A road condition risk early warning method, device and control equipment

Abstract

The application provides a road network risk early warning method, device and control equipment, which are used for early warning of risks in front of vehicles during driving. The road condition risk early warning method comprises the following steps: acquiring at least one risk area in a current road network, and determining secondary risk areas of each risk area; the secondary risk area refers to an area containing the corresponding risk area and being spaced apart from the edge of the corresponding risk area by a preset distance; for each online vehicle entering the secondary risk area, the probability of each online vehicle passing through the risk area is respectively predicted in combination with the actual driving track of each online vehicle and the distance between each online vehicle and the corresponding risk area; and risk early warning is performed on the online vehicle whose probability of passing through the risk area exceeds a preset probability.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of Internet of Vehicles, and particularly relates to a road condition early warning method, system and control device. BACKGROUND

[0002] At present, the early warning in the process of vehicle driving includes but is not limited to collision early warning, lane deviation early warning, fire early warning, emergency braking early warning, tire detection early warning, etc. For example, CN217347777U mentions that collision early warning is performed by comparing the infrared ranging and image analysis ranging with a preset threshold value; CN108099819B mentions that the vehicle deviation state is judged by the minimum distance between the real-time collected driving route of the vehicle and the lane line being less than the deviation threshold, and early warning is performed; CN113053054A mentions that the data is uploaded to the vehicle cloud through the vehicle-mounted sensor, the vehicle cloud discriminates through a fire prediction model, and the discrimination result is sent back to the vehicle networking module for early warning; CN218159326U mentions that the braking information of the front vehicle is transmitted to the rear vehicle through the vehicle-mounted communication module for braking early warning; CN207931400U mentions that the tire pressure sensor signal is uploaded to the vehicle networking platform, and through aggregation analysis, not only the tire pressure of the vehicle itself can be early warned, but also the tire pressure early warning information of the side vehicle can be monitored.

[0003] Through comparison, it is found that the technology in the field of road condition risk early warning in the process of vehicle driving is relatively blank in the prior art. SUMMARY

[0004] The present application provides a road condition risk early warning method, device and control device for early warning of the front risk in the process of vehicle driving.

[0005] The technical scheme of the present application is as follows:

[0006] The present application provides a road condition risk early warning method, which comprises:

[0007] At least one risk area in the current road network is acquired, and the secondary risk areas of each risk area are determined; the secondary risk area refers to an area containing the corresponding risk area and being spaced apart from the edge of the corresponding risk area by a preset distance;

[0008] For each online vehicle entering the secondary risk area, the probability of each online vehicle passing through the risk area is predicted in combination with the actual driving track of each online vehicle and the distance between each online vehicle and the corresponding risk area;

[0009] The online vehicle whose probability of passing through the risk area exceeds a preset probability is early warned. Preferably, the step of acquiring at least one risk area in the current road network comprises:

[0010] Real-time risk index data in the current road network is acquired;

[0011] extracting, from the real-time risk indicator data, risk points of a point type and geographical areas of an area type;

[0012] expanding outwardly from each risk point a preset distance to form a geographical area, taking the geographical area as a risk area, and dividing each geographical area into a risk area.

[0013] Preferably, the step of predicting the probability of each online vehicle passing through the risk area in combination with the actual driving track of each online vehicle and the distance between each online vehicle and the corresponding risk area comprises:

[0014] determining a distance influence factor parameter value based on the straight-line distance between the online vehicle and the edge of the risk area, and

[0015] determining a path matching degree influence factor parameter value based on the matching degree between all drivable paths from the current position of the online vehicle to the edge of the risk area and the actual driving track of the online vehicle;

[0016] predicting the probability of the online vehicle passing through the risk area according to the distance influence factor parameter value and the path matching degree influence factor parameter value.

[0017] Preferably, the smaller the straight-line distance between the online vehicle and the edge of the risk area, the greater the distance influence factor parameter value, and the greater the predicted probability of the online vehicle passing through the risk area.

