Route safety assessment method and device, electronic device, and computer-readable storage medium

Abstract

The present application discloses a route safety assessment method and device, an electronic device, and a computer-readable storage medium. The method comprises: obtaining segment data for characterizing the state of at least one target segment included in the route; calculating a first safety probability of the target segment using a safety calculation model based on driving behavior according to the segment data; and determining a safety assessment result of the route according to the calculation result of the first safety probability of the target segment. The embodiment of the present application calculates safety by introducing driving behavior data, so that compared with the solution of using only static road data to calculate and evaluate road safety, it is possible to calculate safety by considering the actual driving behavior on the road, so that a more suitable model can be used to calculate safety, thereby improving the accuracy of safety calculation.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a route safety assessment method and device, an electronic device, and a computer-readable storage medium. Background Art

[0002] With the popularization and widespread application of smart devices, more and more people choose devices with navigation functions as auxiliary tools when traveling. When they can't find the destination or are unfamiliar with the road conditions, they can use the navigation function to plan a more reasonable travel route to reach the destination smoothly.

[0003] Existing navigation devices or related applications generally plan one or more travel routes for users to choose based on the user's starting location, end location, and real-time road conditions. In at least one navigation planning route provided to the user, the user is usually provided with information such as the total mileage, estimated driving time, and number of traffic lights along the route for reference.

[0004] However, in the prior art, it is impossible to provide the user with the driving safety risk level of each path, that is, the user cannot know the risk of driving on each path, and it is impossible to prevent accidents. Summary of the invention

[0005] The embodiments of the present application provide a route safety assessment method and device, an electronic device, and a computer-readable storage medium to address the defect in the prior art that a user cannot know the safety of a route.

[0006] To achieve the above objectives, the present application provides a route safety assessment method, including:

[0007] Acquire segment data for characterizing a state of at least one target segment included in the route;

[0008] Calculating a first safety probability of the target road section using a safety calculation model based on driving behavior according to the road section data;

[0009] A safety assessment result of the route is determined according to a calculation result of the first safety probability of the target road section.

[0010] The present application also provides a route safety assessment device, including:

[0011] An acquisition module, used for acquiring segment data for characterizing a state of at least one target segment included in the route;

[0012] A first calculation module, configured to calculate a first safety probability of the target road section according to the road section data using a safety calculation model based on driving behavior;

[0013] The first determination module is used to determine the safety assessment result of the route according to the calculation result of the first safety probability of the target road section.

[0014] The present application also provides an electronic device, including:

[0015] Memory, used to store programs;

[0016] The processor is used to run the program stored in the memory, and the route safety assessment method provided in the embodiment of the present application is executed when the program is run.

[0017] The embodiment of the present application also provides a computer-readable storage medium, on which a computer program executable by a processor is stored, wherein the program, when executed by the processor, implements the route safety assessment method provided in the embodiment of the present application.

[0018] The route safety assessment method and device, electronic device and computer-readable storage medium provided in the embodiments of the present application can calculate the safety probability of each road section based on the road section information that characterizes the status of each road section in the route planned for the user using a driving behavior-based safety calculation model, and then determine the overall safety assessment result of the route according to the safety probability of each road section, thereby calculating safety by introducing driving behavior data, and thus, compared with the solution of using only static road data to calculate and evaluate road safety, it is possible to calculate safety by considering the actual driving behavior on the road, so that a more suitable model can be used to calculate safety, thereby improving the accuracy of safety calculation.

[0019] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present application. Also, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0021] Figure 1 A schematic diagram of an application scenario of the route safety assessment method provided in an embodiment of the present application;

[0022] Figure 2 A flowchart of an embodiment of a route safety assessment method provided by the present application;

[0023] Figure 3 A flowchart of another embodiment of the route safety assessment method provided by the present application;

[0024] Figure 4a A schematic diagram of the structure of an embodiment of a route safety assessment device provided in this application;

[0025] Figure 4b This is an application example of the security assessment solution according to an embodiment of the present application;

[0026] Figure 5 A schematic diagram of the structure of an electronic device embodiment provided in this application. DETAILED DESCRIPTION

[0027] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0028] Embodiment 1

[0029] The solution provided in the embodiment of the present application can be applied to any route safety assessment system with data processing capabilities. Figure 1 A schematic diagram of an application scenario of the route safety assessment method provided in an embodiment of the present application, Figure 1 The scenario shown is only one example of a scenario to which the technical solution of the present application can be applied.

[0030] With the popularization and widespread application of smart devices, more and more people choose devices with navigation functions as auxiliary tools when traveling. When they can't find the destination or are unfamiliar with the road conditions, they can use the navigation function to plan a more reasonable travel route to reach the destination smoothly.

[0031] Existing navigation devices or related applications generally plan one or more travel routes for users to choose based on the user's starting location, end location, and real-time road conditions. In at least one navigation planning path provided to the user, the user is usually provided with information such as the total mileage, estimated driving time, and number of traffic lights along the way for reference. However, in the prior art, it is impossible to provide users with the driving safety risk level of each path, that is, users cannot know the risks of driving on each path, and cannot prevent accidents.

