Potential collision risk prediction method, device and computer equipment

Abstract

The present application provides a method, apparatus and computer equipment for predicting potential collision risks. In various Internet of Vehicles applications, after obtaining a first prediction value and a second prediction value of the potential collision risk between a first vehicle and a second vehicle at the current moment, a plurality of first historical prediction values ​​and a plurality of second historical prediction values ​​of the potential collision risks that the two vehicles have encountered are combined, and a difference operation is performed on the potential collision risk between the two vehicles using a normal distribution function to obtain a difference prediction range, i.e., a correction value range, that is, a difference prediction range that is less than the historical traffic accident rate of the section. Thereafter, based on the difference prediction value, i.e., the correction value, included in the difference prediction range, the potential collision risk prediction value of the corresponding vehicle at the current moment is calibrated to obtain a more accurate and reliable target prediction value, thereby solving the problem that the potential collision risk for one vehicle may be too conservative and the potential collision risk for another vehicle may be too aggressive, resulting in inaccurate collision warning.

Description

Technical Field

[0001] The present application relates to the field of safe driving applications, and specifically to a potential collision risk prediction method, device and computer equipment. Background Art

[0002] At present, the vehicle safety assisted driving system is an important part of the current international intelligent transportation system research. It uses artificial intelligence (AI) technologies such as machine vision to realize real-time notification of the driver's surrounding environment status, and when the vehicle may be in danger of potential collision, it promptly warns the driver in advance to take effective response measures, eliminate potential accident hazards, improve vehicle driving safety, and effectively reduce the incidence of vehicle accidents.

[0003] Among them, for the potential collision risk between vehicles, due to certain errors in the detection, transmission and processing of vehicle information, the potential collision intensity between two vehicles with potential collision risk may be different, and there may even be a large difference, resulting in the potential collision risk prediction for one vehicle being too conservative, and failing to warn the driver of the vehicle in time to avoid a collision accident. The potential collision risk prediction for another vehicle is too aggressive, causing the driver of the vehicle to take measures based on the warning prompt even when there is no collision, affecting the normal driving of the vehicle.

[0004] It can be seen that how to reliably and accurately predict the potential collision risk of vehicles has become one of the important research topics in the field of vehicle safety assisted driving and vehicle-road collaboration. Summary of the invention

[0005] In view of this, in order to achieve the above purpose, the embodiment of the present application provides the following technical solutions:

[0006] On the one hand, the present application proposes a potential collision risk prediction method, the method comprising:

[0007] Obtaining a first prediction value of a potential collision risk of a first vehicle with a second vehicle at a current moment, a second prediction value of a potential collision risk of a second vehicle with the first vehicle, and a historical traffic accident rate corresponding to a road section where the first vehicle and the second vehicle are located;

[0008] Retrieving a plurality of first historical prediction values ​​of the potential collision risk corresponding to the first vehicle and a plurality of second historical prediction values ​​of the potential collision risk corresponding to the second vehicle acquired at different times;

[0009] Based on the first prediction value, the plurality of first historical prediction values, the second prediction value, and the plurality of second historical prediction values, a potential collision risk difference calculation between the first vehicle and the second vehicle is performed using a normal distribution function to obtain a difference prediction range that is less than the historical traffic accident rate, and a mean value of the normal distribution function is zero;

[0010] Based on the difference prediction value within the difference prediction range, the first prediction value and the second prediction value are calibrated respectively, and a first target prediction value of the potential collision risk of the first vehicle with the second vehicle and a second target prediction value of the potential collision risk of the second vehicle with the first vehicle are output at the current moment;

[0011] The first target prediction value and the second target prediction value are used to implement collision warning for corresponding vehicles.

[0012] Optionally, the calculation of the potential collision risk difference between the first vehicle and the second vehicle using a normal distribution function based on the first predicted value, the multiple first historical predicted values, the second predicted value, and the multiple second historical predicted values ​​to obtain a difference prediction range that is less than the historical traffic accident rate includes:

[0013] Obtaining a difference variable of a potential collision risk between the first vehicle and the second vehicle at a current moment based on the first prediction value, the plurality of first historical prediction values, the second prediction value, and the plurality of second historical prediction values;

[0014] The difference variable is operated using a normal distribution function to obtain a difference prediction range of the potential collision risk between the first vehicle and the second vehicle that is less than the historical traffic accident rate at the current moment.

[0015] Optionally, obtaining a difference variable of a potential collision risk between the first vehicle and the second vehicle at a current moment based on the first prediction value, the plurality of first historical prediction values, the second prediction value, and the plurality of second historical prediction values ​​includes:

[0016] Based on the first prediction value and the plurality of first historical prediction values, obtaining a first risk variable of the first vehicle at a current moment;

[0017] Based on the second predicted value and the plurality of second historical predicted values, obtaining a second risk variable of the second vehicle at a current moment;

[0018] The first risk variable and the second risk variable are summed to obtain a difference variable of the potential collision risk between the first vehicle and the second vehicle at the current moment.

[0019] Optionally, the first risk variable and the second risk variable are both sample variances.

[0020] Optionally, the first prediction value and the second prediction value are calibrated respectively based on the difference prediction value within the difference prediction range, and a first target prediction value of the potential collision risk of the first vehicle to the second vehicle and a second target prediction value of the potential collision risk of the second vehicle to the first vehicle are output at the current moment, including:

[0021] Select any first difference prediction value and any second difference prediction value from the difference prediction range;

[0022] Summing the first difference prediction value and the first prediction value to obtain a first target prediction value of the potential collision risk of the first vehicle to the second vehicle at the current moment;

[0023] The second difference prediction value and the second prediction value are summed to obtain a second target prediction value of the potential collision risk of the second vehicle with the first vehicle at the current moment.

[0024] Optionally, the obtaining of a first prediction value of a potential collision risk of the first vehicle with the second vehicle and a second prediction value of a potential collision risk of the second vehicle with the first vehicle at a current moment includes:

[0025] Acquire vehicle information of the first vehicle and the second vehicle corresponding to the current moment, wherein the vehicle information includes driving speed, geographic location, vehicle type, and vehicle configuration attribute information corresponding to the vehicle type;

[0026] Based on the vehicle information corresponding to the current moment, a first prediction value of the potential collision risk of the first vehicle with the second vehicle and a second prediction value of the potential collision risk of the second vehicle with the first vehicle are obtained.

[0027] Optionally, the process of obtaining the predicted value of the potential collision risk between two vehicles includes any of the following implementations:

[0028] Obtain the predicted value of the potential collision risk between two vehicles at each moment;

[0029] In response to a prediction request for a potential collision risk of two target vehicles, obtaining a predicted value of the potential collision risk between the two target vehicles;

[0030] Based on the acquired vehicle information, it is determined that the target vehicle has entered a road section with a high incidence of collision accidents, and a predicted value of the potential collision risk between the target vehicle and its adjacent vehicles is obtained. The road section with a high incidence of collision accidents refers to a traffic geographical range where the historical traffic accident rate is greater than the accident threshold.

[0031] Optionally, the method further includes:

[0032] Get the vehicle collision threshold;

[0033] detecting that the first target prediction value is greater than the vehicle collision threshold, and outputting first collision prompt information for the first vehicle;

[0034] It is detected that the second target prediction value is greater than the vehicle collision threshold, and second collision prompt information for the second vehicle is output.