[0018] Preferably, the step of determining the distance influence factor parameter value based on the straight-line distance between the online vehicle and the edge of the risk area comprises: determining the distance influence factor parameter value according to the straight-line distance between the online vehicle and the edge of the risk area, a pre-labeled maximum safe straight-line distance between the online vehicle and the edge of the risk area, and a pre-labeled minimum safe straight-line distance between the online vehicle and the edge of the risk area.

[0019] Preferably, the step of determining the distance influence factor parameter value according to the straight-line distance between the online vehicle and the edge of the risk area, a pre-labeled maximum safe straight-line distance between the online vehicle and the edge of the risk area, and a pre-labeled minimum safe straight-line distance between the online vehicle and the edge of the risk area comprises:

[0020] by the formula:

[0021] calculating the distance influence factor parameter value p ss is a straight-line distance between the online vehicle and the edge of the risk area, Max(S) is a maximum safe straight-line distance between the online vehicle and the edge of the risk area, and Min(S) is a minimum safe straight-line distance between the online vehicle and the edge of the risk area.

[0022] Preferably, based on the matching degrees of all drivable paths between the online vehicle and the edge of the risk area from the current position respectively with the actual driving trajectory of the online vehicle, the step of determining the path matching degree influence factor parameter value comprises:

[0023] The one matching degree with the largest all-matching-degree median is determined as the path matching degree influence factor parameter value.

[0024] Preferably, according to the distance influence factor parameter value and the path matching degree influence factor parameter value, the step of predicting the probability of the online vehicle passing through the risk area comprises:

[0025] a first product of the distance influence factor parameter value and a first weight proportion is calculated, and a second product of the path matching degree influence factor parameter value and a second weight proportion is calculated;

[0026] the first product and the second product are added to obtain the probability of the online vehicle passing through the risk area;

[0027] the sum of the first weight proportion and the second weight proportion is 1.

[0028] The application also provides a road condition risk early warning device, which comprises:

[0029] an acquisition module, configured to acquire at least one risk area in a current road network and determine secondary risk areas of each risk area; the secondary risk area refers to an area containing the corresponding risk area and being spaced apart from the edge of the corresponding risk area by a preset distance;

[0030] a prediction module, configured to, for each online vehicle entering the secondary risk area, combine the actual driving trajectory of each online vehicle and the distance between each online vehicle and the corresponding risk area to respectively predict the probability of each online vehicle passing through the risk area;

[0031] an early warning module, configured to make a risk early warning for the online vehicle whose probability of passing through the risk area exceeds a preset probability.

[0032] The application also provides a control device, which comprises a processor, a memory, and a program or instruction stored on the memory and executable on the processor, and the program or instruction is executed by the processor to implement the steps of the road condition risk early warning method as described above.

[0033] The application has the following beneficial effects:

[0034] The vehicle networking platform divides a secondary risk area based on a risk area generated in real time in a current road network, predicts a probability of a vehicle entering the risk area when the vehicle enters the secondary risk area, in combination with a driving path of the vehicle and a distance between the vehicle and the risk area, and issues a warning to the vehicle when the probability exceeds a preset probability, thereby playing a role of warning a driver and helping the driver to re-plan a driving path in advance. BRIEF DESCRIPTION OF DRAWINGS

[0035] Figure 1 A schematic diagram of a road condition risk prediction method in an embodiment of the present application;

[0036] Figure 2 A detailed schematic diagram of a road condition risk prediction method in an embodiment of the present application;

[0037] Figure 3 A schematic diagram of a road condition risk prediction device in an embodiment of the present application. DETAILED DESCRIPTION

[0038] An embodiment of the present application provides a road condition risk prediction method, as shown in Figure 1 and Figure 2 , the method comprising:

[0039] S1, acquiring at least one risk area in a current road network and determining secondary risk areas of the risk areas; the secondary risk area refers to an area containing a corresponding risk area and being spaced apart from an edge of the corresponding risk area by a preset distance;

[0040] S2, for each online vehicle entering the secondary risk area, in combination with an actual driving trajectory of each online vehicle and a distance between each online vehicle and a corresponding risk area, respectively predicting a probability of each online vehicle passing through the risk area;

[0041] S3, when there is an online vehicle passing through the risk area with a probability exceeding a preset probability, making a risk warning to the online vehicle.