[0032] More and more users choose to use travel assistance tools when traveling, such as navigation software or built-in driving assistance software in the vehicle to help users determine the route planning from the user's current departure place or the designated departure place to the user's designated destination. At present, in the prior art, multiple route planning can be provided to users for selection based on the different sections of the road from the departure place to the destination, so that users can choose the route planning with the lowest fee or the shortest time. With the development of Internet technology, travel assistance tools in the prior art can further provide users with road congestion information to facilitate users to choose routes. However, in the prior art, when choosing a route, users can only make decisions based on the type of roads, toll conditions and time conditions provided by these travel assistance tools. For users, travel safety is also a very important factor that needs to be considered. However, in the current prior art, there is no technical solution to provide users with safety risk assessment of planned routes.

[0033] For example, when a user drives a vehicle, the environmental conditions on the route, such as weather, etc., and the road conditions, such as the number of lanes and lighting conditions, will affect the safety of the user's driving the vehicle. As the number of vehicles on the road increases, the driver's driving behavior will also affect the safety of driving on the road. Therefore, the embodiments of the present application can be applied to various travel assistance tools based on data collection, such as navigation software or a vehicle's built-in safe travel assistance program to provide travel suggestions to the user by understanding the road the user is on or even the road ahead, as well as the environmental conditions and the conditions of the vehicle driven by the user.

[0034] For example, Figure 1 As shown in , before traveling, user 1 can send a travel request containing his destination information to the server through his mobile terminal, such as a travel application on a mobile phone. Therefore, the route safety assessment system of the embodiment of the present application can calculate the route planning information from the user's departure point to the destination based on the travel destination information contained in the travel request sent by the user and the user's current location information. For example, Figure 1As shown in , the route planning information may include, for example, three route information from the user's departure place to the destination specified by the user. In this case, the section information of each route can be obtained, that is, which sections each route consists of, and therefore the road information about these sections can be obtained through a data source such as the road information of a road database, and the environmental information when the user is traveling can also be obtained, for example, the weather conditions at that time can be obtained according to the time information when the user issues a route planning request, in particular, if the destination specified by the user is far away, the weather conditions at the destination can be obtained, in particular, the time to arrive at the destination can be calculated based on the current time when the user issues the request and the arrival time calculated according to the planned route, and then the weather condition forecast for the time to arrive at the destination can be obtained. In addition, the user's vehicle information can also be collected by various sensors set on the user's vehicle. Afterwards, the safety evaluation of the three routes planned for the user can be calculated based on the acquired section information, environmental information and vehicle information. For example, a section vector can be generated for each section according to the acquired information, and the safety evaluation result of the route can be obtained by splicing or LSTM neural network calculation based on each generated section vector, and it can be presented to the user. For example, the security risk value score may be displayed on the screen of the user's mobile terminal, or the routes may be sorted according to the security risk value score of each route, and the route with the highest score, for example, may be recommended to the user.

[0035] In addition, with the authorization of the user, with the development of Internet technology, Internet of Things technology, mobile communication and various sensor technologies, it is possible to obtain various driving behavior data of the user when driving the vehicle. In this application, in addition to considering the road section information, environmental information and vehicle information of the vehicle user's travel route, the driving behavior information can be further considered. For example, when training the parameters for calculating the safety risk value of the road section, the user's braking behavior data can be used as verification data of the road safety risk to compare with the road safety risk value calculated based on the road section information, environmental information and vehicle information as described above. For example, if on a certain road section, for vehicles that have traveled on the road section, they all involve braking behaviors of varying degrees every day, or even sudden braking behaviors, it means that the risk index of the road section is high, that is, it is not very safe. In an embodiment of the present application, the result of the safety calculation can be verified based on driving behavior data such as braking behavior, or the safety can be calculated based on driving behavior data, that is, if there are more braking situations in the road section, the safety of the road section is low. In addition, in an embodiment of the present application, the accuracy of the calculated result can also be calculated. If the accuracy does not meet the condition, the road section vector can be recalculated.

[0036] In the embodiment of the present application, driving behavior data can be further generated by various driving behavior models. For example, various driving behavior models can be established using road-related data, so that when evaluating the safety of a specific route, relevant road parameters can be input into the model, so as to use the model to generate the user's simulated driving behavior data, and use it as a supplementary reference data of the user's actual driving behavior together with the safety assessment scheme of the embodiment of the present application to evaluate the safety of the route. Or when evaluating a specific route, some or even all of the roads included in the route do not have the user's actual driving behavior data at the time of evaluation, then such a driving model can also be used to generate simulated driving behavior data for these roads, so as to evaluate the safety of the route, and after the user actually drives through the route, the collected user's actual driving behavior can be fed back to the above model to train the model to improve the accuracy of the model-generated simulated driving behavior data.

