Traffic light scene road condition recognition method, device, electronic device and storage medium

Abstract

The present disclosure provides a method, device, electronic device and storage medium for identifying road conditions in a traffic light scene, and relates to technical fields such as intelligent transportation and autonomous driving. The specific implementation scheme is: collecting driving information of vehicles passing through traffic lights on a specified turn in each time window of a plurality of consecutive time windows before the current moment; obtaining the driving characteristics of vehicles in each time window on the specified turn based on the driving information of vehicles passing through traffic lights on the specified turn in each time window; predicting the road conditions of the road affected by the traffic light at the current moment and on the specified turn based on the driving characteristics of vehicles in each time window on the specified turn; identifying the road conditions corresponding to each road section mark at the current moment and on the specified turn based on the road conditions affecting the road at the current moment and on the specified turn and the road section marks corresponding to the affected roads. The technology disclosed in the present disclosure can effectively improve the accuracy of the road conditions of the road sections identified in the traffic light scene.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, specifically to the technical fields of intelligent transportation, automatic driving, etc., and more particularly to a method, device, electronic device and storage medium for identifying road conditions in a traffic light scene. Background Art

[0002] When users are driving and navigating, map products will comprehensively consider the distance, estimated arrival time, etc. when planning routes for users, and plan the best driving route for users. The estimated arrival time depends on the congestion of the road. Therefore, road condition recognition is crucial.

[0003] Since vehicle speed limits are different for different road grades, the road condition of a road section can be estimated based on the road grade and the statistical characteristics of the speeds of all vehicles passing on the road section. Summary of the invention

[0004] The present invention provides a method, device, electronic device and storage medium for identifying road conditions in a traffic light scene.

[0005] According to one aspect of the present disclosure, a method for identifying a road condition in a traffic light scene is provided, comprising:

[0006] Collecting driving information of vehicles passing through traffic lights on a specified turn in each time window of a plurality of consecutive time windows before the current moment;

[0007] Based on the driving information of vehicles passing through the traffic light in each of the time windows on the designated turn, obtaining the driving characteristics of vehicles in each of the time windows on the designated turn;

[0008] Based on the vehicle driving characteristics in each of the multiple time windows on the designated turn, predict the road condition affecting the traffic light on the designated turn at the current moment;

[0009] Based on the traffic conditions of the road affected by the traffic light at the current moment and the designated turn and the road segment identification corresponding to the affected road, the traffic conditions corresponding to each of the road segment identifications at the current moment and the designated turn are identified.

[0010] According to another aspect of the present disclosure, a method for training a road condition prediction model is provided, comprising:

[0011] Collecting driving information of vehicles passing through traffic lights at each turn in each of a plurality of consecutive time windows before each of a plurality of training moments from a historical driving information database;

[0012] Based on the driving information of vehicles passing through the traffic lights in each of the time windows on each of the turns and before each of the training moments, the driving characteristics of vehicles in each of the time windows on each of the turns and before each of the training moments are obtained;

[0013] Based on the driving information database, obtaining the real road conditions of the roads affected by the traffic lights at each turning point and each training moment;

[0014] The road condition prediction model is trained based on the vehicle driving characteristics in each time window at each turn and before each training moment, and the actual road conditions affecting the road at each turn and at each training moment.

[0015] According to another aspect of the present disclosure, a road condition determination device for a traffic light scene is provided, comprising:

[0016] A collection module, used to collect driving information of vehicles passing through the traffic lights on a specified turn in each time window of a plurality of consecutive time windows before the current moment;

[0017] An acquisition module, configured to acquire driving characteristics of vehicles in each of the time windows on the designated turn based on driving information of vehicles passing through the traffic light in each of the time windows on the designated turn;

[0018] A prediction module, configured to predict the road condition affecting the traffic light on the designated turn at the current moment based on the vehicle driving characteristics in each of the multiple time windows on the designated turn;

[0019] The identification module is used to identify the road conditions corresponding to each road section mark at the current moment and the designated turn based on the road conditions affected by the traffic light at the current moment and the designated turn and the road section marks corresponding to the affected road.

[0020] According to another aspect of the present disclosure, a training device for a road condition prediction model is provided, comprising:

[0021] A collection module, used to collect driving information of vehicles passing through traffic lights on each turn in each time window in a plurality of consecutive time windows before each training moment in a plurality of training moments from a historical driving information database;

[0022] A feature acquisition module, for acquiring vehicle driving features in each of the time windows on each of the turns and before each of the training moments based on vehicle driving information passing through the traffic lights in each of the time windows on each of the turns and before each of the training moments;

[0023] A road condition acquisition module, for acquiring the real road condition of the road affected by the traffic lights at each of the turns and at each of the training moments based on the driving information database;

[0024] The training module is used to train the road condition prediction model based on the vehicle driving characteristics in each time window at each turn and before each training moment, and the actual road conditions affecting the road at each turn and at each training moment.

[0025] According to yet another aspect of the present disclosure, there is provided an electronic device, comprising: at least one processor; and

[0026] a memory communicatively connected to the at least one processor; wherein,

[0027] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method of any possible implementation manner and the aspects described above.

[0028] According to yet another aspect of the present disclosure, there is provided a non-transient computer having computer instructions stored therein.

[0029] A computer-readable storage medium, wherein the computer instructions are used to cause the computer to execute the method of the aspects and any possible implementation manner as described above.

[0030] According to yet another aspect of the present disclosure, a computer program product is provided, including a computer program, wherein when the computer program is executed by a processor, the computer program implements the method of the above-mentioned aspects and any possible implementation manner.

[0031] According to the technology disclosed in the present invention, the accuracy of the road conditions of the road sections identified in the traffic light scene can be effectively improved.

[0032] It should be understood that the contents described in this section are not intended to identify the key aspects of the embodiments of the present disclosure.

[0033] Or important features, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following 5 descriptions. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0035] Figure 1 is a schematic diagram according to a first embodiment of the present disclosure;

[0036] Figure 2 is a schematic diagram according to a second embodiment of the present disclosure;

[0037] Figure 3 is a schematic diagram according to a third embodiment of the present disclosure;

[0038] Figure 4 is a schematic diagram according to a fourth embodiment of the present disclosure;

[0039] Figure 5 is a schematic diagram according to a fifth embodiment of the present disclosure;

[0040] Figure 6 is a schematic diagram according to a sixth embodiment of the present disclosure;

[0041] Figure 7 is a schematic diagram according to a seventh embodiment of the present disclosure;

[0042] Figure 8 The block diagram is a block diagram of an electronic device for implementing the method of the embodiment of the present disclosure. DETAILED DESCRIPTION

[0043] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0044] Obviously, the described embodiments are only part of the embodiments of the present disclosure, but not all of them. Based on the embodiments of the present disclosure, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present disclosure.

[0045] It should be noted that the terminal devices involved in the embodiments of the present disclosure may include but are not limited to mobile phones, personal digital assistants (PDAs), wireless handheld devices, tablet computers and other smart devices; display devices may include but are not limited to personal computers, televisions and other devices with display functions.

[0046] In addition, the term "and / or" in this article is only a description of the association relationship between the associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0047] In the prior art, the road condition of a road section is identified based on the speed statistics of all vehicles passing on the road section. When this method is used to identify the road condition of a road section in a traffic light scene, the accuracy is poor. For example, at a traffic light intersection, even if the road section at the traffic light intersection is unobstructed, the average speed of vehicles is very slow because vehicles need to wait for the red light, which in turn causes the road condition of the corresponding road section to be identified as congested. Therefore, the above method of the prior art has very low accuracy in identifying the road condition of a road section in a traffic light scene.

[0048] Figure 1 is a schematic diagram according to the first embodiment of the present disclosure; Figure 1 As shown, this embodiment provides a method for identifying road conditions in a traffic light scene, which can be used in a navigation server and may specifically include the following steps:

[0049] S101, collecting driving information of vehicles passing through a traffic light on a specified turn in each time window of a plurality of consecutive time windows before a current moment;

[0050] S102, based on the driving information of vehicles passing through the traffic lights in each time window on the designated turn, obtaining the driving characteristics of vehicles in each time window on the designated turn;

[0051] S103, predicting the road conditions affecting the traffic lights on the designated turn at the current moment based on the vehicle driving characteristics in each of the multiple time windows on the designated turn;

[0052] S104, based on the road conditions of the road affected by the traffic lights at the designated turn at the current moment and the road section signs corresponding to the affected roads, identifying the road conditions corresponding to the road section signs at the designated turn at the current moment.

[0053] The road section in this embodiment is Link, which is the smallest unit for identifying a road in an electronic map. When the navigation server performs navigation, it is necessary to refer to the road conditions of each road section link, and then accurately plan navigation. The road describes the road on which the vehicle travels from the perspective of vehicle travel. The road can include one, two or more road sections link in series. Moreover, some roads may also include only one section of a road section link. In an electronic map, when identifying the road conditions, a road section link cannot be split again, and there is no different road conditions corresponding to different segments in the same road section link. The purpose of this embodiment is to accurately identify the road conditions of each road section link in a traffic light scene, especially the road conditions of each road section link on the road affected by the traffic light, so as to provide a more accurate navigation service for the navigation application. The road affected by the traffic light refers to a section of road that is directly connected to the traffic light and affects the vehicle passing through the traffic light. The road affected by the traffic light may include one, two or more road sections link.