[0035] On the other hand, the present application also proposes a potential collision risk prediction device, the device comprising:

[0036] An information acquisition module, used to acquire a first prediction value of a potential collision risk of a first vehicle with a second vehicle at a current moment, a second prediction value of a potential collision risk of a second vehicle with the first vehicle, and a historical traffic accident rate corresponding to a road section where the first vehicle and the second vehicle are located;

[0037] a historical prediction value retrieval module, used to retrieve a plurality of first historical prediction values ​​of the potential collision risk corresponding to the first vehicle obtained at different times within a preset historical time period from the current time, and a plurality of second historical prediction values ​​of the potential collision risk corresponding to the second vehicle;

[0038] a difference prediction range obtaining module, configured to perform a potential collision risk difference calculation between the first vehicle and the second vehicle using a normal distribution function based on the first prediction value, the plurality of first historical prediction values, the second prediction value, and the plurality of second historical prediction values, to obtain a difference prediction range that is less than the historical traffic accident rate, wherein the mean of the normal distribution function is zero;

[0039] a prediction value calibration module, configured to calibrate the first prediction value and the second prediction value respectively based on the difference prediction value within the difference prediction range, and output a first target prediction value of the potential collision risk of the first vehicle with the second vehicle and a second target prediction value of the potential collision risk of the second vehicle with the first vehicle at a current moment;

[0040] The first target prediction value and the second target prediction value are used to implement collision warning for corresponding vehicles.

[0041] On the other hand, the present application further proposes a computer device, characterized in that the computer device comprises:

[0042] Communication interface;

[0043] A memory, used to store a program for implementing the potential collision risk prediction method as described above;

[0044] The processor is used to load and execute the program stored in the memory to implement the various steps of the potential collision risk prediction method as described above.

[0045] It can be seen that the present application provides a method, device and computer equipment for predicting potential collision risks. For a first vehicle and a second vehicle that are close to each other, after obtaining a first prediction value of the potential collision risk of the first vehicle to the second vehicle at the current moment, and a second prediction value of the potential collision risk of the second vehicle to the first vehicle, the present application does not directly issue a collision warning to the vehicle driver based on this, but takes into account the difference in potential collision risks between the first vehicle and the second vehicle, and the law of change of the difference, and proposes to calibrate the potential collision risk prediction values ​​corresponding to each of the two vehicles, improve the accuracy and reliability of each potential collision risk prediction value, so that the calibrated potential collision risks between the two vehicles are relatively symmetrical (that is, the difference is within a reasonable range), which solves the technical problem that the prior art may be too conservative for the potential collision risk of one vehicle and too aggressive for the potential collision risk of another vehicle, resulting in inaccurate collision warnings for the corresponding vehicles, thereby improving vehicle driving safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0047] Figure 1 A schematic diagram of an optional application scenario for implementing the potential collision risk prediction method proposed in this application is shown;

[0048] Figure 2 A schematic diagram showing an optional example of a process flow of a potential collision risk prediction method proposed in the present application;

[0049] Figure 3 A schematic flow chart showing another optional example of the potential collision risk prediction method proposed in the present application;

[0050] Figure 4 A schematic flow chart showing another optional example of the potential collision risk prediction method proposed in the present application;

[0051] Figure 5 A schematic diagram of an application scenario applicable to the potential collision risk prediction method proposed in this application is shown;

[0052] Figure 6Shows the application Figure 5 A scenario flow chart of an application scenario of the potential collision risk prediction method proposed in the present application is shown;

[0053] Figure 7 A schematic structural diagram of an optional example of a potential collision risk prediction device proposed in the present application is shown;

[0054] Figure 8 A schematic structural diagram of another optional example of the potential collision risk prediction device proposed in the present application is shown;

[0055] Fig. 9 A schematic diagram of the hardware structure of an optional example of a computer device suitable for the potential collision risk prediction method proposed in the present application is shown. DETAILED DESCRIPTION

[0056] In response to the issues raised in the background technology section, it was discovered during the research process that when the potential collision intensity of vehicle A against vehicle B is not equal to the potential collision intensity of vehicle B against vehicle A, the existing technology does not correct the potential collision intensity between vehicles, but directly compares the calculated potential collision intensity with the intensity threshold to decide whether to warn the driver of the corresponding vehicle, which leads to the problem that the potential collision risk for one vehicle may be too conservative, while the potential collision risk for another vehicle may be too aggressive.

[0057] Moreover, as mentioned in the background technology section, errors that may be caused by measurement methods, delays or other factors during the detection, transmission and processing of vehicle information are difficult to measure accurately, and it is impossible to use the traditional data correction concept, that is, to use the measured errors to accurately correct the potential collision risks between vehicles to improve the prediction accuracy of the potential collision risks of vehicles.

[0058] In order to improve the above technical problems, improve the prediction accuracy of the potential collision risk of vehicles, ensure timely, reliable and accurate warning of vehicles, and eliminate potential accidents. This application proposes a new potential collision risk prediction method, which can use the laws of the above error factors and the theoretical symmetry of potential collision risks to reduce the adverse effects of errors from a statistical perspective and realize the correction of potential collision risks between vehicles, so that the potential collision risk between any two vehicles is statistically symmetrical, thereby avoiding the above problems caused by being too conservative for the collision risk of one vehicle and too aggressive for the collision risk of another vehicle.

[0059] In practical applications, the potential collision risk prediction method proposed in this application can be applied to the fields of intelligent transportation, artificial intelligence, Internet of Vehicles, Internet of Things, etc., and specifically can be applied to application scenarios such as safe assisted driving and smart highways.

[0060] Among them, the Internet of Things (IOT) refers to the real-time collection of any object or process that needs to be monitored, connected, and interacted through various devices and technologies such as various information sensors, radio frequency identification technology, global positioning system, infrared sensor, laser scanner, etc., and the collection of various required information such as sound, light, heat, electricity, mechanics, biology, location, etc., and the realization of ubiquitous connection between objects and objects, objects and people through various possible network access, and the realization of intelligent perception, identification and management of objects and processes. It can be seen that the Internet of Things is an information carrier based on the Internet, traditional telecommunications networks, etc., which allows all ordinary physical objects that can be independently addressed to form an interconnected network.

[0061] The Internet of Vehicle (IOV) is an integrated network that can realize intelligent traffic management, intelligent dynamic information services and vehicle control. It is usually a huge interactive network composed of information such as vehicle location, speed and route. It is a typical application of Internet of Things technology in the field of transportation systems.

[0062] Generally speaking, the above-mentioned application fields proposed in this application can be realized in combination with CloudIOT technology. The CloudIOT aims to connect the information sensed and instructions received by sensor devices in the traditional Internet of Things to the Internet, truly realize networking, and realize massive data storage and computing through cloud computing technology. Due to the characteristics of the Internet of Things, objects are connected to objects, and the current operating status of each "object" is perceived in real time. In this process, a large amount of data information will be generated. How to aggregate this information and how to filter out useful information from the massive information for decision-making support for subsequent development have become key issues affecting the development of the Internet of Things. Therefore, the Internet of Things Cloud based on cloud computing and cloud storage technology has become a strong support for Internet of Things technology and applications.

[0063] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0064] It should be understood that the "system", "device", "unit" and / or "module" used in this application is a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the word can be replaced by other expressions.

[0065] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "a kind" and / or "the" do not refer to the singular, but also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of clearly identified steps and elements, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements. The elements defined by the sentence "includes a..." do not exclude the existence of other identical elements in the process, method, commodity or device that includes the elements.

[0066] Among them, in the description of the embodiments of the present application, unless otherwise specified, " / " means or, for example, A / B can mean A or B; "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships, for example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, in the description of the embodiments of the present application, "multiple" refers to two or more than two. The following terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features.