[0037] In addition, in the embodiment of the present application, when the user makes a route planning request, the safety assessment scheme of the embodiment of the present application can be used to assess the safety of each section of the user's travel route, and the safety assessment is performed based on the actual driving behavior of other users obtained from the database or the simulated driving behavior generated by the above-mentioned use model, so that the safety assessment result is displayed to the user before the user actually travels using the planned route, so as to prompt the user of the overall safety of the route or provide the user with safety tips for some sections of the route, so that the user pays attention when driving. Or in the embodiment of the present application, when the user actually starts to drive on the planned route, the safety assessment can be performed in real time according to the user's actual driving behavior, and the safety assessment result can be dynamically updated. For example, when the user just starts to drive, the actual driving behavior of other users obtained from the database or the simulated driving behavior generated by the above-mentioned use model is used to perform a safety assessment, and as the user travels on the planned route, the user's actual driving behavior can be collected in real time, so the already calculated evaluation result can be updated, so that the user's actual driving behavior is reflected in the evaluation result, so that the evaluation result can be updated in real time and presented to the user in real time during the user's driving process, so as to provide the user with safety tips that are more in line with the user's driving habits.

[0038] Therefore, by utilizing the safety assessment scheme of the embodiment of the present application, it is possible to calculate the safety probability of each road section based on the road section information that characterizes the status of each road section in the route planned for the user using a driving behavior-based safety calculation model, and then the overall safety assessment result of the route can be determined based on the safety probability of each road section. By introducing driving behavior data to calculate safety, compared with the scheme that only uses static road data to calculate and evaluate road safety, it is possible to consider the actual driving behavior on the road to calculate safety, so that a more suitable model can be used to calculate safety, thereby improving the accuracy of the safety calculation.

[0039] The above embodiments are illustrations of the technical principles and exemplary application frameworks of the embodiments of the present application. The specific technical solutions of the embodiments of the present application are further described in detail below through multiple embodiments.

[0040] Embodiment 2

[0041] Figure 2 This is a flowchart of an embodiment of the route safety assessment method provided by the present application. The execution subject of the method can be various terminals or server devices with data processing capabilities, or it can be a device or chip integrated on these devices. Figure 2 As shown, the route safety assessment method includes the following steps:

[0042] S201, obtaining segment data for characterizing a state of at least one target segment included in a route.

[0043] In an embodiment of the present application, when a route is planned for a user according to a user's travel request, multiple routes from, for example, the user's location or the departure place specified by the user to the destination specified by the user can be obtained, and each route can be composed of multiple sections. In this case, the section information of each route can be obtained, that is, which sections each route consists of, and therefore the section information about these sections can be obtained through a data source of road information such as a road database. The section data may include one or more of the basic section data for characterizing the basic state of the target section, the environmental data for characterizing the environmental state of the target section, and the vehicle type data for characterizing the type of vehicle traveling on the target section. For example, the basic section data such as the length, width and direction of the road can be obtained, and the environmental information when the user travels can also be obtained. For example, the weather conditions at that time can be obtained according to the time information when the user issues a route planning request. In particular, if the destination specified by the user is far away, the weather conditions at the destination can be obtained, and in particular, the time to arrive at the destination can be calculated based on the current time when the user issues the request and the arrival time calculated according to the planned route, and then the weather forecast for the time to arrive at the destination can be obtained. In addition, the user's vehicle information can also be collected through various sensors installed on the user's vehicle.

[0044] S202: Calculate a first safety probability of the target road section using a safety calculation model based on driving behavior according to the road section data.

[0045] After obtaining the segment data of the multiple segments constituting the route in step S201, the safety probabilities of the three routes planned for the user can be calculated based on the obtained segment data. For example, a segment vector can be generated for each segment based on the obtained information, and the safety probability of the segment can be calculated based on each generated segment vector.

[0046] For example, in the embodiment of the present application, the safety calculation model based on driving behavior can be used to calculate the safety probability of each road section. For example, the road section data of each road section can be input into each decision tree of the safety calculation model, so as to determine the safety probability of each road section using a voting algorithm according to the safety probability of each road section calculated by each decision tree.

[0047] For example, in the embodiment of the present application, the safety calculation model may use, for example, a random forest algorithm, and may also use various probability calculation models that can use driving behavior data, or use a combination of these models for calculation.

[0048] S203: Determine a safety assessment result of the route according to the calculation result of the first safety probability of the target road section.

[0049] After the calculation result of the safety probability of each road section is obtained in step S202, the safety evaluation result of the route as a whole can be determined by summarizing the safety probability results of each road section. For example, the safety probabilities of the road sections can be averaged and the average value can be taken as the safety evaluation result of the route.

[0050] Therefore, by utilizing the safety assessment scheme of the embodiment of the present application, it is possible to calculate the safety probability of each road section based on the road section information that characterizes the status of each road section in the route planned for the user using a driving behavior-based safety calculation model, and then the overall safety assessment result of the route can be determined based on the safety probability of each road section. By introducing driving behavior data to calculate safety, compared with the scheme that only uses static road data to calculate and evaluate road safety, it is possible to consider the actual driving behavior on the road to calculate safety, so that a more suitable model can be used to calculate safety, thereby improving the accuracy of the safety calculation.

[0051] Embodiment 3

[0052] Figure 3 This is a flowchart of another embodiment of the route safety assessment method provided by the present application. The execution subject of the method can be various terminals or server devices with data processing capabilities, or it can be a device or chip integrated on these devices. Figure 3 As shown, the route safety assessment method includes the following steps:

[0053] S301, receiving a route planning request.

