident and the loss data of the accident extracted from the accident investigation results.

[0051] In step 22, the parameters in the obtained characteristic parameter set are used to construct a risk assessment matrix using the risk matrix method, and then the risk factors are sorted using the Borda ordinal value method. According to the preset threshold, a fixed number of risk factors are intercepted to construct a risk factor data set.

[0052] The risk matrix method comprehensively analyzes, weights, and classifies risk loss levels and risk probability levels to derive a risk index. A matrix is ​​constructed based on the risk index. This risk assessment matrix is ​​a semi-quantitative analysis tool that uses the frequency of risk occurrence to characterize the likelihood. The frequency of risk occurrence is statistically obtained from a traffic accident database. Based on this statistical data, the likelihood of an accident is divided into five levels: ultra-high risk, high risk, medium risk, low risk, and ultra-low risk, as shown in Table 1.

[0053] Table 1 Risk level classification standards

[0054]

[0055] Comprehensively analyze, weight and classify risk loss levels and risk levels to derive a risk index R , whose expression is:

[0056] R = F C ;

[0057] Where, F is the risk level; C The risk loss level.

[0058] The risk loss is classified into five levels according to the severity of the consequences of traffic safety accidents, namely: extremely serious loss, serious loss, moderate loss, slight loss and tiny loss, as shown in Table 2.

[0059] Table 2 Risk loss level classification standards

[0060]

[0061] The specific steps of the Borda ordinal value method are as follows: calculate the risk probability ranking value based on the probability level of risk occurrence and the number of risk factors appearing in the risk level; calculate the risk loss level ranking value based on the severity level of risk occurrence and the number of risk factors appearing in the severity level; and obtain the ranking index of the risk factor based on the risk probability ranking value and the risk loss level ranking value. The expression is:

[0062] ;

[0063] Where, N is the number of risk factors; i For the i risk factors; k For the preset rules, n is the number of preset rules. In this embodiment, n Take 2, where 1 is the risk probability ranking value; 2 is the risk loss level ranking value.

[0064] The risk factors are sorted according to their C values, and a fixed number of risk factors are intercepted according to the preset threshold to construct a risk factor data set.

[0065] In step 23, based on the C value of each risk factor in the risk factor data set, the hierarchical analysis method is used to calculate the initial weight of each risk factor. Specifically, according to the Borda order value of each risk factor, a judgment matrix of any two risk factors is constructed to calculate the weight of any risk factor, that is, the initial weight of the risk factor. Each risk factor and its corresponding weight as well as the cause of the accident and the loss data of the accident are input into the neural network for iterative training until the preset target is reached, thereby obtaining a safety prediction model.

[0066] Step 3: Obtain historical data and driving data of the vehicle to be controlled, and use the risk factor dataset and safety prediction model to obtain safe driving parameters, specifically:

[0067] Based on the historical data and driving data of the vehicle to be controlled, the risk factor dataset is used according to expert experience to estimate the safe driving parameters of the vehicle to be controlled in different scenarios. Then, the safety estimation model is used to predict the risks of the safe driving parameters. Based on the prediction results, the predicted safe driving parameters in different scenarios are adjusted to make them meet the safety standards. Then, 80% of the estimated safe driving parameters that meet the safety standards are used as the final safe driving parameters.

[0068] Safe driving parameters include, but are not limited to, one or more safe driving values ​​including the vehicle's maximum speed, maximum steering wheel angle, and current minimum following distance. The safe driving parameters of the vehicle to be controlled are estimated in different scenarios using a risk factor dataset based on expert experience. Specifically, expert experience is first used to determine a basic driving value for any scenario (e.g., poor weather, poor road conditions, insufficient driving experience, etc.). The basic driving value is then adjusted based on the weight of each risk factor in the risk factor dataset and the historical data and driving data of the vehicle to be controlled. For example, if visibility is poor due to weather conditions as a risk factor, the vehicle's maximum speed and current minimum following distance may be limited based on safe driving conditions. If driving experience is insufficient as a risk factor, the vehicle's maximum speed and current minimum following distance may be limited based on safe driving conditions. Multiple safe driving values ​​are obtained based on each risk factor, and constraints are set using historical experience to determine the estimated safe driving parameters. The constraints are: the vehicle's maximum speed is the minimum value among multiple maximum speeds of motor vehicles, and the current minimum following distance is the maximum value among multiple maximum speeds of motor vehicles.