[0067] In addition, flow charts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, the various steps may be processed in reverse order or simultaneously. At the same time, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0068] Reference Figure 1 , is a schematic diagram of an optional application scenario applicable to the potential collision risk prediction method proposed in this application, such as application scenarios such as smart highways and safe assisted driving, etc. This application does not limit the specific application scenarios applicable to the potential collision risk prediction method. Figure 1 As described above, in specific application scenarios, in order to facilitate traffic monitoring and management, the collection system 11 of the traffic system usually sets up multiple cameras and various types of sensing devices on the traffic section to obtain the driving speed of the passing vehicle 12, locate the geographical location of the vehicle, collect images of the passing vehicle, so as to determine the type of vehicle based on this, etc. This application does not limit the number and type of collection devices included in the collection system 11 in the traffic system, such as cameras, speed sensors, gravity sensors, etc., which can be determined according to the specific application requirements of the corresponding road section, and this embodiment will not be described in detail here.

[0069] It should be understood that the vehicle type of the vehicle 12 is not limited to Figure 1The automobile shown may also be other types of vehicles, which will not be described in detail in this embodiment.

[0070] In addition, the collection system 11 in the traffic system is not limited to the collection of vehicle information, but can also collect relevant information of other traffic participants, such as pedestrians, etc., in order to meet different application needs. This application does not list them one by one here. The following embodiments mainly illustrate the application needs of potential collision risk prediction between vehicles.

[0071] The vehicle information collected by the above-mentioned collection system 11 can be sent to the database 13 through a wireless network for storage, so as to be subsequently called by the computer device 14 or other devices. This application does not limit the storage method of the vehicle information of each vehicle, such as segmented storage according to different physical sections, etc. It should be noted that since some vehicle information will change during the driving process of the vehicle, it is necessary to update the stored vehicle information accordingly, such as replacing the original vehicle information, or placing the latest collected vehicle information in other storage spaces for continued storage, retaining historical vehicle information, etc., which can be determined according to the storage method and storage requirements of the vehicle information, and is not limited to the several implementation methods listed in this embodiment.

[0072] In combination with the above analysis, the database 13 can be a data storage device for storing data, such as a server or a terminal device with a large storage space, and the present application does not limit the type and structure of the database 13. If the database 13 is a server, the server can be an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server providing cloud computing services, etc., which can be determined according to specific storage requirements.

[0073] Figure 1 The computer device 14 in the application scenario shown can be used to execute the potential collision risk prediction method proposed in the present application to obtain a more accurate prediction value of the potential collision risk of each vehicle, so as to provide a reliable warning to the vehicle driver, avoid being too conservative or too aggressive in predicting the collision risk of the vehicle, and ensure the safety of vehicle driving.

[0074] Among them, the computer device 14 can be a server and / or a terminal with certain computing capabilities. The server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services, etc.; the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, etc., but is not limited to this. This application does not limit the specific device type of the computer device 14.

[0075] In practical applications, if the computer device 14 is a server, according to the potential collision risk prediction method proposed in this application, a more accurate prediction value of the potential collision risk of each vehicle is obtained, and then the risk threshold is compared with the prediction value to generate risk warning information. The server can directly or indirectly send the risk warning information to the terminal pre-bound to the vehicle driver or the vehicle controller output through wired or wireless communication to remind the driver to take corresponding measures in time to avoid collision accidents. It should be noted that this application does not limit the communication connection method between the server and the driver terminal.

[0076] If the computer device 14 is a terminal, the application that implements the potential collision risk prediction method proposed in the present application can be embedded in other applications installed in the terminal, such as map applications, social applications (such as used as xx applet), and other applications that are usually activated by the vehicle driver during driving. Taking the map application as an example, the driver usually activates the map application for navigation before driving the vehicle. During this period, the map application will reliably predict the potential collision risk of the vehicle and promptly feedback the generated risk warning information to the driver, reminding the driver to take measures in advance to eliminate potential collision risks that may occur.

[0077] It should be noted that the application that implements the potential collision risk prediction method proposed in the present application can also be an independent application. The driver can download and install the application through the terminal's application store or web page, start the application before driving, or configure the application to the power-on state so that it can execute the potential collision risk prediction method proposed in the present application to improve driving safety.

[0078] In some other embodiments, the potential collision risk prediction method proposed in the present application may also be implemented by a vehicle-mounted control device. That is, the computer device 14 may also include a vehicle-mounted control device. Figure 1 Not shown, in this case, the above-mentioned vehicle may include an on-board control device with certain computing capabilities. The specific structure of the on-board control device is not described in detail in this application. It can be embedded in the vehicle or can be an independent device.

[0079] It should be understood that the composition of the application scenarios for implementing the potential collision risk prediction method proposed in the present application is not limited to the parts listed above. In actual applications, the selection of the software and hardware platform architecture, development environment, development language, message acquisition source, etc. for implementing the potential collision risk prediction method proposed in the present application can be varied. On the basis of the technical scheme of the potential collision risk prediction method described in the embodiments below of the present application, any improvement or equivalent transformation of a certain part proposed according to the inventive concept of the present application should not be excluded from the scope of protection of the present application.

[0080] Reference Figure 2 , is a flow chart of an optional example of a potential collision risk prediction method proposed in this application. This method can be applied to a computer device. As analyzed above, this application does not limit the type of the computer device. For example, Figure 2 As shown, the method may include:

[0081] Step S11, obtaining vehicle information of the first vehicle and the second vehicle corresponding to the current moment;

[0082] Among them, the vehicle information of each vehicle may include driving speed, geographic location, vehicle type and its corresponding vehicle configuration attribute information, etc. This application does not limit the specific content contained in the vehicle information, which can be determined based on the specific calculation requirements of potential collision risks.

[0083] Combined with Figure 1 As described in the application scenario shown, the collection system of the traffic system can send the collected vehicle information of each vehicle to the database for storage. Therefore, in the actual application of this embodiment, when potential collision risk prediction is required, the computer device can read the vehicle information corresponding to the first vehicle and the second vehicle at the current moment from the server. The first vehicle and the second vehicle are usually any two adjacent vehicles. The two vehicles can be driving relative to each other, driving in the same lane and in the same direction, driving in different lanes and in the same direction, etc. This application does not limit the position relationship and driving relationship between the first vehicle and the second vehicle.

[0084] Step S12, obtaining a first prediction value of a potential collision risk of the first vehicle with the second vehicle and a second prediction value of a potential collision risk of the second vehicle with the first vehicle based on the vehicle information at the current moment;

[0085] In this embodiment, based on the vehicle information of the first vehicle and the vehicle information of the second vehicle, a first prediction value E of the potential collision risk of the first vehicle to the second vehicle can be calculated respectively. 1,estmate , and a second prediction value E of the potential collision risk of the second vehicle to the first vehicle 2,estmate The potential collision risk can be represented by the collision probability or collision intensity between vehicles. Taking the vehicle as an unmanned vehicle as an example, the following formula (1) can be used but is not limited to:

[0086]

[0087] In the above formula (1), SPE v_bj It can represent the potential collision risk between unmanned autonomous driving type vehicles b and j. b and j are only used to distinguish the parameters of different vehicles. M represents the vehicle mass, r represents the distance between vehicles, v represents the driving data of the vehicles, and θ bjrepresents the relative direction of the vehicle, R (do not pay attention to the subscript) can represent the road surface factors of the road on which the vehicle is traveling, and the road surface factors can include the viscosity, humidity, slope, temperature, etc. of the road surface, and k1 and k3 can both represent constants, and the values ​​are determined according to the actual situation. For example, in some embodiments, k1=3, k3=the speed of light (i.e., the propagation speed of electromagnetic waves).