[0054] S302: Calculate a route according to the route planning request.

[0055] In the present application, a user can input a travel request through his mobile terminal according to his travel needs. For example, the user can use voice to issue a voice command "I want to go to xx" to the terminal, so that the user's mobile terminal can trigger an application installed on the mobile terminal according to the voice command, such as a navigation application, so that the route planning request can be received in step S301. In particular, the route planning request can include at least the destination information of the route, and may also include the location information and / or time information of the user's location when the request is issued.

[0056] After receiving the request, a route may be calculated for the user according to the route planning request in step S302. For example, multiple routes may be planned for the user as candidates according to the types and quantities of the sections passed through.

[0057] S303: Acquire the segment data representing the status of each target segment included in the route according to the segment data selection condition.

[0058] In the embodiment of the present application, when the user's travel request is received in step S301 and multiple routes from, for example, the user's location or the departure place specified by the user to the destination specified by the user are obtained in step S302. In this case, the section information of each route can be obtained in step S303, that is, which sections each route consists of, and therefore the section information about these sections can be obtained through a data source of road information such as a road database. The section data may include one or more of the basic section data for characterizing the basic state of the target section, the environmental data for characterizing the environmental state of the target section, and the vehicle type data for characterizing the type of vehicle traveling on the target section. For example, the basic section data such as the length, width and direction of the road can be obtained, and the environmental information when the user travels can also be obtained. For example, the weather conditions at that time can be obtained according to the time information when the user issues a route planning request. In particular, if the destination specified by the user is far away, the weather conditions at the destination can be obtained, and in particular, the time to arrive at the destination can be calculated based on the current time when the user issues the request and the arrival time calculated according to the planned route, and then the weather condition forecast at the time of arrival at the destination can be obtained. In addition, the user's vehicle information can also be collected through various sensors installed on the user's vehicle.

[0059] S304: for each target road section included in the route, a plurality of decision trees in the safety calculation model are applied to calculate the safety classification probability of each target road section based on the road section data of each target road section.

[0060] After obtaining the section data of the multiple sections constituting the route in step S303, the safety probabilities of the three routes planned for the user can be calculated based on the obtained section data. For example, the safety probability of each section can be calculated based on the section data using a classification model such as a random forest algorithm as the safety calculation model of the present application. Specifically, for example, the section data of each section can be input into each decision tree of the safety calculation model, so that the classification probability results given by multiple decision trees can be obtained for each section.

[0061] For example, in an embodiment of the present application, the safety calculation model may not be limited to the above-mentioned random forest algorithm, but may use various probability calculation models that can use driving behavior data, or use a combination of these models for calculation.

[0062] S305 , performing voting selection processing according to the safety classification probability of each target road section calculated by each decision tree to determine the first safety probability of each target road section.

[0063] After obtaining the classification probability results given by multiple decision trees for each road section in step S304, the safety probability of each road section can be determined by using a voting algorithm according to the safety probability of each road section calculated by each decision tree in step S305.

[0064] S306: Acquire driving behavior data of the target road section as training sample data.

[0065] After the first safety probability of each target road section is obtained in step S305, various driving behavior data of the user driving the vehicle on the road section can be further obtained. In an embodiment of the present application, the driving behavior data may include safety behavior data for characterizing the safe driving behavior of the vehicle traveling on the target road section and dangerous behavior data for characterizing the dangerous driving behavior of the vehicle. Such driving behavior data can reflect the safety of the road section. Therefore, in the present application, the driving behavior data on the target road section can be used as training sample data to verify the first safety probability calculated in step S305. On the one hand, this can verify the accuracy of the calculation, and on the other hand, the training sample data can also be used to further improve the parameters of the safety calculation model or other models such as the random forest model using a decision tree in step S304.

[0066] S307: Calculate a second safety probability of the target road section according to the driving behavior data.

[0067] In the embodiment of the present application, as described above, the driving behavior data can reflect the safety of the road section that the driver passes through. For example, if on a certain road section, for vehicles that have traveled on the road section, they all involve braking behaviors of varying degrees every day, or even sudden braking behaviors, it means that the risk index of the road section is high, that is, it is not very safe. Therefore, in the present application, the safety probability of the target road section can be calculated based on the training sample data obtained in step S306, as a reference for verifying the first safety probability calculated in step S305.

[0068] S308: Perform regression calculation on the safety calculation model using the first safety probability and the second safety probability.

[0069] In the embodiment of the present application, as described above, the result of the safety calculation can be verified based on driving behavior data such as braking behavior, or the safety can be calculated based on the driving behavior data, that is, if there are more braking situations in the road section, the safety of the road section is lower. Therefore, in step S308, the first safety probability of each road section calculated in step S305 can be compared with the second safety probability obtained based on the driving behavior data in step S307 to verify the calculation result in step S305. And the comparison result can be applied to the regression model to further optimize the parameters of the safety calculation model used in step S305.

[0070] Therefore, since the previous comparison result can be used through the regression model during the model training process, the training effect can be improved and the calculation accuracy can be improved through the iterative calculation of the regression model.