[0069] Step 4: Control the vehicle to be controlled according to the safe driving parameters. Figure 3 As shown, specifically:

[0070] Step 41: The controlled vehicle drives based on the safe driving parameters and collects real-time data of the driving vehicle, information about adjacent vehicles in the protection identification zone, and driving status information of the left and right lanes in real time;

[0071] Step 42: When the lane does not meet the safe driving parameters or an obstacle appears, determine whether to change lanes, slow down, or stop to avoid it based on the collected data.

[0072] In step 41, if Figure 4As shown, the protection identification zone divides the area within a preset range around the vehicle (set according to needs) into eight directions according to the vehicle's driving direction, namely front (fourth protection identification zone), rear (fifth protection identification zone), left (second protection identification zone), right (seventh protection identification zone), left front (third protection identification zone), right front (eighth protection identification zone), left rear (first protection identification zone) and right rear (sixth protection identification zone), which are determined as eight adjacent zones. Different numbers of adjacent zones are selected as protection identification zones according to the driving conditions and driving conditions of the lane, and there are at least two protection identification zones.

[0073] The selection of protection identification zones is to first determine the importance level of each protection identification zone based on the Road Traffic Safety Law and historical traffic accident data, and then select the protection identification zones for real-time monitoring according to the traffic conditions and driving conditions of the driving lanes according to the importance level. For example, in a congested urban area, you can choose to collect data from eight protection identification zones. If you are on a smooth highway, you can choose to collect data from the front, rear, left and right protection identification zones, or you can choose to collect data from the front, rear and left protection identification zones, or even fewer.

[0074] The real-time data of the vehicle and the information of the adjacent vehicles include: the vehicle's speed, acceleration, angular velocity, geographic location, yaw angle, pitch angle and roll angle, etc.

[0075] The real-time data of the driving vehicle, information about adjacent vehicles in the protection identification zone, and driving status information of the left and right lanes are all obtained by the on-board unit. By extracting multiple data streams from the on-board unit, the real-time data of the driving vehicle, information about adjacent vehicles in the protection identification zone, and driving status information of the left and right lanes are obtained.

[0076] In step 42, when the on-board unit detects that an obstacle appears in the driving direction or that the driving time in the adjacent lane is shorter than the driving time in the current lane, it first determines whether the left and right lanes are available (i.e., whether they are lanes with the same driving direction) based on the driving status information of the left and right lanes. If available, it determines whether there are adjacent vehicles in the left and right protection identification zones. If not, it determines based on the left rear or right rear protection identification zones that no adjacent vehicles have passed within a fixed time period. If not, it changes lanes. If the obstacle ahead is on the vehicle's route, but the vehicle is driving slowly, the left and right lanes are unavailable, or there are adjacent vehicles in the left and right protection identification zones, it decelerates to avoid the obstacle. If there is a fixed obstacle ahead on the vehicle's route, the left and right lanes are unavailable, or there are adjacent vehicles in the left and right protection identification zones, it stops to avoid the obstacle.

[0077] After the driving strategy is determined in step 42, the safety prediction model performs a risk assessment. When the vehicle's safety risk reaches different levels, the onboard unit (OVU) can issue corresponding warnings based on the real-time prediction of the intelligent connected vehicle's behavior. The OVU also provides different response strategies for different levels of danger. Within an acceptable risk range, the OVU automatically controls the vehicle's state or prompts the driver on how to operate based on the predicted intelligent connected vehicle behavior. Conversely, if the risk exceeds the acceptable range, the OVU issues an alarm and enforces mandatory operation.

[0078] The present invention also includes a vehicle driving control system based on intelligent networking, such as Figure 5 As shown, including:

[0079] Training data acquisition module: used to obtain data on the time of occurrence of multiple traffic accidents, historical data and accident result data;

[0080] A safety prediction model building module is used to build a safety prediction model based on the data of multiple traffic accident vehicles at the time of the accident, historical data, and accident result data;

[0081] A safe driving parameter estimation module is used to obtain historical data and driving data of the vehicle to be controlled, and obtain safe driving parameters based on the historical data and driving data of the vehicle to be controlled using a risk factor dataset and a safety estimation model;

[0082] The driving control module is used to control the driving of the vehicle to be controlled according to safe driving parameters.

[0083] A storage medium, characterized in that a computer program is stored in the storage medium, wherein the computer program is configured to execute the above-mentioned vehicle driving control method based on intelligent networking when running.

[0084] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Reference can be made to the common and similar parts between the various embodiments. For the devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method description.