[0088] If the first vehicle and the second vehicle are vehicles with drivers, the potential collision risk between vehicle b and vehicle j can be obtained using formula (2):

[0089]

[0090] Regarding the difference between the above formula (2) and formula (1), DR (driving risk) can be a numerical value, indicating the driver risk factor, and the specific numerical value is not limited. Therefore, DR b The risk factor of the driver of vehicle b can be expressed as DR j The risk factor of the driver of vehicle j can be expressed as

[0091] Combined with the above analysis of the inventive concept of the present application, theoretically, the potential collision risk between the two vehicles should be equal. However, due to the random errors caused by the measurement and transmission described above, the predicted values ​​of the potential collision risk between the two vehicles are not equal, that is, the first predicted value E 1,estmate and the second predicted value E 2,estmate There is a difference between them, and the first prediction value E 1,estmate The true value E of the potential collision risk of the first vehicle to the second vehicle 1,real There is an error between the second prediction value E 2,estmate The true value E of the potential collision risk of the second vehicle to the first vehicle 2,real There are also errors in between.

[0092] In order to improve the accuracy and reliability of the prediction value of the potential collision risk between two vehicles, the present application proposes to obtain the first prediction value E 1,estmate and the second predicted value E 2,estmate Calibration is performed, and then the calibrated prediction value is used to make risk warning judgments. The specific calibration process can refer to the description of the corresponding part of the embodiment below.

[0093] Step S13, obtaining the historical traffic accident rate of the road section corresponding to the geographical location included in the vehicle information;

[0094] In practical applications, the purpose of predicting the potential collision risk of a vehicle is usually to effectively warn the vehicle itself and prevent it from colliding with other vehicles, so as to reduce the traffic accident rate of the road section where the vehicle is currently located. 1,estmate and the second predicted value E 2,estmate The difference between them needs to be within a certain range to help reduce the traffic accident rate on the road section where the vehicle is currently located.

[0095] Based on the above analysis, this embodiment can first obtain the historical traffic accident rate of the road section where the vehicle is currently located, and then calculate the critical value that can reduce the traffic accident rate of the road section, that is, the first prediction value E 1,estmate and the second predicted value E 2,estmate The critical range of the difference between .

[0096] It should be noted that this application does not limit the statistical method and storage method of traffic accident rate, which can usually be obtained from the management system of the traffic management department. The specific implementation process will not be described in detail.

[0097] Step S14, retrieving a plurality of first historical prediction values ​​of the potential collision risk corresponding to the first vehicle and a plurality of second historical prediction values ​​of the potential collision risk corresponding to the second vehicle acquired at different times;

[0098] Following the above analysis, the predicted values ​​of the potential collision risk between the two vehicles are not equal, or in other words, the error between the predicted value and the true value of the potential collision risk of the vehicle is a random error. In order to calibrate the predicted values ​​of the potential collision risk of each vehicle, the present application proposes to perform regular statistics on the difference between the potential collision risks of the two vehicles. Specifically, according to the central limit theorem, some phenomena are affected by many independent random factors. If the influence of each factor is very small, the total influence can be regarded as obeying a normal distribution.

[0099] Based on the above analysis, the present application may consider the first predicted value E of the potential collision risk of the first vehicle to the second vehicle obtained by the above calculation as follows: 1,estmate , and a second prediction value E of the potential collision risk of the second vehicle to the first vehicle 2,estmate All obey the normal distribution. Specifically, the first predicted value E 1,estmate Can obey the true value E 1,real is the mean, with α 2 is the normal distribution of variance, that is, E 1,estmate ~N(E 1,real , α 2 ); The second predicted value E 2,estmate Can obey the true value E2,real is the mean, and β 2 is the normal distribution of variance, that is, E 2,estmate ~N(E 2,real , β 2 ), this application does not elaborate on the principle of the normal distribution function.

[0100] It can be seen that in the normal distribution corresponding to the first vehicle and the second vehicle, E 1,real , α 2 、E 2,real and β 2 These four parameters are all unknown, but since this application does not require the calculation of E 1,real and E 2,real , then we only need to calculate α 2 and β 2 .

[0101] In some embodiments, in order to facilitate the calculation of the above α 2 and β 2 Based on the above analysis, the first predicted value E 1,estmate and the second predicted value E 2,estmate The normal distribution they obey is further deformed. Specifically, since the potential collision risk between two vehicles is symmetrical in theory, as shown in the above E 1,real =E 2,real =E real , after the above analysis, E 1,estmate ~N(E 1,real , α 2 ), E 2,estmate ~N(E 2,real , β 2 ), therefore, (E 1,estmate -E 2,estmate )~N(0,α 2 +β 2 ), or in other words, (E 2,estmate –E 1,estmate )~N(0,α 2 +β 2 ), that is, the difference between the potential collision risk prediction values ​​between the first vehicle and the second vehicle has a mean of 0 and a variance of α 2 +β 2 The normal distribution of .

[0102] Among them, since the potential collision risk of the vehicles at different times may change, that is, the predicted value of the potential collision risk may be a dynamic value, in order to obtain α that can reflect the difference in the potential collision risk between the two vehicles as a whole, 2 +β 2Therefore, this embodiment can retrieve multiple first historical prediction values ​​of the potential collision risk corresponding to (i.e., encountered by) the first vehicle obtained at different times, and multiple second historical prediction values ​​of the potential collision risk corresponding to (i.e., encountered by) the second vehicle, to calculate the above α 2 , β 2 ,

[0103] In some embodiments, specifically, multiple first historical prediction values ​​of potential collision risks corresponding to a first vehicle obtained at different times within a first historical time period from the current moment, and multiple second historical prediction values ​​of potential collision risks corresponding to a second vehicle obtained at different times within a second historical time period from the current moment can be retrieved.

[0104] Specifically, in one possible implementation, for each vehicle, multiple historical prediction values ​​obtained from the beginning of calculating the potential collision risk to the current moment can be retrieved, that is, the corresponding number of historical prediction values ​​of the potential collision risk are retrieved according to the number of calculations. It should be understood that the historical prediction values ​​of different potential collision risks may be the potential collision risk prediction values ​​of different vehicles for the vehicle. The multiple historical prediction values ​​may include several potential collision risk prediction values ​​of the same vehicle for the vehicle. The present application does not elaborate on the method for obtaining multiple historical prediction values ​​corresponding to the same vehicle.

[0105] Among them, it should be understood that since different vehicles enter the traffic road at different times, the starting time for calculating the potential collision risk is often different. Therefore, even for two vehicles, multiple historical prediction values ​​of potential collision risks obtained within the historical time period from the start time to the current moment are retrieved, the historical time period for different vehicles may also be different, that is, the above-mentioned first historical time period and the second historical time period may be different. Of course, these two historical time periods may also be the same, and this application does not limit this.

[0106] In another possible implementation, for each vehicle, the present application can also retrieve multiple historical prediction values ​​obtained within the historical time period from a preset moment to the current moment, and is not limited to the start time for calculating the potential collision risk mentioned above. In this case, the above-mentioned first historical time period and the second historical time period may be the same.

[0107] In summary, in the scenario of calculating the predicted value of the potential collision risk between two vehicles, the present application can obtain the predicted value of the potential collision risk between the two vehicles once at each moment; the computer device can respond to the prediction request for the potential collision risk of the two target vehicles and obtain the predicted value of the potential collision risk between the two target vehicles. At this time, the prediction request can be sent by an external device or automatically generated at a time interval; or, the present application can also determine that the target vehicle has entered a high-incidence section of a collision accident based on the obtained vehicle information, and then obtain the predicted value of the potential collision risk between the target vehicle and its adjacent vehicles. That is to say, the potential collision risk of the vehicle will be calculated only after it is determined that the vehicle has entered a high-incidence section of a collision accident. The specific calculation process can refer to the description of the corresponding part of the above embodiment. Among them, the high-incidence section of a collision accident can refer to a traffic geographical range where the historical traffic accident rate is greater than the accident threshold.