[0071] In addition, the accuracy check may also be performed on the model used in the calculation in step S305, and in particular, the accuracy check calculation may be performed when the model is trained. For example, the route safety assessment method of the present application may further include:

[0072] S309, obtaining a sample result of a safety assessment of the target road section.

[0073] In an embodiment of the present application, a safety assessment result of a training sample can be obtained based on a training sample of a target road section, and the result can be used as a supervisory reference during model training.

[0074] S310, calculating the calculation result accuracy of the safety calculation model according to the first safety probabilities of the plurality of target road sections and the safety evaluation sample results.

[0075] S311, determining a security calculation model according to the accuracy of the calculation result.

[0076] In the present application, the calculation accuracy of the model can be calculated based on the first safety probability of the target road section calculated in step S305 and the sample results of the training samples. For example, the calculation accuracy of the model can be evaluated by calculating the mean absolute error (MAE), the mean square error (MSE) or the mean absolute percentage error (MAPE). In other words, in the present application, the accuracy of the model can be evaluated by inputting multiple samples during training, calculating these samples through the model and comparing the calculation results with the sample results of the samples, that is, the true value, to perform accuracy calculation, so that a model with higher accuracy or precision can be selected for actual calculations.

[0077] S312: Determine a safety assessment result of the route according to a calculation result of a first safety probability of the target road section.

[0078] After the calculation result of the safety probability of each road section is obtained in step S305, the safety evaluation result of the route as a whole can be determined by summarizing the safety probability results of each road section. For example, the safety probabilities of the road sections can be averaged and the average value can be taken as the safety evaluation result of the route.

[0079] Therefore, by utilizing the safety assessment scheme of the embodiment of the present application, it is possible to calculate the safety probability of each road section based on the road section information that characterizes the status of each road section in the route planned for the user using a driving behavior-based safety calculation model, and then the overall safety assessment result of the route can be determined based on the safety probability of each road section. By introducing driving behavior data to calculate safety, compared with the scheme that only uses static road data to calculate and evaluate road safety, it is possible to consider the actual driving behavior on the road to calculate safety, so that a more suitable model can be used to calculate safety, thereby improving the accuracy of the safety calculation.

[0080] Embodiment 4

[0081] Figure 4a The schematic diagram of the structure of the route safety assessment device embodiment provided in the present application can be used to perform the following steps: Figure 2 and Figure 3 The method steps shown. Figure 4a As shown, the route safety assessment device may include: an acquisition module 41 , a first calculation module 42 and a first determination module 43 .

[0082] The acquisition module 41 may be used to acquire segment data for characterizing a state of at least one target segment included in the route.

[0083] In an embodiment of the present application, when a route is planned for a user according to a travel request of the user, the acquisition module 41 can acquire multiple routes from, for example, the user's location or the departure place specified by the user to the destination specified by the user, and each route can be composed of multiple sections. In this case, the acquisition module 41 can acquire the section information of each route, that is, which sections each route is composed of, and thus the section information about these sections can be acquired through a data source of road information such as a road database. The section data may include one or more of the section basic data for characterizing the basic state of the target section, the environmental data for characterizing the environmental state of the target section, and the vehicle type data for characterizing the type of vehicle traveling on the target section. In this case, the acquisition module 41 may include a selection condition acquisition unit 411 and a section data acquisition unit 412. The selection condition acquisition unit 411 may be used to acquire the selection condition according to the section data. And the section data acquisition unit 412 may be used to acquire the section data representing the state of each target section included in the route according to the selection condition.

[0084] For example, the user can specify the road section data selection conditions through the selection condition acquisition unit 411, or the selection condition acquisition unit 411 can also obtain the selection conditions by itself according to various information input by the user. The road section data acquisition unit 412 can obtain the basic road section data such as the length, width and direction of the road according to the selection conditions obtained by the selection condition acquisition unit 411, and can also obtain the environmental information when the user is traveling. For example, the acquisition module 41 can obtain the weather conditions at that time according to the time information when the user sends the route planning request. In particular, if the destination specified by the user is far away, the weather conditions at the destination can be obtained, especially based on the current time when the user sends the request and the arrival time calculated according to the planned route. The time to arrive at the destination can be calculated, and then the weather condition forecast at the time of arrival at the destination can be obtained. In addition, the user's vehicle information can also be collected through various sensors set on the user's vehicle.

[0085] The first calculation module 42 may be configured to calculate a first safety probability of the target road section according to the road section data using a safety calculation model based on driving behavior.

[0086] After the acquisition module 41 acquires the segment data of the multiple segments constituting the route, the first calculation module 42 can calculate the safety probability of the three routes planned for the user based on the acquired segment data. For example, a segment vector can be generated for each segment based on the acquired information, and the safety probability of the segment can be calculated based on each generated segment vector.

[0087] For example, in an embodiment of the present application, the first calculation module 42 may include a classification calculation unit 421 and a probability determination unit 422. The classification calculation unit 421 may be used to apply a plurality of decision trees in the safety calculation model based on driving behavior to each target road section included in the route, respectively, to calculate the safety classification probability of each target road section based on each road section data of the target road section. And the probability determination unit 422 may be used to perform voting selection processing according to the safety classification probability of each target road section calculated by each decision tree, so as to determine the first safety probability of each target road section.