[0054] The granularity of the current moment can be selected based on the cycle of road condition changes in actual traffic application scenarios. For example, if it is assumed that the road condition can be updated once every minute, the granularity of the current moment can be at the minute level. The current moment refers to the current minute, such as the current moment of 1:10 can refer to the time from 1:10:00 to 59 seconds.

[0055] The multiple continuous time windows may include 5, 8, 10 or other number of time windows. The specific number can be determined based on the historical driving information, the time length of the vehicle driving information before a moment of analysis that can predict the road conditions at that moment, and the size of each time window. For example, after analysis, it is determined that the vehicle driving information 10 minutes before each moment can predict the road conditions at that moment, and the time window is 2 minutes, then 5 continuous time windows can be set. Specifically, in actual applications, the size of each time window can also be set based on experience, for example, it can be 1 minute, it can also be 2 minutes, 90 seconds or other time lengths, which are not limited here.

[0056] In this embodiment, a statistical method can be used to obtain the vehicle driving characteristics of each time window on the designated turn based on the vehicle driving information of the traffic lights in each time window on the designated turn. The designated turn in this embodiment refers to any turn when passing the traffic light, such as straight, left turn or right turn. Since the left turn and the U-turn are usually on the same lane, the U-turn can be included in the left turn in this embodiment. And further refer to the vehicle driving characteristics of each time window in multiple time windows to predict the road conditions of the road affected by the traffic light on the designated turn at the current moment. The continuous multiple time windows can refer to the multiple time window sequences of the nearest neighbors before the current moment, so the vehicle driving characteristics of the continuous multiple time windows have a great impact on the road conditions at the current moment. In this embodiment, the vehicle driving characteristics of each time window in the continuous multiple time windows on the designated turn can be analyzed to accurately predict the road conditions of the road affected by the traffic light on the designated turn at the current moment. Then, combined with the road section identification corresponding to the road affected, the road conditions corresponding to each road section identification on the designated turn at the current moment are identified. Different from the prior art in which each road section corresponds to only one road condition, in this embodiment, the road section corresponding to the affected road may include the road conditions of each turn. Since the affected road is directly adjacent to the traffic light, the impact on vehicles passing through the traffic light is very large, and the impact on vehicles with different turns is also different. By adopting the technical solution of this embodiment, the road conditions of each road section on any specified turn at the current moment can be accurately identified, effectively improving the accuracy of road condition identification of the road section in the traffic light scene, and thus improving the accuracy of navigation planning.

[0057] The road condition recognition method for the traffic light scene of this embodiment can obtain the vehicle driving characteristics of each time window on the specified turn based on the driving information of vehicles passing through the traffic lights in each time window. Since the influence of the traffic lights on the vehicle driving information is fully considered, the accuracy of the obtained vehicle driving characteristics can be effectively ensured. Furthermore, based on the vehicle driving characteristics of each time window in the specified turn and multiple consecutive time windows before the current moment, the road condition of the road affected by the traffic light at the current moment and the specified turn can be accurately predicted; and then based on the road condition of the road affected by the traffic light at the current moment and the specified turn and the road section identification corresponding to the affected road, the road condition corresponding to each road section identification can be accurately identified. The technical solution of this embodiment, since it fully considers the information of the traffic lights, can effectively improve the accuracy of the road condition of the road section in the traffic light scene identified compared with the method of the prior art that does not consider the road condition of the road section identified by the traffic lights.

[0058] Figure 2 is a schematic diagram according to the second embodiment of the present disclosure; the road condition recognition method of the traffic light scene of this embodiment, in the above Figure 1 Based on the technical solutions of the embodiments shown in the figure, the technical solutions of the present disclosure are further described in more detail. Figure 2 As shown, the road condition recognition method for the traffic light scene of this embodiment may specifically include the following steps:

[0059] S201, based on the driving information database, analyzing the road corresponding to the longest distance of vehicles queuing at the traffic light as the influencing road of the traffic light;

[0060] Specifically, the driving information database may record all historical driving information that passes the traffic light. Based on the analysis of features such as vehicle speed, the section of road corresponding to the longest queue distance when the vehicles pass the traffic light may be analyzed as the affected road of the traffic light. In other words, the affected road refers to a section of road that the vehicle travels on before passing the traffic light. The affected road is connected to the traffic light, that is, one endpoint of the affected road is the location of the traffic light, and the other endpoint is a point in the road section. In the analysis of the affected road, the turning problem of the queued vehicles passing through the traffic light is not distinguished, and the road corresponding to the longest queue distance is comprehensively analyzed. Since the intersection of the affected road is greatly affected by the traffic light, in this embodiment, it is necessary to analyze the affected road to improve the accuracy of the road conditions identified in the traffic light scene.

[0061] Moreover, the road affected by the traffic light of this embodiment may include one, two or more road sections link. Therefore, the road affected by the traffic light may correspond to a road section identification sequence, and the road section identification sequence may include identifications of each road section link included in sequence on the affected road from the traffic light to the direction away from the traffic light. It should be noted that the road section corresponding to the last road section link identification in the road section identification sequence, that is, the road section link farthest from the traffic light on the affected road, may be fully included in the affected road, or may be only partially included in the affected road. In addition, the traffic light of this embodiment may refer to a traffic light set at an intersection.

[0062] The length of the affected road can be called the affected length or the affected distance. In practical applications, different traffic lights and different road conditions correspond to different affected roads, and the affected lengths may also be different.

[0063] S202, collecting the number of stops, length of time of passing, and average speed of each vehicle among all vehicles passing through the traffic light on the specified turn in each time window before the current moment;

[0064] Optionally, in an embodiment of the present disclosure, only at least one of the number of stops, the length of time of passing, and the average speed may be collected as required. In practical applications, the more features collected, the more accurate the road conditions of the road section under the traffic light scene are identified. In this embodiment, the simultaneous collection of three features is taken as an example.

[0065] In this embodiment, the number of stops of a vehicle refers to the number of stops of the vehicle from entering the affected road to passing the traffic light. In specific implementation, the number of stops of the vehicle can be analyzed based on the speed of the vehicle during driving. For example, if the speed is 0, it corresponds to 1 stop.

[0066] The passing time refers to the length of time it takes for a vehicle to enter the affected road and pass the traffic light. In specific implementation, the time when the vehicle enters the affected road and the time when it arrives at the traffic light can be obtained from the vehicle's historical driving information, and then the length of time it takes for the vehicle to enter the affected road and pass the traffic light can be obtained.

[0067] The average speed refers to the average speed of the vehicle from entering the affected road to passing the traffic light. At the same time, in specific implementation, it is to count the average speed of the vehicle on the affected road.

[0068] S203, obtaining an average number of stops in each time window on the specified turn based on the number of stops of each vehicle among all vehicles passing through the traffic light in each time window on the specified turn;

[0069] S204, obtaining an average of the length of passing time in each time window on the specified turn based on the length of passing time of each vehicle among all vehicles passing through the traffic light in each time window on the specified turn;

[0070] S205, obtaining an average value of the average speed in each time window on the specified turn based on the average speed of each vehicle among all vehicles passing through the traffic light in each time window on the specified turn;

[0071] S206, constructing vehicle driving characteristics in each time window on the specified turn based on the average number of stops, the average length of passing time and the average speed in each time window on the specified turn;

[0072] In the above manner, the vehicle driving characteristics of each time window on the designated turn are constructed to be very reasonable and accurate.

[0073] S207, using a pre-trained traffic condition prediction model, based on the vehicle driving characteristics in each of the multiple time windows on the designated turn, predicting the traffic condition of the road affected by the traffic light on the designated turn at the current moment;

[0074] Specifically, each time window corresponds to a vehicle driving feature, and a plurality of consecutive time windows constitute a vehicle driving feature sequence, and the vehicle driving feature sequence is arranged in sequence according to the order of the corresponding time windows.

[0075] When in use, the vehicle driving feature sequences corresponding to multiple time windows are input into the road condition prediction model, and the road condition prediction model can predict and output the road condition affecting the traffic lights at the current moment and the specified turn.

[0076] Since the vehicle driving characteristics in the above-mentioned time windows all correspond to the vehicle driving characteristics that affect the road, the road conditions affected by the traffic lights are also predicted here.

[0077] In this embodiment, the road condition prediction model may include road condition prediction sub-models for different turns. When this step is implemented, the vehicle driving characteristics of each time window in multiple time windows of three specified turns can be obtained at the same time, and input into the road condition prediction sub-model of the corresponding turn respectively, so that the road condition of the road affected by the traffic lights at the current moment of each turn can be accurately predicted. Of course, in actual applications, if it is only necessary to obtain the road condition on one turn, only the vehicle driving characteristics of each time window in multiple time windows of the turn are obtained, and input into the road condition prediction sub-model of the turn, so that the road condition of the road affected by the traffic lights at the current moment of the turn can be accurately predicted.