[0108] It should be noted that the present application does not limit the specific implementation of the above step S14, and is not limited to the several implementations described above, and can be flexibly determined according to actual needs. The calculation triggering conditions for the potential collision risk of the vehicle are also not limited to the several methods listed above.

[0109] Step S15, obtaining a difference variable of a potential collision risk between the first vehicle and the second vehicle at the current moment based on the first prediction value, the plurality of first historical prediction values, the second prediction value, and the plurality of second historical prediction values;

[0110] Combined with the above (E 1,estmate -E 2,estmate )~N(0,α 2 +β 2 ), or in other words, (E 2,estmate –E 1,estmate )~N(0,α 2 +β 2 ) analysis, this application can be α 2 +β 2 The difference variable of the potential collision risk between the first vehicle and the second vehicle at the current moment is recorded as follows. Based on this, in some embodiments, the above step S15 may include but is not limited to the following implementation methods:

[0111] Based on the first predicted value and the multiple first historical predicted values, the first risk variable of the first vehicle at the current moment is obtained, such as the above α 2 , and based on the second predicted value and the plurality of second historical predicted values, obtain the second risk variable of the second vehicle at the current moment, such as the above β 2 , then, the first risk variable and the second risk variable can be summed to obtain the difference variable of the potential collision risk between the first vehicle and the second vehicle at the current moment, as shown in the above α2 +β 2 .

[0112] In some embodiments, the above step S15 is implemented in combination with the calculation principle of sample variance to obtain the difference variable of the potential collision risk between the first vehicle and the second vehicle at the current moment. Specifically, refer to Figure 3 As shown, the method may include:

[0113] Step S21, performing variance calculation on the first prediction value and the plurality of first historical prediction values, and determining the obtained first sample variance as the first risk variable;

[0114] Step S22, performing variance calculation on the second predicted value and the plurality of second historical predicted values, and determining the obtained second sample variance as the second risk variable;

[0115] Step S23, summing the first risk variable and the second risk variable to obtain a difference variable of the potential collision risk between the first vehicle and the second vehicle at the current moment.

[0116] It can be seen that in this embodiment, the first sample variance can be the above α 2 , the second sample variance can be the above β 2 , the difference variable of the potential collision risk between the first vehicle and the second vehicle at the current moment can be the above α 2 +β 2 , after the above variance summation operation, the α 2 +β 2 becomes known data, then, according to (E 1,estmate -E 2,estmate )~N(0,α 2 +β 2 ), the difference value of the potential collision risk between the first vehicle and the second vehicle can be obtained in combination with the characteristics of the normal distribution function.

[0117] Step S16, using a normal distribution function to calculate the difference variable, to obtain a difference prediction range of the potential collision risk between the first vehicle and the second vehicle that is less than the historical traffic accident rate at the current moment;

[0118] Continuing with the above analysis, the smaller the difference in potential collision risk between the first vehicle and the second vehicle, the more conducive it is to reducing the traffic accident rate on the current road section. Therefore, if ξ is used to represent the difference prediction value, then the difference in potential collision risk between the first vehicle and the second vehicle is in the range of [-ξ,ξ], the purpose of reducing the traffic accident rate on the current road section can be achieved.

[0119] Combined with the above analysis, ξ is a random variable and obeys the normal distribution N(0, α2 +β 2 ), that is, the deviation between ξ and 0 is a random variable. In order to make each vehicle correct the predicted value of the potential collision risk it encounters according to the deviation, so as to reduce the traffic accident rate, combined with the expression of the normal distribution function, the present application can calculate the range of the potential collision risk difference between the first vehicle and the second vehicle that is less than the historical traffic accident rate p of the current road section, that is:

[0120]

[0121] In this embodiment, the above inequality can be solved to obtain ξ, thereby obtaining a predicted range [-ξ,ξ] of the difference in potential collision risk between the first vehicle and the second vehicle that is less than the historical traffic accident rate at the current moment. The specific solution process of the inequality is not described in detail in this embodiment.

[0122] In summary, the present application can use a normal distribution function to calculate the difference in potential collision risk between the first vehicle and the second vehicle based on the first prediction value, multiple first historical prediction values, the second prediction value, and multiple second historical prediction values, and obtain a difference prediction range that is less than the historical traffic accident rate. At this time, the mean of the normal distribution function is zero. The specific implementation process can be, but is not limited to, the implementation method described in the above steps S15 and S16.

[0123] In some embodiments, in the process of solving the difference ξ of the potential collision risk between the first vehicle and the second vehicle, the sample variance α can be used in the manner described in the above embodiment. 2 and β 2 To achieve this, other random variables that can represent an unbiased estimate of the variance can also be used to solve the problem, which will not be described in detail in this application.

[0124] Step S17, based on the difference prediction value within the difference prediction range, calibrate the first prediction value and the second prediction value respectively, and output a first target prediction value of the potential collision risk of the first vehicle with the second vehicle and a second target prediction value of the potential collision risk of the second vehicle with the first vehicle at the current moment;

[0125] In some instances, according to the above analysis method, a correction range for realizing the correction of the potential collision risk prediction of the first vehicle with the second vehicle, and a correction range for the potential collision risk prediction of the second vehicle with the first vehicle, that is, the above-mentioned difference prediction range [-ξ,ξ] is obtained. Then, the present application can select a correction amount for the potential collision risk prediction value corresponding to each of the first vehicle and the second vehicle from the difference prediction range [-ξ,ξ]. For example, a difference prediction value can be randomly sampled from the difference prediction range [-ξ,ξ] as the correction amount, and added to the potential collision risk prediction value of the corresponding vehicle to obtain the target prediction value.

[0126] Specifically, in some embodiments, Figure 4 But not limited to Figure 4 The method shown in the figure implements the above step S17, as shown in the figure Figure 4 As shown, based on the above embodiment, after obtaining the difference prediction range of the potential collision risk between the first vehicle and the second vehicle, which can reduce the traffic accident rate of the corresponding section as a whole, the potential collision risk prediction method proposed in the present application may also include:

[0127] Step S31, selecting any first difference prediction value and a second difference prediction value from the difference prediction range;

[0128] Among them, the first difference prediction value and the second difference prediction value can both be randomly selected from the difference prediction range obtained above, and the two can be equal or unequal, and this application does not limit this.

[0129] Step S32, summing the first difference prediction value and the first prediction value to obtain a first target prediction value of the potential collision risk of the first vehicle to the second vehicle at the current moment;

[0130] Step S33: summing the second difference prediction value and the second prediction value to obtain a second target prediction value of the potential collision risk of the second vehicle to the first vehicle at the current moment.

[0131] To sum up, the present application considers the application scenarios with random factors, and adopts statistical distribution (i.e. the normal distribution mentioned above) to describe the above random factors. From a statistical sense, the present application can generally reduce the actual difference in the potential collision risk between the first vehicle and the second vehicle. The average difference can be used to measure the effect of reducing the traffic accident rate achieved by the potential collision risk prediction method proposed in the present application. The specific verification process is not described in detail in this application.