[0042] In the above step S1, the specific process of acquiring at least one risk area in a current road network is as follows:

[0043] Obtain real-time risk index data in the current road network; these real-time risk index data can come from the vehicle networking platform self-service system, third-party weather station data, third-party map data, third-party traffic road equipment data, and online vehicle data in the road network, such as weather, disaster, road conditions, broken-down vehicles, accident points, etc. The specific information format of real-time risk index data from different devices is different, and the information needed is extracted from the real-time risk index data, mainly the risk type of the real-time risk index data (such as weather warning type, geological disaster warning type, road congestion type, road construction type, etc. The division of risk type is determined by pre-marking) to identify the risk area needed in the above steps.

[0044] Specifically, as Figure 2 This step S1 includes:

[0045] Obtain real-time risk index data in the current road network;

[0046] Extract risk points with a risk type of point type and geographical areas with a risk type of area type from the real-time risk index data;

[0047] The geographical area formed by extending a preset distance outward from each risk point as the center is taken as a risk area, and each geographical area is divided into a risk area.

[0048] Among them, the risk type of point type refers to the risk of a small area or a small mileage road section such as traffic accidents, traffic congestion, road maintenance, traffic control, steep slope road sections, rockfall points, and continuous bend road sections in real-time risk index data. This kind of risk type with small occurrence range is divided into point type; while the risk type of area type refers to the risk of a large area such as heavy fog, heavy rain, high temperature, flood, earthquake, etc. in real-time risk index data. This kind of risk type with large occurrence range is divided into area type. Whether each kind of risk belongs to the risk type of point type or area type needs to be defined in advance.

[0049] Among them, for this kind of data with a risk type of point type, the geographical area formed by extending a preset distance outward from each risk point as the center is taken as a risk area.

[0050] Thus, the acquisition of the risk area is realized.

[0051] As Figure 2 The above step S2 is specifically:

[0052] Determine the distance influence factor parameter value based on the straight-line distance between the online vehicle and the edge of the risk area, and

[0053] Based on the matching degree between all drivable paths of online vehicles from their current location to the edge of the risk area and the actual driving trajectory of online vehicles, the parameter values ​​of the path matching degree influencing factors are determined.

[0054] Based on the distance-related and path-matching factors, the probability of online vehicles passing through risk areas is predicted.

[0055] The steps for determining the distance influencing factor parameter values ​​based on the straight-line distance between the online vehicle and the edge of the risk area include: determining the distance influencing factor parameter values ​​based on the straight-line distance between the online vehicle and the edge of the risk area, the pre-calibrated maximum safe straight-line distance between the online vehicle and the edge of the risk area, and the pre-calibrated minimum safe straight-line distance between the online vehicle and the edge of the risk area.

[0056] Specifically, through the formula:

[0057] Calculate the parameter value p of the distance influencing factors s s is the straight-line distance between the online vehicle and the edge of the risk area, Max(S) is the pre-calibrated maximum safe straight-line distance between the online vehicle and the edge of the risk area, and Min(S) is the pre-calibrated minimum safe straight-line distance between the online vehicle and the edge of the risk area.

[0058] The steps for determining the parameter values ​​of path matching factors, based on the matching degree between all drivable paths of online vehicles from their current location to the edge of the risk area and the actual driving trajectory of the online vehicles, include:

[0059] The highest matching score among all matching scores is determined as the parameter value of the path matching score influencing factors. That is, it is determined by the formula:

[0060] p r =Max(P r )

[0061] Determine the parameter value p of the factors affecting path matching degree r P r The calculation formula is:

[0062] P r ={f(n)|f(n)=S(R) n ,G)(n>0,n∈Z)}

[0063] n represents the number of passable paths, S(R) n G) represents a passable path R n The degree of matching with the actual driving trajectory G, S(R) nG) can be obtained using methods including but not limited to Longest Common Subsequence (LCSS), Dynamic Time Planning (DTW), Edit Distance Based (EDR), Cell Similarity (CSIM), etc., which will not be elaborated here.