[0078] S208, detecting whether the road condition of the road affected by the traffic light at the current moment and the designated turn is congested, unobstructed or slow-moving; if it is congested, executing step S209; if it is unobstructed, executing step S210; if it is slow-moving, executing step S211;

[0079] S209, determining that the road condition corresponding to the road section identifiers of the road sections affected by the specified turn at the current moment is congested; executing step S212;

[0080] In combination with traffic scenarios in actual applications, when the road condition affecting the road is congested, the road conditions of all sections included in the affected road must be congested.

[0081] The road section corresponding to the influencing road includes two situations, one is that the road section is completely included in the influencing road, and the other is that the road section is partially included in the influencing road.

[0082] In specific implementation, reference can be made to the section identification sequence of the affected roads, wherein the roads corresponding to all section identifications before the last section identification in the section identification sequence are included in the affected roads, and the section corresponding to the last section identification may also be included in the affected roads, or may only be partially included in the affected roads. The specific analysis needs to be carried out based on the specific location of the last section and the specific conditions of all sections covered by the affected roads.

[0083] S210, determining that the road condition corresponding to the road segment identifiers of each road segment affected by the specified turn at the current moment is smooth; end.

[0084] In combination with traffic scenarios in actual applications, when the road condition affecting the road is smooth, the road conditions of all road sections included in the affected road must be smooth.

[0085] S211, based on the impact length of the impacted road, the road section identification corresponding to the impacted road and the preset length ratio, determine the road conditions corresponding to each road section identification on the specified turn at the current moment, and end.

[0086] In combination with traffic scenarios in actual applications, when the road condition affecting the road is slow, the road conditions of the sections included in the road may be different. For example, the road condition of the section near the traffic light may be slow, while the road condition of the section far from the traffic light may be smooth.

[0087] For example, when the step S211 is implemented specifically, it may include the following steps:

[0088] (1) Obtaining a reference length based on the ratio of the impact length to the preset length;

[0089] (2) Detect whether the road section corresponding to the road section mark corresponding to the affected road is included in the road within the reference length range before the traffic light; if so, execute step (3); otherwise, if not included, execute step (4);

[0090] (3) determining that the road condition corresponding to the road section sign on the designated turn at the current time is slow traffic;

[0091] (4) Determine that the road condition corresponding to the road section sign on the specified turn at the current time is smooth.

[0092] In this embodiment, by setting a preset length ratio, it is considered that the road condition on the section of the road near the traffic light and the preset length ratio is still slow; while the road condition on the section of the road far from the traffic light and outside the preset length ratio in the affected road is smooth. The preset length ratio of this embodiment can be set based on experience, for example, it can be 2 / 3, 3 / 5, etc., which is not limited here. Based on this principle, according to the above steps (1)-(4), it is possible to accurately identify the road conditions of each section corresponding to the affected road.

[0093] Furthermore, in an embodiment of the present disclosure, when step S211 is specifically implemented, the following steps may also be included:

[0094] (a) detecting whether only the first part of the road section corresponding to the road section identifier corresponding to the affecting road is included in the road within the reference length range before the traffic light; if so, executing step (b);

[0095] (b) obtaining a first ratio of the length of the first part of the road segment to the road segment corresponding to the road segment identifier; executing step (c);

[0096] (c) detecting whether the first ratio is greater than or equal to a first preset ratio threshold, and if so, executing step (d); otherwise, if the first ratio is less than the first preset ratio threshold, executing step (e);

[0097] (d) determining that the road condition corresponding to the road section sign on the designated turn at the current time is slow traffic;

[0098] (e) Determine that the road condition corresponding to the road section sign on the specified turn at the current time is smooth.

[0099] Combined with the traffic scenarios in actual applications, there is a road section where only part of the first part of the road section is within the reference length range. In this case, further analysis can be performed to determine the first proportion of the length of the first part of the road section to the length of the entire road section link. If the first proportion is large, such as greater than or equal to the first

[0100] The preset ratio threshold can set the road condition of the section to be the same as the road condition of the other five sections within the reference length range, i.e., slow traffic. If the first ratio is smaller, the road condition of the section can be set to be the same as the road condition of the roads on the affected roads and outside the reference length range, i.e., unblocked. The first preset ratio threshold can be set based on experience, such as 55%, 60%, etc. Based on this principle, combined with the above steps (a)-(e), it is possible to accurately identify the road conditions of each section corresponding to the affected road at the current moment and the specified turn.

[0101] 0S212, detect the road section corresponding to the road section mark corresponding to the affected road, whether only the second part of the road

[0102] The segment is included in the affected road; if so, executing step S213;

[0103] S213, obtaining the length of the second part of the road segment and the second part of the road segment corresponding to the road segment identifier.

[0104] ratio; executing step S214;

[0105] S214, detecting whether the second ratio is greater than or equal to a second preset ratio threshold, if so, executing step S215; otherwise, if the second ratio is less than the second preset ratio threshold, executing step S216;

[0106] S215, determining that the traffic condition corresponding to the road section mark on the designated turn at the current moment is congested; end.

[0107] S216. Determine the road condition corresponding to the road section identifier based on the road conditions of other sections except the second part of the road section in the road section corresponding to the road section identifier, and end.

[0108] 0 When the intersection corresponding to the affected road is congested, if only the second part of the road segment in the link includes

[0109] In the affected road, that is, the road section is located at the tail of the affected road away from the traffic light. Other road sections outside the second part of the road section are not within the affected road range, are less affected by the traffic light, and do not need to identify the road conditions by turning. For the prediction of specific road conditions, refer to the prediction method of the road conditions of general road sections, which will not be repeated here.

[0110] 5. When the road conditions of the sections of the affected road are smooth, the sections after the affected road

[0111] The road condition must be smooth. Therefore, the road condition of the last section in the road section identification sequence corresponding to the affecting road, whether it is fully included in the affecting road or partially included in the affecting road, must also be smooth if the road conditions of the adjacent sections before and after it are smooth.

[0112] When the road conditions of the sections included in the affected road are congested, the road conditions of the sections after the affected road may be congested or slow. Since the sections after the affected road are far from the traffic lights, the impact of the traffic lights is not considered, and the prediction method of the road conditions of the general sections is adopted to achieve the prediction. For the section corresponding to the last section identifier in the sequence of section identifiers corresponding to the affected road, if only part of it is included in the affected road, according to steps S212-S216, the road conditions of the last section corresponding to the affected road at the current moment and on the specified turn can be accurately identified.

[0113] The road condition identification method for the traffic light scene of the present embodiment can fully consider the conditions that affect the corresponding road sections under various circumstances based on the road conditions affected by the traffic lights, accurately analyze the road conditions that affect the corresponding road sections, and effectively improve the accuracy of identifying the road conditions of the sections in the traffic light scene, thereby effectively improving the accuracy of navigation planning.

[0114] In actual applications, more than half of map navigation users will encounter traffic lights during the navigation process. On average, each user can pass through more than 20 traffic lights a day. Map navigation users across the country can pass through more than 500 million traffic lights a day. The above-mentioned technical solution disclosed in the present invention can effectively improve the accurate judgment of road conditions in traffic light scenarios, bring users a more comfortable user experience, and enhance users' overall reputation for maps.

[0115] Figure 3 is a schematic diagram according to the third embodiment of the present disclosure; Figure 3 As shown, this embodiment provides a method for training a road condition prediction model, which may specifically include the following steps:

[0116] S301, collecting driving information of vehicles passing through traffic lights in each turn in each of a plurality of consecutive time windows before each of a plurality of training moments from a driving information database;

[0117] S302, based on the driving information of vehicles passing through the traffic lights in each time window on each turn and before each training moment, obtaining the driving characteristics of vehicles in each time window on each turn and before each training moment;

[0118] S303, based on the historical driving information database, obtaining the actual road conditions affecting the traffic lights at each turn and each training time;

[0119] S304: Train a road condition prediction model based on vehicle driving characteristics in each time window at each turn and before each training moment, and actual road conditions affecting the road at each turn and at each training moment.

[0120] In this embodiment, the specific implementation of step S301-step S302 can refer to the above Figure 1 The relevant records of the illustrated embodiment will not be repeated here.

[0121] During training, the vehicle driving characteristics in each time window at each turn and before each training moment are used as input feature data, and the actual road conditions affecting the road at each turn and at each training moment are used as label data to train the road condition prediction model, so that the road condition prediction model learns the ability to predict the road conditions that affect the road.

[0122] In this embodiment, taking the simultaneous collection of data of multiple turns as an example, when training the road condition prediction model, the road condition prediction model is enabled to simultaneously learn the road condition prediction capability of each turn that affects the road.

[0123] In this embodiment, the actual road conditions affecting the road may include congestion, smooth traffic or slow traffic.

[0124] The training method of the traffic condition prediction model of this embodiment can accurately train the traffic condition prediction model by adopting the above method, so that the traffic condition prediction model can learn the ability to predict the traffic conditions that affect the road, and obtain a traffic condition prediction module with high accuracy, thereby effectively improving the accuracy of the traffic conditions of the road section in the traffic light scene in the traffic condition prediction.