[0132] In the actual application of this embodiment, the first target prediction value and the second target prediction value can be used to implement collision warning for the corresponding vehicle driver. Specifically, the vehicle collision threshold can be obtained first, which is the force that the driver can withstand. If it is greater than the vehicle collision threshold, it will threaten the driver's life safety. This application does not limit the specific value of the vehicle collision threshold, which can be expressed by the work done by the vehicle at the moment of collision, such as 22.8KN.

[0133] Afterwards, in actual application, during the driving process of the vehicle, after obtaining the target prediction value of the potential collision risk of the surrounding vehicles to it in the above-mentioned manner, the target prediction value can be compared with the above-mentioned vehicle collision threshold. If the target prediction value is greater than the vehicle collision threshold, the corresponding collision prompt information can be output to the driver of the vehicle to remind the driver to collect the corresponding anti-collision measures in time, or the collision prompt information can be directly sent to the vehicle (such as an unmanned vehicle), and the vehicle-mounted controller of the vehicle responds to the collision prompt information and executes the preset anti-collision measures, or the collision prompt information is sent to the management platform of the vehicle (such as an unmanned vehicle), and the management platform controls the vehicle to execute the corresponding anti-collision measures, etc., to eliminate the risk of collision with the surrounding vehicles and improve the driving safety of the vehicle. Based on this, for the above-mentioned first vehicle, it can be detected that the first target prediction value is greater than the vehicle collision threshold, and the first collision prompt information for the first vehicle can be output; the second target prediction value can be detected to be greater than the vehicle collision threshold, and the second collision prompt information for the second vehicle can be output.

[0134] It should be noted that this application does not limit the content of the above-mentioned collision warning information and its output method. If the computer device is a server, the server can send the generated collision warning information to the vehicle-mounted alarm output of the corresponding vehicle, or send it to the terminal device pre-bound by the driver, so as to remind the driver to take anti-collision measures in time when the risk of collision with surrounding vehicles is predicted to be high, that is, the target prediction value is greater than the vehicle collision threshold. The specific content is not limited.

[0135] To summarize, for the first vehicle and the second vehicle that are close to each other, after obtaining the first prediction value of the potential collision risk of the first vehicle to the second vehicle at the current moment, and the second prediction value of the potential collision risk of the second vehicle to the first vehicle, the present application does not directly provide a collision warning to the vehicle driver based on this, but takes into account the difference in potential collision risk between the first vehicle and the second vehicle, and the changing pattern of the difference, and proposes to calibrate the potential collision risk prediction values ​​corresponding to each of the two vehicles, so as to improve the accuracy and reliability of their respective potential collision risk prediction values, thereby improving the reliability and accuracy of vehicle warnings and ensuring the safety of vehicle driving.

[0136] Specifically, the present embodiment can utilize the first prediction value and the second prediction value obtained above, as well as a plurality of first historical prediction values ​​of the potential collision risks that the first vehicle has encountered, and a plurality of second historical prediction values ​​of the potential collision risks that the second vehicle has encountered, to obtain a difference variable of the potential collision risk between the first vehicle and the second vehicle at the current moment, thereby utilizing a normal distribution function to operate on the difference variable, and obtain a difference prediction range of the potential collision risk between the first vehicle and the second vehicle when the current moment is less than the historical traffic accident rate of the location, that is, a correction range for calibrating the first prediction value and the second prediction value. Afterwards, a difference prediction value, that is, a correction value, can be selected from the difference prediction range to calibrate the potential collision risk prediction value of the corresponding vehicle at the current moment, and obtain a more accurate and reliable target prediction value of the potential collision risk that can reduce the traffic accident rate, and make the potential collision risk between the two vehicles relatively symmetrical (that is, the difference is within a reasonable range), thereby avoiding the problem that the potential collision risk of one vehicle may be too conservative, while the potential collision risk of another vehicle may be too aggressive, and a reliable warning cannot be given for the potential collision risk that the corresponding vehicle may encounter.

[0137] Based on the description of the application scenarios of the potential collision risk prediction method proposed in the present application in the above embodiments, refer to Figure 5 The traffic scenario shown in FIG. 1 is used in this application to predict and calibrate the potential collision risk between vehicle A and vehicle B using a computer device as a server. For details, refer to FIG. Figure 6 The scenario flow diagram shown in the figure shows that the potential collision risk prediction method is executed as follows:

[0138] When vehicle A and vehicle B (such as any two adjacent vehicles on the road) are driving on the road, various sensing devices deployed by the traffic system and sensors of the vehicle itself can report the collected vehicle information related to the vehicle to the database for storage. The specific implementation process will not be described in detail. In this way, in order to ensure driving safety, when it is necessary to predict the potential collision risk between the two, the computer device can obtain the vehicle information of vehicle A and vehicle B from the database, and calculate the first prediction value of the potential collision risk of vehicle A to vehicle B and the second prediction value of the potential collision risk of vehicle B to vehicle A. Then, combined with the multiple first historical prediction values ​​of the potential collision risk encountered by vehicle A from the beginning of calculating the potential collision risk to the current moment retrieved from the database, correspondingly, the multiple second historical prediction values ​​of the potential collision risk encountered by vehicle B from the beginning of calculating the potential collision risk to the current moment are retrieved.

[0139] Based on the data obtained above, combined with the description of the corresponding part of the above embodiment, the difference variable of the potential collision risk between vehicle A and vehicle B at the current moment can be calculated, and then the difference variable can be operated using the normal distribution function to obtain the difference correction range between vehicle A and vehicle B at the current moment when it is less than the historical collision accident rate. After that, the correction amount can be randomly selected from the difference correction range to calibrate the potential collision risk prediction value of the corresponding vehicle, and use the calibrated target prediction value to make an early warning judgment. After generating the corresponding collision prompt information, it is sent to the corresponding vehicle output through the wireless network. For example, when it is determined that the target prediction value of the potential collision risk of vehicle A to vehicle B is greater than the vehicle collision threshold, if the two vehicles collide, the life safety of the driver of vehicle B may be threatened. The computer device sends the generated corresponding collision prompt information to the on-board controller of vehicle B, or the pre-bound terminal output of the driver. The on-board controller or the driver can execute the pre-configured anti-collision measures in advance to avoid the collision between vehicle B and vehicle A, thereby ensuring the life safety of the driver of vehicle B.

[0140] Similarly, the prediction process of the potential collision risk of vehicle A or other vehicles on the road is similar to the prediction process of the potential collision risk of vehicle B, and this application will not describe them in detail. In addition, the output method of the above-mentioned collision prompt information can be determined according to the vehicle type and the pre-configured collision avoidance measures execution rules, etc., and is not limited to the implementation method described above.

[0141] In some embodiments, in order to understand the effect of anti-collision warning by using the potential collision risk prediction method proposed in the present application, the present application can further count the collision false alarm rate (also known as the missed alarm rate), false alarm rate and other information. Specifically, 1000 calculations and calibration times are used as an example for statistics. According to the method described in the above embodiment, after completing the calibration of the potential collision risk prediction of the vehicle each time, it is possible to observe whether the corresponding vehicle in the simulator finally collides. If a collision occurs, but no early warning is given based on the previous prediction results, it can be considered that a missed alarm has occurred; if no collision actually occurs, but an early warning is given based on the previous prediction results, it can be considered that a false alarm has occurred. In this way, after 1000 statistics, the total number of missed alarms and the total number of false alarms can be obtained, and the missed alarm rate and the false alarm rate can be obtained by dividing them by 1000 respectively.