[0085] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A vehicle driving control method based on intelligent networking, characterized in that: The specific steps are: Step 1: Obtain data on the time of occurrence, historical data, and accident result data of multiple traffic accident vehicles; Step 2: Build a safety prediction model based on the data of multiple traffic accident vehicles at the time of the accident, historical data, and accident result data; Step 3: Obtain historical data and driving data of the vehicle to be controlled, and use the risk factor dataset and safety prediction model to obtain safe driving parameters; Step 4: Control the vehicle to be controlled according to the safe driving parameters; The steps to obtain the safety assessment model are: Step 21: extracting characteristic parameters based on the data of the multiple traffic accident vehicles at the time of the accident, historical data, and accident result data, and constructing a characteristic parameter set; Step 22: Use the risk matrix method and Borda ordinal value method to select risk factors from the characteristic parameter set to construct a risk factor data set; Step 23: Input the risk factors in the risk factor data set into the neural network for iterative training to generate a safety prediction model; The steps to obtain safe driving parameters are: Step 31: Obtain historical data and driving data of the vehicle to be controlled; Step 32: Based on the historical data and driving data of the vehicle to be controlled and expert experience, the risk factor dataset is used to estimate the safe driving parameters of the vehicle to be controlled in different scenarios; Step 33: Correct the estimated safe driving parameters using a safety estimation model; Step 34: setting safe driving parameters of the vehicle to be controlled according to the corrected estimated safe driving parameters; The specific steps of driving control are: Step 41: The controlled vehicle drives based on the safe driving parameters and obtains information about adjacent vehicles in the protection identification zone of the driving vehicle and driving status information of the left and right lanes in real time; Step 42: Determine whether to change lanes, slow down, or stop to avoid an obstacle based on the real-time data obtained.

2. The method for controlling vehicle driving based on intelligent networking according to claim 1, characterized in that: The characteristic parameters include driving data, vehicle data and environmental data, which come from the vehicle driving log data of the vehicle involved in the traffic accident.

3. The method for controlling vehicle driving based on intelligent networking according to claim 1, characterized in that: The risk matrix method is constructed based on risk loss level and risk probability level.

4. The method for controlling vehicle driving based on intelligent networking according to claim 1, characterized in that: The expression of Borda ordinal value method is: ； Where, N is the number of risk factors; i For the i risk factors; k For the preset rules, n is the number of preset rules, For the i Risk factors in the preset rules k The sort value below.

5. The method for controlling vehicle driving based on intelligent networking according to claim 1, characterized in that: After the driving strategy is made based on the real-time data in step 42, the driving strategy is sent to the safety prediction model for risk assessment, and the driving strategy is corrected based on the assessment results.

6. An automobile driving control system based on intelligent networking, characterized in that: include: Training data acquisition module: used to obtain data on the time of occurrence of multiple traffic accidents, historical data and accident result data; A safety prediction model building module is used to build a safety prediction model based on the data of multiple traffic accident vehicles at the time of the accident, historical data, and accident result data; A safe driving parameter estimation module is used to obtain historical data and driving data of the vehicle to be controlled, and obtain safe driving parameters based on the historical data and driving data of the vehicle to be controlled using a risk factor dataset and a safety estimation model; A driving control module, used to control the driving of the controlled vehicle according to safe driving parameters; The steps to obtain the safety assessment model are: Extract characteristic parameters based on the data of multiple traffic accident vehicles at the time of the accident, historical data and accident result data, and construct a characteristic parameter set; The risk matrix method and Borda ordinal value method are used to screen risk factors from the characteristic parameter set to construct a risk factor data set; Input the risk factors in the risk factor dataset into the neural network for iterative training to generate a safety prediction model; The steps to obtain safe driving parameters are: Obtain historical data and driving data of the vehicle to be controlled; Based on the historical data and driving data of the vehicle to be controlled, the risk factor dataset is used according to expert experience to estimate the safe driving parameters of the vehicle to be controlled in different scenarios; Correcting the estimated safe driving parameters using a safety prediction model; setting safe driving parameters of the vehicle to be controlled according to the corrected estimated safe driving parameters; The specific steps of driving control are: The vehicle to be controlled drives based on safe driving parameters and obtains information about adjacent vehicles in the protection identification zone of the driving vehicle and the driving status information of the left and right lanes in real time; Determine whether to change lanes, slow down, or stop to avoid a maneuver based on the real-time data obtained.

7. A storage medium, characterized in that: The storage medium stores a computer program, wherein the computer program is configured to execute the method according to any one of claims 1 to 5 when executed.

Images