[0125] Figure 4 is a schematic diagram according to the fourth embodiment of the present disclosure; this embodiment also provides a method for training a road condition prediction model. Figure 3 Based on the technical solutions of the embodiments shown in the figure, the technical solutions of the present disclosure are further described in more detail. Figure 4 As shown, the training method of the traffic condition prediction model of this embodiment may specifically include the following steps:

[0126] S401, collecting from the driving information database the number of stops, the length of time for passing through, and the average speed of each vehicle among all vehicles passing through the traffic lights at each turn in each time window before each training moment;

[0127] Among them, the number of vehicle stops refers to the number of times the vehicle stops from entering the affected road to passing the traffic light; the passing time refers to the length of time the vehicle consumes from entering the affected road to passing the traffic light; the average speed refers to the average speed of the vehicle from entering the affected road to passing the traffic light.

[0128] Similarly, in one embodiment of the present disclosure, only at least one of the number of stops, the length of time it takes to pass, and the average speed may be collected as required. In practical applications, the more features collected, the more accurate the road conditions of the road section identified in the traffic light scene. In this embodiment, the simultaneous collection of three features is taken as an example.

[0129] S402, based on the number of stops of each vehicle among all vehicles passing through the traffic lights in each time window before each training moment on each turn, obtaining the average number of stops in each time window before each training moment on each turn;

[0130] S403, based on the passing time length of each vehicle among all vehicles passing through the traffic light in each time window before each training moment on each turn, obtaining the average passing time length of each time window before each training moment on each turn;

[0131] S404, based on the average speed of each vehicle among all vehicles passing through the traffic light in each time window before each training moment on each turn, obtain the average value of the average speed in each time window before each training moment on each turn;

[0132] S405, constructing vehicle driving characteristics at each turn and each time window based on the average number of stops, the average length of passing time and the average speed at each turn and each time window before each training moment;

[0133] For the specific implementation of steps S401-S405 of this embodiment, please refer to the above Figure 2 The relevant implementation methods of the illustrated embodiment are described and will not be repeated here.

[0134] S406, acquiring historical driving information of a plurality of reference vehicles corresponding to each turn and each training moment from a driving information database;

[0135] For example, in one embodiment of the present disclosure, this step can be implemented in at least one of the following ways:

[0136] The first method is to obtain historical driving information of multiple reference vehicles passing through traffic lights at each training time and each turn from a driving information database;

[0137] This method obtains the historical driving information of reference vehicles passing through traffic lights at the current time of training. The driving status of these reference vehicles can identify the road conditions that affect the road.

[0138] The second way is to obtain, from a driving information database, driving information of a plurality of reference vehicles that are on the affected road at each training moment and pass through traffic lights via each turn after the corresponding training moment.

[0139] This method obtains the driving information of reference vehicles that are on the affected road at the time of training but have not passed the traffic lights. Since the driving status of these reference vehicles is on the affected road at the current time, the driving information of these vehicles can also identify the road conditions of the affected road.

[0140] Since there is less data in the first method mentioned above, in practical applications, it is preferred to use the first method and the second method in combination to more accurately identify the actual road conditions that affect the road.

[0141] S407, determining the actual road conditions affecting the road at each turn and at each training moment based on the historical driving information of multiple reference vehicles corresponding to each turn and each training moment;

[0142] Specifically, when this step is implemented, it may include the following steps:

[0143] (a1) for each reference vehicle corresponding to each training moment on each turn, based on the historical driving information of the corresponding reference vehicle, obtaining the predicted road condition affecting the road on the corresponding turn and at the corresponding training moment;

[0144] In this way, at each turn, at each training moment, and for each reference vehicle, a predicted intersection affecting the road can be obtained.

[0145] For a reference vehicle, multiple predicted road conditions can be obtained accordingly.

[0146] (b1) based on a plurality of predicted road conditions corresponding to a plurality of reference vehicles, obtaining a corresponding real road condition affecting the road at a corresponding turning point and a corresponding training moment,

[0147] For example, based on a voting mechanism, according to multiple predicted road conditions corresponding to multiple reference vehicles, the actual road conditions affecting the road at the corresponding turning point and the corresponding training moment can be obtained.

[0148] Among the predicted road conditions of 100 reference vehicles at a certain training moment and a certain turn, 80 are unobstructed, 15 are slow-moving, and 5 are congested. The road condition with the largest proportion can be taken as the standard, and the actual road condition affecting the road at the current training moment and the current turn can be taken as unobstructed.

[0149] Specifically, when step (a1) is implemented, the following steps may be included: 5 (a2) for each reference vehicle corresponding to each training moment on each turn, based on the corresponding reference vehicle

[0150] The historical driving information of the reference vehicle is used to obtain the vehicle driving characteristics of the corresponding reference vehicle;

[0151] (b2) Determining the predicted road conditions affecting the road at the corresponding turning point and the training time based on the vehicle driving characteristics of each reference vehicle and a preset road condition determination strategy.

[0152] Among them, when step (b2) is specifically implemented, it may include the following situations: 0 Situation 1: For each reference vehicle, if the number of stops of the reference vehicle is less than or equal to the first preset

[0153] The number threshold determines the corresponding turn and the predicted road condition affecting the road at the training time as smooth;

[0154] Scenario 2: If the number of stops of the reference vehicle is greater than or equal to the first preset number threshold and less than the second preset number threshold, but the length of the passing time of the reference vehicle is less than the first preset time threshold

[0155] value, determine the corresponding turning, the predicted road condition affecting the road at the training time is smooth; 5 Case 3, if the number of stops of the reference vehicle is greater than or equal to the first preset number threshold and less than

[0156] At the second preset number threshold, the average speed of the reference vehicle is greater than or equal to the first preset average speed threshold, and the predicted road condition affecting the road at the corresponding turning point and the training time is determined to be smooth;

[0157] Scenario 4: If the number of stops of the reference vehicle is greater than the first preset number threshold, and the length of the passing time of the reference vehicle is greater than or equal to the first preset time threshold and less than the second preset time threshold, it is determined that the predicted road condition affecting the road at the corresponding turning point and the training time is slow moving;

[0158] Scenario 5: If the number of stops of the reference vehicle is greater than the first preset number threshold, the average speed of the reference vehicle is less than the first preset average speed threshold and greater than the second preset average speed threshold, it is determined that the predicted road condition affecting the road at the corresponding turning point and the training time is slow moving;

[0159] Scenario 6: If the number of stops of the reference vehicle is greater than the second preset number threshold, and the length of time for the reference vehicle to pass is greater than the second preset time threshold, it is determined that the predicted road condition affecting the road at the corresponding turning point and the training time is congested; and the second preset number threshold is greater than the first preset number threshold; or

[0160] Scenario 7: If the number of stops of the reference vehicle is greater than the second preset number threshold, and the average speed of the reference vehicle is less than the second preset average speed threshold, it is determined that the predicted road condition affecting the road at the corresponding turning point and the training time is congested.

[0161] In actual applications, it is also possible to accurately determine the predicted road conditions that affect the road at the corresponding turning and training time based on the vehicle driving characteristics of each reference vehicle and in combination with other preset road conditions to determine the strategy, which will not be elaborated in examples here.

[0162] By adopting the above method, the predicted road conditions of the roads affecting each turn at each training moment can be accurately determined; and further, the actual road conditions of the roads affecting each turn and each training moment can be accurately determined.

[0163] S408: Training a road condition prediction model based on the vehicle driving characteristics in each time window at each turn and before each training moment, and the actual road conditions affecting the road at the corresponding turn and the corresponding training moment.

[0164] The traffic condition prediction model of this embodiment may include three traffic condition prediction sub-models. The three traffic condition prediction sub-models may share a feature processing layer, and only use different prediction modules during prediction.

[0165] For example, the road condition prediction model can be constructed using an encoder-decoder structure. The encoder stage uses basic networks such as GRU and transformer, and three decoders are used to estimate the road conditions of three turns. In this embodiment, the estimation of the three directions is made more accurate. The decoder stage uses basic network structures such as GRU and transformer. Since the overall proportion of congestion and slow traffic is relatively low, focal loss is used in training to make the model fit better.

[0166] It should be noted that the training data obtained above include the training data on each turn, and the training data of different turns in this embodiment need to train the corresponding road condition prediction sub-model separately, so before training, the training data needs to be divided into three groups according to the turn. In order to facilitate training, the identification of the turn can also be added when the training data is input during the training process, so as to identify the road condition prediction sub-model of which turn the input training data is used to train. Each piece of training data on each turn can include the vehicle driving characteristics of each time window corresponding to the training time, and the real road conditions affecting the road at the training time. The road condition prediction sub-model of the turn can predict the predicted road conditions affecting the road at the training time based on the input vehicle driving characteristics. Then, based on the predicted road conditions and the real road conditions in the label data, a loss function is constructed, and the parameters of the model are adjusted in the direction of convergence of the super loss function to make the model converge. According to the above method, through continuous training, the road condition prediction sub-models on the three turns can all converge to obtain the final road condition prediction model.