[0142] In actual application, according to the above statistical method, after 10 repeated tests, the comparison results of the prior art and the technical solution of the present application are obtained as shown in Table 1 below:

[0143] Table 1

[0144]

[0145] As shown in Table 1 above, the missed alarm rate and false alarm rate of anti-collision warning using the potential collision risk prediction method proposed in the present application are lower than the missed alarm rate and false alarm rate of anti-collision warning using the potential collision risk prediction method of the prior art. Therefore, the potential collision risk prediction method proposed in the present application greatly improves the system performance and reduces the collision accident rate.

[0146] Reference Figure 7 , is a schematic diagram of a structure of an optional example of a potential collision risk prediction device proposed in the present application, which can be applied to a computer device, such as Figure 7 As shown, the device may include:

[0147] An information acquisition module 21 is used to obtain a first prediction value of a potential collision risk of a first vehicle with a second vehicle at a current moment, a second prediction value of a potential collision risk of the second vehicle with the first vehicle, and a historical traffic accident rate corresponding to a road section where the first vehicle and the second vehicle are located;

[0148] A historical prediction value retrieval module 22 is used to retrieve a plurality of first historical prediction values ​​of the potential collision risk corresponding to the first vehicle obtained at different times within a preset historical time period from the current time, and a plurality of second historical prediction values ​​of the potential collision risk corresponding to the second vehicle;

[0149] a difference prediction range obtaining module 23, configured to perform a potential collision risk difference calculation between the first vehicle and the second vehicle using a normal distribution function based on the first prediction value, the plurality of first historical prediction values, the second prediction value, and the plurality of second historical prediction values, to obtain a difference prediction range that is less than the historical traffic accident rate, wherein the mean of the normal distribution function is zero;

[0150] A prediction value calibration module 24 is used to calibrate the first prediction value and the second prediction value respectively based on the difference prediction value within the difference prediction range, and output a first target prediction value of the potential collision risk of the first vehicle to the second vehicle and a second target prediction value of the potential collision risk of the second vehicle to the first vehicle at a current moment;

[0151] The first target prediction value and the second target prediction value are used to implement collision warning for corresponding vehicles.

[0152] In some embodiments, Figure 8 As shown, the above-mentioned difference prediction range obtaining module 23 may include:

[0153] a difference variable obtaining unit 231, configured to obtain a difference variable of a potential collision risk between the first vehicle and the second vehicle at a current moment based on the first prediction value, the plurality of first historical prediction values, the second prediction value, and the plurality of second historical prediction values;

[0154] In a possible implementation, the difference variable obtaining unit 231 may include:

[0155] A first risk variable obtaining unit, configured to obtain a first risk variable of the first vehicle at a current moment based on the first predicted value and the plurality of first historical predicted values;

[0156] A second risk variable obtaining unit, configured to obtain a second risk variable of the second vehicle at a current moment based on the second predicted value and the plurality of second historical predicted values;

[0157] The risk variable summing unit is used to sum the first risk variable and the second risk variable to obtain a difference variable of the potential collision risk between the first vehicle and the second vehicle at a current moment.

[0158] In practical applications, both the first risk variable and the second risk variable may be sample variances, but are not limited thereto.

[0159] The difference prediction range obtaining unit 232 is used to calculate the difference variable using a normal distribution function to obtain a difference prediction range of the potential collision risk between the first vehicle and the second vehicle that is less than the historical traffic accident rate at the current moment.

[0160] In some embodiments, reference Figure 8 , the above-mentioned prediction value calibration module 24 may include:

[0161] A random selection unit 241, configured to select any one of a first difference prediction value and a second difference prediction value from within the difference prediction range;

[0162] A first target prediction value obtaining unit 242 is used to sum the first difference prediction value and the first prediction value to obtain a first target prediction value of the potential collision risk of the first vehicle to the second vehicle at the current moment;

[0163] The second target prediction value obtaining unit 243 is used to sum the second difference prediction value and the second prediction value to obtain a second target prediction value of the potential collision risk of the second vehicle to the first vehicle at the current moment.

[0164] In some embodiments, based on the potential collision risk prediction device described in the above embodiments, the information acquisition module 21 may include:

[0165] A vehicle information acquisition unit, used to acquire vehicle information of the first vehicle and the second vehicle corresponding to the current moment;

[0166] The vehicle information includes driving speed, geographic location, vehicle type, and vehicle configuration attribute information corresponding to the vehicle type;

[0167] a collision risk prediction unit, configured to obtain, based on the vehicle information corresponding to the current moment, a first prediction value of a potential collision risk of the first vehicle with the second vehicle, and a second prediction value of a potential collision risk of the second vehicle with the first vehicle;

[0168] In some embodiments, the above device may use but is not limited to any of the following acquisition units to trigger acquisition of a predicted value of a potential collision risk between two vehicles:

[0169] A first acquisition unit, used for acquiring a predicted value of a potential collision risk between two vehicles at every moment;

[0170] A second acquisition unit, configured to respond to a prediction request for a potential collision risk of two target vehicles and acquire a predicted value of a potential collision risk between the two target vehicles;

[0171] The third acquisition unit is used to determine, based on the acquired vehicle information, that the target vehicle has entered a road section with a high incidence of collision accidents, and obtain a predicted value of the potential collision risk between the target vehicle and its adjacent vehicles. The road section with a high incidence of collision accidents refers to a traffic geographical range where the historical traffic accident rate is greater than an accident threshold.

[0172] On the basis of the potential collision risk prediction device described in the above embodiments, the potential collision risk prediction device proposed in the present application may further include:

[0173] A vehicle collision threshold acquisition module, used to acquire a vehicle collision threshold;

[0174] A first collision prompt information output module, configured to detect that the first target prediction value is greater than the vehicle collision threshold, and output first collision prompt information for the first vehicle;

[0175] The second collision prompt information output module is used to detect that the second target prediction value is greater than the vehicle collision threshold and output second collision prompt information for the second vehicle.

[0176] It should be understood that each module and unit in the above-mentioned device embodiment can be a functional module composed of an application program, and the specific process of realizing the corresponding function can refer to the description of the corresponding part of the above-mentioned method embodiment, which will not be repeated in this embodiment.

[0177] An embodiment of the present application also provides a storage medium on which a program is stored. The program can be called and executed by a processor to implement the various steps of the above-mentioned potential collision risk prediction method. The specific implementation method can refer to the description of the corresponding part of the above-mentioned embodiment.

[0178] Reference Fig. 9 , is a schematic diagram of a hardware structure of an optional example of a computer device applicable to the potential collision risk prediction method proposed in the present application, and the computer device may include: a communication interface 31, a memory 32 and a processor 33;

[0179] In the embodiment of the present application, the communication interface 31, the memory 32, and the processor 33 can communicate with each other through a communication bus, and the number of the communication interface 31, the memory 32, the processor 33 and the communication bus can be at least one.

[0180] Optionally, the communication interface 31 can be an interface of a communication module, such as a communication module for implementing a wired network or a wireless network, such as a GPS module, a WIFI module, a GSM module, an interface for 4G / 5G / 6G network communication, etc. It can also include a USB interface for realizing internal data interaction of computer devices, various serial / parallel interfaces, etc. This application does not limit the type and number of the communication interface.

[0181] In actual applications, the above-mentioned communication interface 31 can be used to realize data communication with databases, vehicle-mounted equipment, and various components of the traffic system, to obtain vehicle information, historical collision accident rates, historical prediction values ​​of potential collision risks and other data, and to feed back the target prediction value of the vehicle's potential collision risk to the on-board controller of the corresponding vehicle, so that the on-board controller can perform anti-collision warnings and output collision prompt information accordingly, or directly send the generated collision prompt information to the vehicle's on-board alarm or driver terminal output, which can be determined according to the needs of the specific scenario and will not be elaborated in this application.