[0088] For example, the classification calculation unit 421 can use the classification model of the safety algorithm to calculate the safety probability of each road section. Specifically, for example, the road section data of each road section can be input into each decision tree of the safety calculation model such as the random forest model, so that the probability determination unit 422 can use the voting algorithm to determine the safety probability of each road section based on the safety probability of each road section calculated by each decision tree.

[0089] For example, in an embodiment of the present application, the safety calculation model may not be limited to the above-mentioned random forest algorithm, but may use various probability calculation models that can use driving behavior data, or use a combination of these models for calculation.

[0090] The first determination module 43 may be configured to determine a safety assessment result of the route according to a calculation result of a first safety probability of the target road section.

[0091] After the first calculation module 42 obtains the calculation result of the safety probability of each road section, the first determination module 43 can determine the safety assessment result of the route as a whole by summarizing the safety probability results of each road section. For example, the safety probabilities of the road sections can be averaged and the average value is taken as the safety assessment result of the route.

[0092] In addition, the route safety assessment device of the present application may further include: a receiving module 44 and a third calculation module 45 .

[0093] The receiving module 44 can be used to receive a route planning request, and the third calculation module 45 can be used to calculate the route according to the route planning request. In the present application, the user can input a travel request through his mobile terminal according to his travel needs. For example, the user can use voice to issue a voice command "I want to go to xx" to the terminal, so that the user's mobile terminal can trigger an application installed on the mobile terminal according to the voice command, such as a navigation application, so that the receiving module 44 can receive the route planning request. In particular, the route planning request can at least include the destination information of the route, and can also include the location information and / or time information of the location where the user issued the request.

[0094] After receiving the request, the third calculation module 45 can calculate a route for the user according to the route planning request. For example, multiple routes can be planned for the user as candidates according to the types and quantities of the sections passed through.

[0095] In addition, after the first calculation module 42 obtains the first safety probability of each target road section, the acquisition module 41 can further obtain the driving behavior data of the target road section as training sample data.

[0096] In an embodiment of the present application, the driving behavior data may include safety behavior data for characterizing the safe driving behavior of the vehicle traveling on the target road section and dangerous behavior data for characterizing the dangerous driving behavior of the vehicle. Such driving behavior data can reflect the safety of the road section. Therefore, in the present application, the driving behavior data on the target road section can be used as training sample data to verify the first safety probability calculated by the first calculation module 42. This can verify the accuracy of the calculation on the one hand, and on the other hand, the training sample data can also be used to further improve the parameters of the random forest model or other models using the decision tree.

[0097] Therefore, the route safety assessment device of the present application may further include: a second calculation module 46 and a regression calculation module 47 .

[0098] The second calculation module 46 can be used to calculate the second safety probability of the target road section according to the driving behavior data, and the regression calculation module 47 can be used to perform regression calculation on the safety calculation model using the first safety probability and the second safety probability. Therefore, since the last comparison result can be used by the regression model during the model training process, the training effect can be improved and the calculation accuracy can be improved by iterative calculation of the regression model.

[0099] In the embodiment of the present application, as described above, the driving behavior data can reflect the safety of the road section that the driver passes through. For example, if on a certain road section, for vehicles that have traveled on the road section, they all involve braking behaviors of varying degrees every day, or even sudden braking behaviors, it means that the risk index of the road section is high, that is, it is not very safe. Therefore, in the present application, the safety probability of the target road section can be calculated based on the training sample data obtained by the acquisition module 41, as a reference for verifying the first safety probability calculated by the first calculation module 42.

[0100] In the embodiment of the present application, as described above, the result of the safety calculation can be verified based on driving behavior data such as braking behavior, or the safety can be calculated based on the driving behavior data, that is, if there are more braking situations in the road section, the safety of the road section is lower. Therefore, the regression calculation module 47 can compare the first safety probability of each road section calculated by the first calculation module 42 with the second safety probability obtained by the second calculation module 46 based on the driving behavior data to verify the calculation result of the first calculation module 42. And the comparison result can be applied to the regression model to further optimize the parameters of the safety calculation model used by the first calculation module 42.

[0101] Therefore, since the previous comparison result can be used through the regression model during the model training process, the training effect can be improved and the calculation accuracy can be improved through the iterative calculation of the regression model.

[0102] In addition, the accuracy check can also be performed on the model used in the calculation of the first calculation module 42, and in particular, the accuracy check calculation can be performed when the model is trained. For example, the route safety assessment device of the present application can further include: an accuracy calculation module 48 and a second determination module 49. And the acquisition module 41 can further acquire the safety assessment sample results of the target road section.

[0103] The accuracy calculation module 48 may be configured to calculate the accuracy of the calculation result of the random forest model according to the first safety probabilities of the plurality of target road sections and the safety assessment sample results.

[0104] The second determination module 49 may be configured to determine the random forest model according to the accuracy of the calculation result.