[0167] The training method of the road condition prediction model of the present embodiment can accurately train the road condition prediction model by adopting the above-mentioned method, so that the road condition prediction model can learn the ability to predict the road conditions that affect the road, and can be trained to obtain an accurate road condition prediction model; then, based on the road condition prediction model, the road conditions of the roads affected by traffic lights can be accurately predicted, and further, the road conditions of each road section corresponding to the affected road can be accurately identified.

[0168] Figure 5 is a schematic diagram according to the fifth embodiment of the present disclosure; Figure 5 As shown, this embodiment provides a road condition determination device 500 for a traffic light scene, including:

[0169] The collection module 501 is used to collect the driving information of vehicles passing the traffic light on the specified turn in each time window of a plurality of consecutive time windows before the current moment;

[0170] An acquisition module 502, configured to acquire a vehicle driving feature in each of the time windows on the designated turn based on the driving information of the vehicles passing through the traffic light in each of the time windows on the designated turn;

[0171] A prediction module 503, configured to predict the road condition affecting the traffic light on the designated turn at the current moment based on the vehicle driving characteristics in each of the multiple time windows on the designated turn;

[0172] The identification module 504 is used to identify the road conditions corresponding to each road section mark at the current moment and the designated turn based on the road conditions of the road affected by the traffic light at the current moment and the designated turn and the road section marks corresponding to the affected road.

[0173] The road condition determination device 500 for the traffic light scene of this embodiment implements the implementation principle and technical effect of road condition determination for the traffic light scene by adopting the above-mentioned module, which is the same as the implementation of the above-mentioned related method embodiments. For details, please refer to the records of the above-mentioned related method embodiments, which will not be repeated here.

[0174] Figure 6 is a schematic diagram according to the sixth embodiment of the present disclosure; Figure 6 As shown, this embodiment also provides a road condition determination device 600 for a traffic light scene, including the above Figure 5 The modules with the same name and function are shown as: a collection module 601 , an acquisition module 602 , a prediction module 603 and an identification module 604 .

[0175] In this embodiment, the acquisition module 601 is used to:

[0176] Collect at least one of the number of stops, the length of time of passing, and the average speed of each of the vehicles among all the vehicles passing through the traffic light on the designated turn within each of the time windows before the current moment;

[0177] Among them, the number of times the vehicle stops refers to the number of times the vehicle stops from entering the affected road to passing the traffic light; the passing time length refers to the length of time consumed by the vehicle from entering the affected road to passing the traffic light; the average speed refers to the average speed of the vehicle from entering the affected road to passing the traffic light.

[0178] Optionally, in an embodiment of the present disclosure, the acquisition module 602 is configured to perform at least one of the following operations:

[0179] Based on the number of stops of each of the vehicles among all the vehicles passing through the traffic light in each of the time windows on the designated turn, obtaining an average number of stops in each of the time windows on the designated turn;

[0180] Based on the passing time length of each of the vehicles among all the vehicles passing through the traffic light in each of the time windows on the designated turn, obtaining the average passing time length of each of the time windows on the designated turn; and

[0181] Based on the average speed of each of the vehicles among all the vehicles passing through the traffic light in each of the time windows on the designated turn, an average value of the average speed in each of the time windows on the designated turn is obtained.

[0182] Optionally, in one embodiment of the present disclosure, the prediction module 603 is used to:

[0183] A pre-trained road condition prediction model is used to predict the road condition affecting the road by the traffic light at the designated turn at the current moment based on the vehicle driving characteristics in each of the multiple time windows at the designated turn.

[0184] Optionally, in one embodiment of the present disclosure, the identification module 604 is used to:

[0185] If the traffic condition of the road affected by the traffic light at the current moment and the designated turn is congested, determine that the traffic condition corresponding to the road segment identifier of each road segment included in the affected road at the current moment and the designated turn is congested;

[0186] If the road condition of the road affected by the traffic light at the current moment and the designated turn is smooth, determine that the road condition corresponding to the road segment identifier of each road segment included in the affected road at the current moment and the designated turn is smooth;

[0187] If the traffic condition of the road affected by the traffic light at the specified turn at the current moment is slow traffic, the traffic condition corresponding to each road section mark at the specified turn at the current moment is determined based on the affected length of the affected road, the road section mark corresponding to the affected road and the preset length ratio.

[0188] Optionally, in one embodiment of the present disclosure, the identification module 604 is used to:

[0189] Based on the impact length and the preset length ratio, obtaining a reference length;

[0190] If the road section corresponding to the road section mark is included in the road within the reference length range before the traffic light, it is determined that the road condition corresponding to the road section mark at the current moment and the designated turn is slow traffic;

[0191] If the road section corresponding to the road section mark is not included in the road within the reference length range before the traffic light, it is determined that the road condition corresponding to the road section mark at the current moment and on the designated turn is unobstructed.

[0192] Optionally, in one embodiment of the present disclosure, the identification module 604 is further configured to:

[0193] If only a first portion of the road section corresponding to the road section identifier is included in the road within the reference length range before the traffic light, obtaining a first ratio of the length of the first portion of the road section to the length of the road section corresponding to the road section identifier;

[0194] If the first ratio is greater than or equal to a first preset ratio threshold, it is determined that the road condition corresponding to the road section mark at the current moment and the designated turn is slow traffic;

[0195] If the first ratio is less than the first preset ratio threshold, it is determined that the road condition corresponding to the road section mark on the designated turn at the current moment is smooth.

[0196] Optionally, in one embodiment of the present disclosure, the identification module 604 is further configured to:

[0197] If the traffic condition of the road affected by the traffic light at the designated turn at the current moment is congested, and only a second part of the road section corresponding to the road section identifier is included in the affected road, obtaining a second ratio of the length of the second part of the road section to the length of the road section corresponding to the road section identifier;

[0198] If the second ratio is greater than or equal to a second preset ratio threshold, it is determined that the road condition corresponding to the road section identifier at the current moment and on the designated turn is congested;

[0199] If the second ratio is less than the second preset ratio threshold, the road condition corresponding to the road section identifier is determined based on the road conditions of other sections of the road section corresponding to the road section identifier other than the second partial section.

[0200] Alternatively, if Figure 6As shown, in one embodiment of the present disclosure, the road condition determination device 600 for a traffic light scene further includes:

[0201] The analysis module 605 is used to analyze the road corresponding to the longest distance that vehicles queue through the traffic light based on the driving information database, as the affected road of the traffic light.

[0202] The road condition determination device 600 for the traffic light scene of this embodiment implements the implementation principle and technical effect of road condition determination for the traffic light scene by adopting the above-mentioned module, which is the same as the implementation of the above-mentioned related method embodiments. For details, please refer to the records of the above-mentioned related method embodiments, which will not be repeated here.

[0203] Figure 7 is a schematic diagram according to the seventh embodiment of the present disclosure; Figure 7 As shown, this embodiment provides a training device 700 for a road condition prediction model, including:

[0204] The collection module 701 is used to collect the driving information of the vehicles passing the traffic lights in each turn in each time window of the multiple consecutive time windows before each training moment in the multiple training moments from the historical driving information database;

[0205] The feature acquisition module 702 is used to acquire the vehicle driving features of each of the time windows on each of the turns and before each of the training moments based on the vehicle driving information passing through the traffic lights in each of the time windows on each of the turns and before each of the training moments;

[0206] A road condition acquisition module 703, for acquiring the real road condition of the road affected by the traffic lights at each turning point and each training moment based on the driving information database;

[0207] The training module 704 is used to train the road condition prediction model based on the vehicle driving characteristics in each time window at each turn and before each training moment, and the actual road conditions affecting the road at each turn and at each training moment.

[0208] The training device 700 of the road condition prediction model of this embodiment realizes the implementation principle and technical effect of the training of the road condition prediction model by adopting the above-mentioned modules, which is the same as the implementation of the above-mentioned related method embodiments. For details, please refer to the records of the above-mentioned related method embodiments, which will not be repeated here.

[0209] Further optionally, in one embodiment of the present disclosure, the acquisition module 701 is used to:

[0210] Collecting at least one of the number of stops, the length of time of passing, and the average speed of each of the vehicles among all the vehicles passing through the traffic lights on each of the turns within each of the time windows before each of the training moments from the driving information database;

[0211] Among them, the number of times the vehicle stops refers to the number of times the vehicle stops from entering the affected road to passing the traffic light; the passing time length refers to the length of time consumed by the vehicle from entering the affected road to passing the traffic light; the average speed refers to the average speed of the vehicle from entering the affected road to passing the traffic light.

[0212] Further optionally, in an embodiment of the present disclosure, the feature acquisition module 702 is configured to perform at least one of the following operations:

[0213] Based on the number of stops of each of the vehicles among all the vehicles passing through the traffic light in each of the time windows before each of the training moments on each of the turns, obtaining an average number of stops in each of the time windows before each of the training moments on each of the turns;

[0214] Based on the length of passing time of each of the vehicles among all the vehicles passing through the traffic light in each of the time windows on each of the turns and before each of the training moments, obtaining the average length of passing time in each of the time windows on each of the turns and before each of the training moments; and

[0215] Based on the average speed of each of the vehicles among all the vehicles passing through the traffic light in each of the time windows on each of the turns and before each of the training moments, the average value of the average speed in each of the time windows on each of the turns and before each of the training moments is obtained.