[0064] Furthermore, the first product of the distance influencing factor parameter value and the first weight ratio is calculated, and the second product of the path matching degree influencing factor parameter value and the second weight ratio is calculated.

[0065] Add the first product and the second product to get the probability that the online vehicle passes through the risk area;

[0066] The sum of the first weight percentage and the second weight percentage is 1.

[0067] That is, through formula p i =p s w s +p r w r The probability p of an online vehicle passing through a risk area is calculated. i p s w represents the distance-influencing factor parameter value. s As the first weighted percentage, w r This is the second weighting percentage. The specific values ​​for the first and second weighting percentages need to be obtained through prior testing.

[0068] Analysis of the above formulas shows that the smaller the straight-line distance between the online vehicle and the edge of the risk area, the larger the distance influencing factor parameter value, and the greater the predicted probability that the online vehicle will pass through the risk area.

[0069] In the aforementioned step S3, the warning to online vehicles can be sent via the vehicle network to the online vehicles, informing the driver that the probability of the vehicle passing through the risk area is high, and prompting the driver whether to switch to a route to avoid the risk area.

[0070] When the probability of an online vehicle passing through a risk area is lower than the preset probability, no further warnings will be issued for the online vehicle.

[0071] The above methods can provide early warnings for online vehicles connected to the vehicle networking platform, thereby improving the driving experience.

[0072] like Figure 3 The present invention also provides a road condition risk warning device, the device comprising:

[0073] The acquisition module 101 is used to acquire at least one risk area in the current road network and determine the secondary risk areas of each risk area; the secondary risk area refers to the area that contains the corresponding risk area and is separated from the edge of the corresponding risk area by a preset distance;

[0074] The prediction module 102 is configured to predict the probability of each online vehicle passing through the risk area according to the actual driving track of each online vehicle and the distance between each online vehicle and the corresponding risk area.

[0075] The early warning module 103 is configured to give a risk warning to the online vehicle when the probability of the online vehicle passing through the risk area exceeds a preset probability.

[0076] The application further provides a control device, which comprises a processor, a memory, and a program or instruction stored in the memory and executable on the processor, and the program or instruction is executed by the processor to realize the steps of the road risk warning method.

[0077] The device can achieve the technical effects corresponding to the method, that is, based on the risk area generated in real time in the current road network, the secondary risk area is divided, when a vehicle enters the secondary risk area, the driving path of the vehicle itself and the distance between the vehicle and the risk area are combined to predict the probability of the vehicle entering the risk area, and when the probability exceeds a preset probability, the vehicle is given a warning, thereby playing a role of warning the driver and helping the driver to choose to re-plan the driving route in advance.

[0078] The above embodiments are the preferred embodiments of the application, but the embodiments of the application are not limited to the above embodiments, and any changes, modifications, substitutions, combinations, simplifications made without departing from the spirit and principles of the application shall be equivalent replacement methods, which are all included in the protection scope of the application.

Claims

1. A road condition risk early warning method, characterized in that, The method includes: Obtain at least one risk area in the current road network and determine the secondary risk areas of each risk area; a secondary risk area is an area that contains the corresponding risk area and is separated from the edge of the corresponding risk area by a preset distance; For each online vehicle entering the secondary risk area, the probability of each online vehicle passing through the risk area is predicted by combining the actual driving trajectory of each online vehicle and the distance between each online vehicle and the corresponding risk area. Risk warnings are issued to online vehicles whose probability of passing through risk areas exceeds a preset probability. For each online vehicle entering a secondary risk area, the steps for predicting the probability of each online vehicle passing through the risk area, based on the actual driving trajectory of each online vehicle and the distance between each online vehicle and the corresponding risk area, include: Based on the straight-line distance between the online vehicle and the edge of the risk area, the parameter values ​​of the distance influencing factors are determined, and Based on the matching degree between all drivable paths of online vehicles from their current location to the edge of the risk area and the actual driving trajectory of online vehicles, the parameter values ​​of the path matching degree influencing factors are determined. Based on the distance-related and path-matching factors, the probability of online vehicles passing through risk areas is predicted.