[0105] In the embodiment of the present application, the acquisition module 41 can acquire the safety assessment results of the training samples based on the training samples of the target road section, and the results can be used as a supervisory reference during model training.

[0106] In the present application, the calculation accuracy of the model can be calculated based on the first safety probability of the target road section calculated by the first calculation module 42 and the sample results of the training samples. For example, the calculation accuracy of the model can be evaluated by calculating the mean absolute error (MAE), the mean square error (MSE) or the mean absolute percentage error (MAPE). In other words, in the present application, the accuracy of the model can be evaluated by inputting multiple samples during training, calculating these samples through the model and comparing the calculation results with the sample results of the samples, that is, the true value, to perform accuracy calculation, so that a model with higher accuracy or precision can be selected for actual calculations.

[0107] Figure 4b This is an application example of the security assessment solution according to the embodiment of the present application. Figure 4b As shown in , the safety assessment scheme of the present application can use a model trained by machine learning to assess safety. For example, the road range and time period for data acquisition can be set first. For example, the time period can be set to April 2019 to July 2019, and the road range can be set to City 1 and City 2, so that the sample data of roads, environments, vehicles and behaviors for generating training samples within the set time period can be obtained according to the set time period and road range, thereby generating a full data training sample set for machine learning. For example, Figure 4bAs shown in , the training sample set may include field classifications such as environment, car, etc., and may have corresponding descriptions such as city, wind speed, ultraviolet intensity, etc. In addition, the sample set may include result data, for example, result data of whether the braking behavior is severe.

[0108] After the data training sample set is generated, on the one hand, it can be applied to a classification model based on a random forest algorithm, for example, to determine whether the label of the predicted classification result is equal to 0, and, on the other hand, it can be determined whether the mean severity of dangerous driving behaviors is less than or equal to a preset threshold, such as 2, based on the set time period and driving behavior occurrence data such as sudden braking within the road range. According to the judgment result, a training sub-sample set representing different driving behaviors can be generated. In an embodiment of the present application, the training sub-samples representing different driving behaviors generated in this way can be applied to a regression model to determine whether the machine learning algorithm meets the model accuracy evaluation index requirements, and only when the index requirements are met, the machine learning model based on the combination of random forest classification and regression can be determined.

[0109] Therefore, after determining a machine learning model based on, for example, a combination of random forest classification and regression, multiple planned paths can be generated according to the navigation request newly initiated by the user, and different planned paths and related variables generated by the newly initiated navigation request can be obtained. Thus, the planned paths and related variables can be given for evaluation, and the driving risk assessment calculation results of each road section included in each path can be output. Of course, in the embodiment of the present application, a machine learning model combining other safety calculation models with regression models can also be used.

[0110] For example, Figure 4b As shown in , the evaluation indicators used to evaluate driving risks may include mean absolute error (MAE), mean square error (MSE), mean absolute percentage error (MAPE), and degree of fit (R^2). After the accuracy requirements are met, a machine learning model based on a safety calculation model and a regression combination is determined; when a new navigation request is initiated to generate different planned paths, the relevant variables used for machine learning model input for each path are obtained, and the generated model is used to complete the driving risk assessment calculation for each road section included in each path; finally, the overall calculation results of the driving risk assessment for different navigation paths are output.

[0111] In addition, based on the RF random forest model, in order to improve the accuracy of effective prediction of dangerous driving behavior, in an embodiment of the present application, the data and the model can be divided into two parts, and calculations based on the random forest algorithm and predictions based on the regression combination model are performed respectively.

[0112] like Figure 4bAs shown in , multiple path plans are generated according to the navigation request from location 1 to location 2 initiated by the user through the above data training method. For example, three routes can be generated in total and the above scheme can be used. For example, the machine learning model trained in the above manner is used to calculate the safety of these path plans, and the calculated estimated risk values ​​are 0.996, 0.974 and 0.960 respectively.

[0113] Therefore, by utilizing the safety assessment scheme of the embodiment of the present application, it is possible to calculate the safety probability of each road section based on the road section information that characterizes the status of each road section in the route planned for the user using a driving behavior-based safety calculation model, and then the overall safety assessment result of the route can be determined based on the safety probability of each road section. By introducing driving behavior data to calculate safety, compared with the scheme that only uses static road data to calculate and evaluate road safety, it is possible to consider the actual driving behavior on the road to calculate safety, so that a more suitable model can be used to calculate safety, thereby improving the accuracy of the safety calculation.

[0114] Embodiment 5

[0115] The internal functions and structure of the text processing device are described above. The device can be implemented as an electronic device. Figure 5 This is a schematic diagram of the structure of an electronic device embodiment provided by this application. Figure 5 As shown, the electronic device includes a memory 51 and a processor 52 .

[0116] The memory 51 is used to store programs. In addition to the above programs, the memory 51 can also be configured to store various other data to support operations on the electronic device. Examples of such data include instructions for any application or method operating on the electronic device, contact data, phone book data, messages, pictures, videos, etc.

[0117] The memory 51 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0118] The processor 52 is not limited to a central processing unit (CPU), but may also be a processing chip such as a graphics processing unit (GPU), a field programmable gate array (FPGA), an embedded neural network processor (NPU) or an artificial intelligence (AI) chip. The processor 52 is coupled to the memory 51 to execute a program stored in the memory 51. When the program is executed, the route safety assessment method of the above-mentioned embodiments 2 and 3 is executed.