[0216] Further optionally, in one embodiment of the present disclosure, the road condition acquisition module 703 is used to:

[0217] Acquiring historical driving information of a plurality of reference vehicles corresponding to each of the turns and each of the training moments from the driving information database;

[0218] Based on the historical driving information of a plurality of reference vehicles corresponding to each of the turns and each of the training moments, the actual road conditions affecting the road at each of the turns and each of the training moments are determined.

[0219] Further optionally, in one embodiment of the present disclosure, the road condition acquisition module 703 is used to:

[0220] Acquire historical driving information of a plurality of reference vehicles passing through the traffic lights at each training moment and each turn from the driving information database; and / or

[0221] Driving information of a plurality of reference vehicles that are in the affected road at each training moment and pass through the traffic light via each turn after the corresponding training moment is obtained from the driving information database.

[0222] Further optionally, in one embodiment of the present disclosure, the road condition acquisition module 703 is used to:

[0223] For each of the reference vehicles corresponding to each of the training moments on each of the turns, based on the historical driving information of the corresponding reference vehicles, the predicted road conditions affecting the road on the corresponding turns and at the corresponding training moments are obtained;

[0224] Based on the multiple predicted road conditions corresponding to the multiple reference vehicles, the corresponding real road conditions affecting the road at the turn and the corresponding training moment are obtained,

[0225] Further optionally, in one embodiment of the present disclosure, the road condition acquisition module 703 is used to:

[0226] For each of the reference vehicles corresponding to each of the training moments on each of the turns, based on the historical driving information of the corresponding reference vehicle, obtaining the vehicle driving characteristics of the corresponding reference vehicle;

[0227] Based on the vehicle driving characteristics of each of the reference vehicles and a preset road condition determination strategy, the predicted road condition affecting the road at the corresponding turning point and the training moment is determined.

[0228] Further optionally, in one embodiment of the present disclosure, the road condition acquisition module is used to:

[0229] For each of the reference vehicles, if the number of stops of the reference vehicle is less than or equal to a first preset number threshold, determining that the predicted road condition affecting the road at the corresponding turning point and the training time is smooth;

[0230] If the number of stops of the reference vehicle is greater than or equal to the first preset number threshold and less than the second preset number threshold, but the passing time of the reference vehicle is less than the first preset time threshold, it is determined that the predicted road condition affecting the road at the corresponding turning point and the training time is unobstructed;

[0231] If the number of stops of the reference vehicle is greater than or equal to the first preset number threshold and less than the second preset number threshold, and the average speed of the reference vehicle is greater than or equal to the first preset average speed threshold, it is determined that the predicted road condition affecting the road at the corresponding turning point and the training time is unobstructed;

[0232] If the number of stops of the reference vehicle is greater than the first preset number threshold, and the length of the passing time of the reference vehicle is greater than or equal to the first preset time threshold and less than the second preset time threshold, it is determined that the predicted road condition affecting the road at the corresponding turning point and the training time is slow moving;

[0233] If the number of stops of the reference vehicle is greater than the first preset number threshold, and the average speed of the reference vehicle is less than the first preset average speed threshold and greater than the second preset average speed threshold, it is determined that the predicted road condition affecting the road at the corresponding turning point and the training time is slow moving;

[0234] If the number of stops of the reference vehicle is greater than a second preset number threshold, and the length of time for the reference vehicle to pass is greater than the second preset time threshold, it is determined that the predicted road condition affecting the road at the corresponding turning point and the training time is congested; and the second preset number threshold is greater than the first preset number threshold; or

[0235] If the number of stops of the reference vehicle is greater than the second preset number threshold, and the average speed of the reference vehicle is less than the second preset average speed threshold, it is determined that the predicted road condition affecting the road at the corresponding turning point and the training time is congested. …

[0236] The training device 700 for the road condition prediction model of the above embodiment, by adopting the above module to realize the implementation principle of the training of the road condition prediction model, and the technical effect are the same as those of the above-mentioned related method embodiments. For details, please refer to the records of the above-mentioned related method embodiments, which will not be repeated here.

[0237] In the technical solution disclosed herein, the acquisition, storage and application of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0238] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

[0239] Figure 8 A schematic block diagram of an example electronic device 800 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0240] like Figure 8 As shown, the device 800 includes a computing unit 801, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 802 or a computer program loaded from a storage unit 808 into a random access memory (RAM) 803. In the RAM 803, various programs and data required for the operation of the device 800 can also be stored. The computing unit 801, the ROM 802, and the RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0241] A number of components in the device 800 are connected to the I / O interface 805, including: an input unit 806, such as a keyboard, a mouse, etc.; an output unit 807, such as various types of displays, speakers, etc.; a storage unit 808, such as a disk, an optical disk, etc.; and a communication unit 809, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 809 allows the device 800 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0242] The computing unit 801 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 801 performs the various methods and processes described above, such as the above-mentioned methods of the present disclosure. For example, in some embodiments, the above-mentioned methods of the present disclosure may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 808. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 800 via ROM 802 and / or communication unit 809. When the computer program is loaded into RAM 803 and executed by the computing unit 801, one or more steps of the above-mentioned methods of the present disclosure described above may be performed. Alternatively, in other embodiments, the computing unit 801 may be configured to perform the above-mentioned methods of the present disclosure in any other appropriate manner (e.g., by means of firmware).

[0243] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0244] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0245] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0246] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0247] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0248] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0249] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0250] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A road condition recognition method for traffic light scenes, include: Collecting driving information of vehicles passing through traffic lights on a specified turn in each time window of a plurality of consecutive time windows before the current moment; Based on the driving information of vehicles passing through the traffic light in each of the time windows on the designated turn, obtaining the driving characteristics of vehicles in each of the time windows on the designated turn; A pre-trained road condition prediction model is used to predict the road condition of the road affected by the traffic light at the current moment and at the designated turn based on the vehicle driving characteristics in each of the multiple time windows at the designated turn; the road affected by the traffic light refers to a section of road directly connected to the traffic light and affecting vehicles passing through the traffic light, and the road affected by the traffic light includes one, two or more sections; Based on the road condition of the road affected by the traffic light at the current moment and the designated turn and the road segment identifier corresponding to the affected road, identify the road condition corresponding to each road segment identifier at the current moment and the designated turn; Wherein, based on the road condition of the road affected by the traffic light at the current moment and the designated turn and the road section mark corresponding to the affected road, identifying the road condition corresponding to each road section mark at the current moment and the designated turn includes: If the road condition of the road affected by the traffic light at the designated turn at the current moment is slow traffic, a reference length is obtained based on the ratio of the affected length to the preset length; If the road section corresponding to the road section mark is included in the road within the reference length range before the traffic light, it is determined that the road condition corresponding to the road section mark at the current moment and the designated turn is slow traffic; If the road section corresponding to the road section mark is not included in the road within the reference length range before the traffic light, it is determined that the road condition corresponding to the road section mark at the current moment and on the designated turn is unobstructed.

2. The method according to claim 1, in, Collect driving information of vehicles passing through traffic lights on a specified turn in each time window of multiple consecutive time windows before the current time, including: Collect at least one of the number of stops, the length of time of passing, and the average speed of each of the vehicles among all the vehicles passing through the traffic light on the designated turn within each of the time windows before the current moment; Among them, the number of times the vehicle stops refers to the number of times the vehicle stops from entering the affected road to passing the traffic light; the passing time length refers to the length of time consumed by the vehicle from entering the affected road to passing the traffic light; the average speed refers to the average speed of the vehicle from entering the affected road to passing the traffic light.

3. The method according to claim 2, in, Based on the driving information of the vehicles passing through the traffic light in each of the time windows on the designated turn, the driving characteristics of the vehicles in each of the time windows on the designated turn are obtained, including at least one of the following: Based on the number of stops of each of the vehicles among all the vehicles passing through the traffic light in each of the time windows on the designated turn, obtaining an average number of stops in each of the time windows on the designated turn; Based on the passing time length of each of the vehicles among all the vehicles passing through the traffic light in each of the time windows on the designated turn, obtaining the average passing time length of each of the time windows on the designated turn; and Based on the average speed of each of the vehicles among all the vehicles passing through the traffic light in each of the time windows on the designated turn, an average value of the average speed in each of the time windows on the designated turn is obtained.

4. The method according to any one of claims 1 to 3, in, Based on the road condition of the road affected by the traffic light at the current moment and the designated turn and the road segment identifier corresponding to the affected road, identifying the road condition corresponding to each of the road segment identifiers at the current moment and the designated turn, further comprising: If the traffic condition of the road affected by the traffic light at the current moment and the designated turn is congested, determine that the traffic condition corresponding to the road segment identifier of each road segment included in the affected road at the current moment and the designated turn is congested; If the traffic condition of the road affected by the traffic light at the current moment and the designated turn is smooth, determine that the traffic condition corresponding to the section identifier of each section included in the affected road at the current moment and the designated turn is smooth.