2. The road condition risk warning method according to claim 1, characterized in that, The steps to identify at least one risk area in the current road network include: Obtain real-time risk indicator data for the current road network; Extract point-type risk points and regional-type geographical areas from the real-time risk indicator data. A risk zone is defined as a geographical area formed by extending a preset distance outward from each risk point, and each geographical area is designated as a risk zone.

3. The road condition risk warning method according to claim 1, characterized in that, The smaller the straight-line distance between the online vehicle and the edge of the risk area, the larger the distance influencing factor parameter value, and the greater the predicted probability that the online vehicle will pass through the risk area.

4. The road condition risk warning method according to claim 1 or 3, characterized in that, The steps for determining the distance influencing factor parameters based on the straight-line distance between the online vehicle and the edge of the risk area include: The distance influencing factor parameter values ​​are determined based on the straight-line distance between the online vehicle and the edge of the risk area, the pre-calibrated maximum safe straight-line distance between the online vehicle and the edge of the risk area, and the pre-calibrated minimum safe straight-line distance between the online vehicle and the edge of the risk area.

5. The road condition risk early warning method according to claim 4, characterized in that, The steps for determining the distance influencing factor parameters based on the straight-line distance between the online vehicle and the edge of the risk area, the pre-calibrated maximum safe straight-line distance between the online vehicle and the edge of the risk area, and the pre-calibrated minimum safe straight-line distance between the online vehicle and the edge of the risk area include: Through the formula: The distance influencing factor parameter value ps is calculated, where s is the straight-line distance between the online vehicle and the edge of the risk area, Max(S) is the pre-calibrated maximum safe straight-line distance between the online vehicle and the edge of the risk area, and Min(S) is the pre-calibrated minimum safe straight-line distance between the online vehicle and the edge of the risk area.

6. The road condition risk early warning method according to claim 1, characterized in that, The steps for determining the parameter values ​​of path matching factors, based on the matching degree between all drivable paths of online vehicles from their current location to the edge of the risk area and the actual driving trajectory of the online vehicles, include: The matching degree with the largest value among all matching degrees is determined as the parameter value of the path matching degree influencing factors.

7. The road condition risk warning method according to claim 1, characterized in that, Based on the distance-influencing factor parameter values ​​and the path matching degree-influencing factor parameter values, the steps for predicting the probability of online vehicles passing through risk areas include: Calculate the first product of the distance influencing factor parameter value and the first weight percentage, and calculate the second product of the path matching degree influencing factor parameter value and the second weight percentage; Add the first product and the second product to get the probability that the online vehicle passes through the risk area; The sum of the first weight percentage and the second weight percentage is 1.

8. A road condition risk warning device, characterized in that, The device includes: The acquisition module is used to acquire at least one risk area in the current road network and determine the secondary risk areas of each risk area; the secondary risk area refers to the area that contains the corresponding risk area and is separated from the edge of the corresponding risk area by a preset distance; The prediction module is used to predict the probability of each online vehicle passing through the risk area by combining the actual driving trajectory of each online vehicle and the distance between each online vehicle and the corresponding risk area. The early warning module is used to issue risk warnings to online vehicles whose probability of passing through risk areas exceeds a preset probability. For each online vehicle entering a secondary risk area, the steps for predicting the probability of each online vehicle passing through the risk area, based on the actual driving trajectory of each online vehicle and the distance between each online vehicle and the corresponding risk area, include: Based on the straight-line distance between the online vehicle and the edge of the risk area, the parameter values ​​of the distance influencing factors are determined, and Based on the matching degree between all drivable paths of online vehicles from their current location to the edge of the risk area and the actual driving trajectory of online vehicles, the parameter values ​​of the path matching degree influencing factors are determined. Based on the distance-related and path-matching factors, the probability of online vehicles passing through risk areas is predicted.

9. A control device, characterized in that, It includes a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the steps of the road condition risk warning method as described in any one of claims 1 to 7.

Images