[0119] Further, if Figure 5 As shown, the electronic device may also include: a communication component 53, a power component 54, an audio component 55, a display 56 and other components. Figure 5 Only some components are shown schematically, which does not mean that the electronic device only includes Figure 5 Components shown.

[0120] The communication component 53 is configured to facilitate wired or wireless communication between the electronic device and other devices. The electronic device can access a wireless network based on a communication standard, such as WiFi, 3G, 4G or 5G, or a combination thereof. In an exemplary embodiment, the communication component 53 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 53 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0121] The power supply component 54 provides power to various components of the electronic device. The power supply component 54 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the electronic device.

[0122] The audio component 55 is configured to output and / or input audio signals. For example, the audio component 55 includes a microphone (MIC), and when the electronic device is in an operating mode, such as a call mode, a recording mode, and a voice recognition mode, the microphone is configured to receive an external audio signal. The received audio signal can be further stored in the memory 51 or sent via the communication component 53. In some embodiments, the audio component 55 also includes a speaker for outputting audio signals.

[0123] The display 56 includes a screen, which may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from a user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundary of a touch or slide action, but also detect the duration and pressure associated with the touch or slide operation.

[0124] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, disk or optical disk and other media that can store program codes.

[0125] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A route safety assessment method, comprising: Acquire segment data for characterizing a state of at least one target segment included in the route; Calculating a first safety probability of the target road section using a safety calculation model based on driving behavior according to the road section data; Determining a safety assessment result of the route according to a calculation result of a first safety probability of the target road section; The method further includes: acquiring driving behavior data of the target road section as training sample data, wherein the driving behavior data includes safe behavior data for characterizing safe driving behavior of a vehicle traveling on the target road section and dangerous behavior data for characterizing dangerous driving behavior of the vehicle, the driving behavior data includes braking behavior data, and the driving behavior data also includes simulated driving behavior data generated using a driving behavior model; Calculating a second safety probability of the target road section according to the driving behavior data; and The first safety probability and the second safety probability are used to perform regression calculation on the safety calculation model to optimize parameters of the safety calculation model.

2. The route safety assessment method according to claim 1, wherein: The acquiring of the segment data for characterizing the state of at least one target segment included in the route comprises: According to the segment data selection conditions, segment data characterizing the status of each target segment included in the route is obtained, wherein the segment data selection conditions at least include one or more of the geographical scope of the segment, the time of data selection and the road type of the segment.

3. The route safety assessment method according to claim 1, wherein: The road section data includes: One or more of basic road segment data for characterizing the basic state of the target road segment, environmental data for characterizing the environmental state of the target road segment, and vehicle type data for characterizing the type of vehicles traveling on the target road segment.

4. The route safety assessment method according to claim 1, wherein: The step of calculating the first safety probability of the target road section according to the road section data using a safety calculation model based on driving behavior includes: For each target road section included in the route, a plurality of decision trees in the safety calculation model are respectively applied to calculate the safety classification probability of each target road section based on each road section data of the target road section; Voting selection processing is performed on the safety classification probabilities of the target road sections calculated by the decision trees to determine the first safety probability of the target road sections.

5. The route safety assessment method according to claim 4, wherein: The method further comprises: Obtaining a sample result of a safety assessment of the target road section; Calculating the calculation result accuracy of the safety calculation model according to the first safety probabilities of the plurality of target road sections and the safety evaluation sample results; The safety calculation model is determined according to the accuracy of the calculation result.

6. The route safety assessment method according to claim 1, wherein: The method further comprises: receiving a route planning request, wherein the route planning request includes at least destination information of the route; The route is calculated according to the route planning request.

7. A route safety assessment device, wherein: include: An acquisition module, used for acquiring segment data for characterizing a state of at least one target segment included in the route; It is also used to obtain driving behavior data of the target road section as training sample data, wherein the driving behavior data includes safe behavior data for characterizing the safe driving behavior of the vehicle traveling on the target road section and dangerous behavior data for characterizing the dangerous driving behavior of the vehicle, the driving behavior data includes braking behavior data, and the driving behavior data also includes simulated driving behavior data generated using a driving behavior model; A first calculation module, configured to calculate a first safety probability of the target road section according to the road section data using a safety calculation model based on driving behavior; A first determination module, configured to determine a safety assessment result of the route according to a calculation result of a first safety probability of the target road section; A second calculation module is used to calculate a second safety probability of the target road section according to the driving behavior data; and A regression calculation module is used to perform regression calculation on the safety calculation model using the first safety probability and the second safety probability to optimize the parameters of the safety calculation model.

8. An electronic device comprising: Memory, used to store programs; A processor is used to run the program stored in the memory, and the program executes the route safety assessment method as described in any one of claims 1 to 6 when running.

9. A computer-readable storage medium having stored thereon a computer program executable by a processor, wherein: When the program is executed by a processor, the route safety assessment method as described in any one of claims 1 to 6 is implemented.

Images