[0182] The memory 32 may include a high-speed RAM memory, and may also include a non-volatile memory, such as at least one disk memory. The processor 33 may be a central processing unit CPU, or an application specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present application.

[0183] In an embodiment of the present application, the memory 32 is used to store a program for implementing the potential collision risk prediction method described in any of the above method embodiments. The processor 33 can load and execute the program stored in the memory 32 to implement the various steps of the potential collision risk prediction method described in the above corresponding method embodiment. The specific implementation process can refer to the description of the corresponding part of the above method embodiment.

[0184] It should be understood that Fig. 9 The structure of the computer device shown does not constitute a limitation on the computer device in the embodiment of the present application. In practical applications, the computer device may be configured according to the specific device type, including the computer device of FIG. Fig. 9 More or fewer components as shown, or combinations of certain components, such as a terminal, may also include various input devices, output devices, etc., which are not listed one by one in this application.

[0185] Finally, it should be noted that the various embodiments in this specification are described in a progressive or parallel manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other. For the devices and computer equipment 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 part.

[0186] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0187] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0188] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the core idea or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A potential collision risk prediction method, characterized in that: The method comprises: Obtaining a first prediction value of a potential collision risk of a first vehicle with a second vehicle at a current moment, a second prediction value of a potential collision risk of a second vehicle with the first vehicle, and a historical traffic accident rate corresponding to a road section where the first vehicle and the second vehicle are located; Retrieving a plurality of first historical prediction values ​​of the potential collision risk corresponding to the first vehicle and a plurality of second historical prediction values ​​of the potential collision risk corresponding to the second vehicle acquired at different times; Based on the first prediction value, the plurality of first historical prediction values, the second prediction value, and the plurality of second historical prediction values, a potential collision risk difference calculation between the first vehicle and the second vehicle is performed using a normal distribution function to obtain a difference prediction range that is less than the historical traffic accident rate, and a mean value of the normal distribution function is zero; Based on the difference prediction value within the difference prediction range, the first prediction value and the second prediction value are calibrated respectively, and a first target prediction value of the potential collision risk of the first vehicle with the second vehicle and a second target prediction value of the potential collision risk of the second vehicle with the first vehicle are output at the current moment; The first target prediction value and the second target prediction value are used to implement collision warning for corresponding vehicles.

2. The method according to claim 1, characterized in that The method of calculating the difference in potential collision risk between the first vehicle and the second vehicle using a normal distribution function based on the first prediction value, the plurality of first historical prediction values, the second prediction value, and the plurality of second historical prediction values ​​to obtain a difference prediction range that is less than the historical traffic accident rate includes: Obtaining a difference variable of a potential collision risk between the first vehicle and the second vehicle at a current moment based on the first prediction value, the plurality of first historical prediction values, the second prediction value, and the plurality of second historical prediction values; The difference variable is operated using a normal distribution function to obtain a difference prediction range of the potential collision risk between the first vehicle and the second vehicle that is less than the historical traffic accident rate at the current moment.

3. The method according to claim 2, characterized in that The obtaining, based on the first prediction value, the plurality of first historical prediction values, the second prediction value, and the plurality of second historical prediction values, a difference variable of the potential collision risk between the first vehicle and the second vehicle at the current moment includes: Based on the first prediction value and the plurality of first historical prediction values, obtaining a first risk variable of the first vehicle at a current moment; Based on the second predicted value and the plurality of second historical predicted values, obtaining a second risk variable of the second vehicle at a current moment; The first risk variable and the second risk variable are summed to obtain a difference variable of the potential collision risk between the first vehicle and the second vehicle at the current moment.

4. The method according to claim 3, characterized in that The first risk variable and the second risk variable are both sample variances.

5. The method according to claim 1, characterized in that The first prediction value and the second prediction value are calibrated respectively based on the difference prediction value within the difference prediction range, and a first target prediction value of the potential collision risk of the first vehicle to the second vehicle and a second target prediction value of the potential collision risk of the second vehicle to the first vehicle are output at the current moment, including: Select any first difference prediction value and any second difference prediction value from the difference prediction range; Summing the first difference prediction value and the first prediction value to obtain a first target prediction value of the potential collision risk of the first vehicle to the second vehicle at the current moment; The second difference prediction value and the second prediction value are summed to obtain a second target prediction value of the potential collision risk of the second vehicle with the first vehicle at the current moment.

6. The method according to claim 1, characterized in that The obtaining, at the current moment, a first predicted value of a potential collision risk between the first vehicle and the second vehicle, and a second predicted value of a potential collision risk between the second vehicle and the first vehicle, comprises: Acquire vehicle information of the first vehicle and the second vehicle corresponding to the current moment, wherein the vehicle information includes driving speed, geographic location, vehicle type, and vehicle configuration attribute information corresponding to the vehicle type; Based on the vehicle information corresponding to the current moment, a first prediction value of the potential collision risk of the first vehicle with the second vehicle and a second prediction value of the potential collision risk of the second vehicle with the first vehicle are obtained.

7. The method according to claim 6, characterized in that The process of obtaining the predicted value of the potential collision risk between two vehicles includes any of the following implementation methods: Obtain the predicted value of the potential collision risk between two vehicles at each moment; In response to a prediction request for a potential collision risk of two target vehicles, obtaining a prediction value of the potential collision risk between the two target vehicles; Based on the acquired vehicle information, it is determined that the target vehicle has entered a road section with a high incidence of collision accidents, and a predicted value of the potential collision risk between the target vehicle and its adjacent vehicles is obtained. The road section with a high incidence of collision accidents refers to a traffic geographical range where the historical traffic accident rate is greater than the accident threshold.

8. The method according to any one of claims 1 to 7, characterized in that: The method further comprises: Get the vehicle collision threshold; detecting that the first target prediction value is greater than the vehicle collision threshold, and outputting first collision prompt information for the first vehicle; It is detected that the second target prediction value is greater than the vehicle collision threshold, and second collision prompt information for the second vehicle is output.

9. A potential collision risk prediction device, characterized in that: The device comprises: An information acquisition module, used to acquire a first prediction value of a potential collision risk of a first vehicle with a second vehicle at a current moment, a second prediction value of a potential collision risk of a second vehicle with the first vehicle, and a historical traffic accident rate corresponding to a road section where the first vehicle and the second vehicle are located; a historical prediction value retrieval module, used to retrieve a plurality of first historical prediction values ​​of the potential collision risk corresponding to the first vehicle obtained at different times within a preset historical time period from the current time, and a plurality of second historical prediction values ​​of the potential collision risk corresponding to the second vehicle; a difference prediction range obtaining module, configured to perform a potential collision risk difference calculation between the first vehicle and the second vehicle using a normal distribution function based on the first prediction value, the plurality of first historical prediction values, the second prediction value, and the plurality of second historical prediction values, to obtain a difference prediction range that is less than the historical traffic accident rate, wherein the mean of the normal distribution function is zero; a prediction value calibration module, configured to calibrate the first prediction value and the second prediction value respectively based on the difference prediction value within the difference prediction range, and output a first target prediction value of the potential collision risk of the first vehicle with the second vehicle and a second target prediction value of the potential collision risk of the second vehicle with the first vehicle at a current moment; The first target prediction value and the second target prediction value are used to provide collision warning to the corresponding vehicle driver.

10. A computer device, characterized in that: The computer device comprises: Communication interface; A memory for storing a program for implementing the potential collision risk prediction method as claimed in claim 1; A processor is used to load and execute the program stored in the memory to implement each step of the potential collision risk prediction method as claimed in claim 1.

Images