5. The method according to claim 1, in, The method further comprises: If only a first portion of the road section corresponding to the road section identifier is included in the road within the reference length range before the traffic light, obtaining a first ratio of the length of the first portion of the road section to the length of the road section corresponding to the road section identifier; If the first ratio is greater than or equal to a first preset ratio threshold, it is determined that the road condition corresponding to the road section mark at the current moment and the designated turn is slow traffic; If the first ratio is less than the first preset ratio threshold, it is determined that the road condition corresponding to the road section mark on the designated turn at the current moment is smooth.

6. The method according to claim 4, in, The method further comprises: If the traffic condition of the road affected by the traffic light at the designated turn at the current moment is congested, and only a second part of the road section corresponding to the road section identifier is included in the affected road, obtaining a second ratio of the length of the second part of the road section to the length of the road section corresponding to the road section identifier; If the second ratio is greater than or equal to a second preset ratio threshold, it is determined that the road condition corresponding to the road section identifier at the current moment and on the designated turn is congested; If the second ratio is less than the second preset ratio threshold, the road condition corresponding to the road section identifier is determined based on the road conditions of other sections of the road section corresponding to the road section identifier other than the second partial section.

7. The method according to any one of claims 1 to 3 and 5 to 6, in, Before predicting the road condition affecting the road by the traffic light at the designated turn at the current moment based on the vehicle driving characteristics in each of the multiple time windows at the designated turn, the method further includes: Based on the driving information database, the road corresponding to the longest distance that vehicles queue through the traffic light is analyzed as the affected road of the traffic light.

8. A training method for a road condition prediction model, include: Collecting driving information of vehicles passing through traffic lights at each turn in each of a plurality of consecutive time windows before each of a plurality of training moments from a historical driving information database; Based on the driving information of vehicles passing through the traffic lights in each of the time windows on each of the turns and before each of the training moments, the driving characteristics of vehicles in each of the time windows on each of the turns and before each of the training moments are obtained; Based on the driving information database, obtaining the real road conditions of the roads affected by the traffic lights at each turning point and each training moment; The road condition prediction model is trained based on the vehicle driving characteristics in each time window at each turn and before each training moment, and the actual road conditions affecting the road at each turn and at each training moment; the road condition prediction model is the road condition prediction model used by any of the methods described in claims 1-7 above.

9. The method according to claim 8, in, From the historical driving information database, driving information of vehicles passing through traffic lights at each turn in each of a plurality of consecutive time windows before each of a plurality of training moments is collected, including: Collecting at least one of the number of stops, the length of time of passing, and the average speed of each of the vehicles among all the vehicles passing through the traffic lights on each of the turns within each of the time windows before each of the training moments from the driving information database; Among them, the number of times the vehicle stops refers to the number of times the vehicle stops from entering the affected road to passing the traffic light; the passing time length refers to the length of time consumed by the vehicle from entering the affected road to passing the traffic light; the average speed refers to the average speed of the vehicle from entering the affected road to passing the traffic light.

10. The method according to claim 9, in, Based on the driving information of vehicles passing through the traffic lights in each of the time windows before each of the training moments on each of the turns, the driving characteristics of vehicles in each of the time windows before each of the training moments on each of the turns are obtained, including at least one of the following: Based on the number of stops of each of the vehicles among all the vehicles passing through the traffic light in each of the time windows before each of the training moments on each of the turns, obtaining an average number of stops in each of the time windows before each of the training moments on each of the turns; Based on the length of passing time of each of the vehicles among all the vehicles passing through the traffic light in each of the time windows before each of the training moments on each of the turns, obtaining the average length of passing time in each of the time windows before each of the training moments on each of the turns; and Based on the average speed of each of the vehicles among all the vehicles passing through the traffic light in each of the time windows on each of the turns and before each of the training moments, the average value of the average speed in each of the time windows on each of the turns and before each of the training moments is obtained.

11. The method according to claim 8, in, Based on the driving information database, the real road conditions of the roads affected by the traffic lights at each turning point and each training moment are obtained, including: Acquiring historical driving information of a plurality of reference vehicles corresponding to each of the turns and each of the training moments from the driving information database; Based on the historical driving information of a plurality of reference vehicles corresponding to each of the turns and each of the training moments, the actual road conditions affecting the road at each of the turns and each of the training moments are determined.

12. The method according to claim 11, in, Acquiring historical driving information of a plurality of reference vehicles corresponding to each of the turns and each of the training moments from the driving information database includes: Acquire historical driving information of a plurality of reference vehicles passing through the traffic lights at each training moment and each turn from the driving information database; and / or Driving information of a plurality of reference vehicles that are in the affected road at each training moment and pass through the traffic light via each turn after the corresponding training moment is obtained from the driving information database.

13. The method according to claim 11 or 12, in, Based on the historical driving information of a plurality of reference vehicles corresponding to each of the turns and each of the training moments, determining the real road conditions affecting the road at each of the turns and each of the training moments, including: For each of the reference vehicles corresponding to each of the training moments on each of the turns, based on the historical driving information of the corresponding reference vehicles, the predicted road conditions affecting the road on the corresponding turns and at the corresponding training moments are obtained; Based on the multiple predicted road conditions corresponding to the multiple reference vehicles, the actual road conditions affecting the road at the corresponding turns and the corresponding training moments are obtained.

14. The method according to claim 13, in, For each of the reference vehicles corresponding to each of the training moments on each of the turns, based on the historical driving information of the corresponding reference vehicles, the predicted road conditions affecting the road on the corresponding turns and at the corresponding training moments are obtained, including: For each of the reference vehicles corresponding to each of the training moments on each of the turns, based on the historical driving information of the corresponding reference vehicle, obtaining the vehicle driving characteristics of the corresponding reference vehicle; Based on the vehicle driving characteristics of each of the reference vehicles and a preset road condition determination strategy, the predicted road condition affecting the road at the corresponding turning point and the training moment is determined.

15. The method according to claim 14, in, Based on the vehicle driving characteristics of each of the reference vehicles and a preset road condition determination strategy, the predicted road condition affecting the road at the corresponding turning point and the training time is determined, including: For each of the reference vehicles, if the number of parking times of the reference vehicle is less than or equal to the first preset number threshold, it is determined that the predicted road condition of the affected road at the corresponding turn and the training moment is unobstructed; If the number of parking times of the reference vehicle is greater than or equal to the first preset number threshold and less than the second preset number threshold, but the passing time length of the reference vehicle is less than the first preset time length threshold, it is determined that the predicted road condition of the affected road at the corresponding turn and the training moment is unobstructed; If the number of parking times of the reference vehicle is greater than or equal to the first preset number threshold and less than the second preset number threshold, and the average speed of the reference vehicle is greater than or equal to the first preset average speed threshold, it is determined that the predicted road condition of the affected road at the corresponding turn and the training moment is unobstructed; If the number of parking times of the reference vehicle is greater than the first preset number threshold, and the passing time length of the reference vehicle is greater than or equal to the first preset time length threshold and less than the second preset time length threshold, it is determined that the predicted road condition of the affected road at the corresponding turn and the training moment is slow; If the number of parking times of the reference vehicle is greater than the first preset number threshold, and the average speed of the reference vehicle is less than the first preset average speed threshold and greater than the second preset average speed threshold, it is determined that the predicted road condition of the affected road at the corresponding turn and the training moment is slow; If the number of parking times of the reference vehicle is greater than the second preset number threshold, and the passing time length of the reference vehicle is greater than the second preset time length threshold, it is determined that the predicted road condition of the affected road at the corresponding turn and the training moment is congested; the second preset number threshold is greater than the first preset number threshold; or If the number of parking times of the reference vehicle is greater than the second preset number threshold, and the average speed of the reference vehicle is less than the second preset average speed threshold, it is determined that the predicted road condition of the affected road at the corresponding turn and the training moment is congested.

16. A road condition recognition device for a traffic light scenario comprising: a collection module, configured to collect the driving information of the vehicles passing through the traffic light in each time window within a plurality of consecutive time windows before the current moment and in a specified direction; an acquisition module, configured to obtain the vehicle driving characteristics of each time window in the specified direction based on the driving information of the vehicles passing through the traffic light in each time window in the specified direction; a prediction module, configured to use a pre-trained road condition prediction model to predict the road condition of the road affected by the traffic light at the current moment and in the specified direction based on the vehicle driving characteristics of each time window in the specified direction and among the plurality of time windows; the road affected by the traffic light refers to a section of road directly connected to the traffic light and affecting the passing of vehicles through the traffic light, and the road affected by the traffic light includes one, two or more sections of road segments; An identification module, configured to identify the road conditions corresponding to the road section identifiers at the current moment and at the designated turn based on the road conditions of the road affected by the traffic light at the current moment and at the designated turn and the road section identifiers corresponding to the affected road; The identification module is used to: If the road condition of the road affected by the traffic light at the designated turn at the current moment is slow traffic, a reference length is obtained based on the ratio of the affected length to the preset length; If the road section corresponding to the road section mark is included in the road within the reference length range before the traffic light, it is determined that the road condition corresponding to the road section mark at the current moment and the designated turn is slow traffic; If the road section corresponding to the road section mark is not included in the road within the reference length range before the traffic light, it is determined that the road condition corresponding to the road section mark at the current moment and on the designated turn is unobstructed.

17. The device according to claim 16, in, The acquisition module is used for: Collect at least one of the number of stops, the length of time of passing, and the average speed of each of the vehicles among all the vehicles passing through the traffic light on the designated turn within each of the time windows before the current moment; Among them, the number of times the vehicle stops refers to the number of times the vehicle stops from entering the affected road to passing the traffic light; the passing time length refers to the length of time consumed by the vehicle from entering the affected road to passing the traffic light; the average speed refers to the average speed of the vehicle from entering the affected road to passing the traffic light.

18. The device according to claim 17, in, The acquisition module is used to perform at least one of the following operations: Based on the number of stops of each of the vehicles among all the vehicles passing through the traffic light in each of the time windows on the designated turn, obtaining an average number of stops in each of the time windows on the designated turn; Based on the passing time length of each of the vehicles among all the vehicles passing through the traffic light in each of the time windows on the designated turn, obtaining the average passing time length of each of the time windows on the designated turn; and Based on the average speed of each of the vehicles among all the vehicles passing through the traffic light in each of the time windows on the designated turn, an average value of the average speed in each of the time windows on the designated turn is obtained.

19. The device according to any one of claims 16 to 18, in, The identification module is further used for: If the traffic condition of the road affected by the traffic light at the current moment and the designated turn is congested, determine that the traffic condition corresponding to the road segment identifier of each road segment included in the affected road at the current moment and the designated turn is congested; If the traffic condition of the road affected by the traffic light at the current moment and the designated turn is smooth, determine that the traffic condition corresponding to the section identifier of each section included in the affected road at the current moment and the designated turn is smooth.

20. The device according to claim 16, in, The identification module is further used for: If only a first portion of the road section corresponding to the road section identifier is included in the road within the reference length range before the traffic light, obtaining a first ratio of the length of the first portion of the road section to the length of the road section corresponding to the road section identifier; If the first ratio is greater than or equal to a first preset ratio threshold, it is determined that the road condition corresponding to the road section mark at the current moment and the designated turn is slow traffic; If the first ratio is less than the first preset ratio threshold, it is determined that the road condition corresponding to the road section mark on the designated turn at the current moment is smooth.

21. The device according to claim 19, in, The identification module is further used for: If the traffic condition of the road affected by the traffic light at the designated turn at the current moment is congested, and only a second part of the road section corresponding to the road section identifier is included in the affected road, obtaining a second ratio of the length of the second part of the road section to the length of the road section corresponding to the road section identifier; If the second ratio is greater than or equal to a second preset ratio threshold, it is determined that the road condition corresponding to the road section identifier at the current moment and on the designated turn is congested; If the second ratio is less than the second preset ratio threshold, the road condition corresponding to the road section identifier is determined based on the road conditions of other sections of the road section corresponding to the road section identifier other than the second partial section.

22. The device according to any one of claims 16 to 18 and 20 to 21, in, The device also includes: The analysis module is used to analyze the road corresponding to the longest distance that vehicles queue through the traffic light based on the driving information database, as the affected road of the traffic light.

23. A training device for a road condition prediction model, include: A collection module, used to collect driving information of vehicles passing through traffic lights on each turn in each time window in a plurality of consecutive time windows before each training moment in a plurality of training moments from a historical driving information database; A feature acquisition module, for acquiring vehicle driving features in each of the time windows on each of the turns and before each of the training moments based on vehicle driving information passing through the traffic lights in each of the time windows on each of the turns and before each of the training moments; A road condition acquisition module, for acquiring the real road condition of the road affected by the traffic lights at each turning point and each training moment based on the driving information database; A training module, used to train a road condition prediction model based on the vehicle driving characteristics in each time window at each turn and before each training moment, and the actual road conditions affecting the road at each turn and at each training moment; the road condition prediction model is the road condition prediction model used by the device described in any one of claims 16-22 above.

24. The device according to claim 23, in, The acquisition module is used for: Collecting at least one of the number of stops, the length of time of passing, and the average speed of each of the vehicles among all the vehicles passing through the traffic lights on each of the turns within each of the time windows before each of the training moments from the driving information database; Among them, the number of times the vehicle stops refers to the number of times the vehicle stops from entering the affected road to passing the traffic light; the passing time length refers to the length of time consumed by the vehicle from entering the affected road to passing the traffic light; the average speed refers to the average speed of the vehicle from entering the affected road to passing the traffic light.

25. The device according to claim 24, in, The feature acquisition module is used to perform at least one of the following operations: Based on the number of stops of each of the vehicles among all the vehicles passing through the traffic light in each of the time windows before each of the training moments on each of the turns, obtaining an average number of stops in each of the time windows before each of the training moments on each of the turns; Based on the length of passing time of each of the vehicles among all the vehicles passing through the traffic light in each of the time windows before each of the training moments on each of the turns, obtaining the average length of passing time in each of the time windows before each of the training moments on each of the turns; and Based on the average speed of each of the vehicles among all the vehicles passing through the traffic light in each of the time windows on each of the turns and before each of the training moments, the average value of the average speed in each of the time windows on each of the turns and before each of the training moments is obtained.

26. The device according to claim 23, in, The road condition acquisition module is used to: Acquiring historical driving information of a plurality of reference vehicles corresponding to each of the turns and each of the training moments from the driving information database; Based on the historical driving information of a plurality of reference vehicles corresponding to each of the turns and each of the training moments, the actual road conditions affecting the road at each of the turns and each of the training moments are determined.

27. The device according to claim 26, in, The road condition acquisition module is used to: Acquire historical driving information of a plurality of reference vehicles passing through the traffic lights at each training moment and each turn from the driving information database; and / or Driving information of a plurality of reference vehicles that are in the affected road at each training moment and pass through the traffic light via each turn after the corresponding training moment is obtained from the driving information database.

28. The device according to claim 26 or 27, in, The road condition acquisition module is used to: For each of the reference vehicles corresponding to each of the training moments on each of the turns, based on the historical driving information of the corresponding reference vehicles, the predicted road conditions affecting the road on the corresponding turns and at the corresponding training moments are obtained; Based on the multiple predicted road conditions corresponding to the multiple reference vehicles, the actual road conditions affecting the road at the corresponding turns and the corresponding training moments are obtained.

29. The device according to claim 28, in, The road condition acquisition module is used to: For each of the reference vehicles corresponding to each of the training moments on each of the turns, based on the historical driving information of the corresponding reference vehicle, obtaining the vehicle driving characteristics of the corresponding reference vehicle; Based on the vehicle driving characteristics of each of the reference vehicles and a preset road condition determination strategy, the predicted road condition affecting the road at the corresponding turning point and the training moment is determined.

30. The device according to claim 29, in, The road condition acquisition module is used to: For each of the reference vehicles, if the number of stops of the reference vehicle is less than or equal to a first preset number threshold, determining that the predicted road condition affecting the road at the corresponding turning point and the training time is smooth; If the number of stops of the reference vehicle is greater than or equal to the first preset number threshold and less than the second preset number threshold, but the passing time of the reference vehicle is less than the first preset time threshold, it is determined that the predicted road condition affecting the road at the corresponding turning point and the training time is unobstructed; If the number of stops of the reference vehicle is greater than or equal to the first preset number threshold and less than the second preset number threshold, and the average speed of the reference vehicle is greater than or equal to the first preset average speed threshold, it is determined that the predicted road condition affecting the road at the corresponding turning point and the training time is unobstructed; If the number of stops of the reference vehicle is greater than the first preset number threshold, and the length of the passing time of the reference vehicle is greater than or equal to the first preset time threshold and less than the second preset time threshold, it is determined that the predicted road condition affecting the road at the corresponding turning point and the training time is slow moving; If the number of stops of the reference vehicle is greater than the first preset number threshold, and the average speed of the reference vehicle is less than the first preset average speed threshold and greater than the second preset average speed threshold, it is determined that the predicted road condition affecting the road at the corresponding turning point and the training time is slow moving; If the number of stops of the reference vehicle is greater than a second preset number threshold, and the length of time for the reference vehicle to pass is greater than the second preset time threshold, it is determined that the predicted road condition affecting the road at the corresponding turning point and the training time is congested; and the second preset number threshold is greater than the first preset number threshold; or If the number of stops of the reference vehicle is greater than the second preset number threshold, and the average speed of the reference vehicle is less than the second preset average speed threshold, it is determined that the predicted road condition affecting the road at the corresponding turn and at the training time is congested.

31. An electronic device, include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-7 or 8-15.

32. A non-transitory computer-readable storage medium storing computer instructions, in, The computer instructions are used to cause the computer to execute the method according to any one of claims 1-7 or 8-15.

33. A computer program product comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1-7 or 8-15.

Images