Method and device for determining intersection passing time, electronic equipment and storage medium

Abstract

The present disclosure provides a method and device for determining intersection passing time, electronic equipment and storage medium, relates to the field of artificial intelligence, especially to the field of intelligent transportation. The specific implementation scheme is: obtaining object information of a target object, the object information including current position information of the target object and path planning information of the target object; determining a reference passing time of a target intersection associated with the object information, wherein the reference passing time is related to a historical passing time of the target intersection; determining signal light information of a target signal light in the target intersection, the signal light information including a correspondence between the light state and time of the target signal light after the current time; and determining a target passing time of the target intersection according to the object information, the reference passing time and the signal light information, so as to display the target passing time.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of artificial intelligence, in particular to the field of intelligent transportation, and more specifically, the present disclosure provides a method and device for determining a road intersection passing time, an electronic device, a storage medium, and a computer program product. BACKGROUND

[0002] When a user is driving a vehicle through a road intersection, the user may not be able to see the countdown information of a signal light due to a long distance, especially in a congested or slow-moving scenario, and the user may have to wait for multiple red lights before being able to pass through the front road intersection. The user is unable to know the time required to pass through the road intersection, which causes psychological uncertainty and anxiety, thereby affecting the driving experience. SUMMARY

[0003] The present disclosure provides a method and device for determining a road intersection passing time, an electronic device, a storage medium, and a computer program product.

[0004] According to an aspect of the present disclosure, a method for determining a road intersection passing time is provided, including: obtaining object information of a target object, the object information including current position information of the target object and path planning information of the target object; determining a reference passing time of a target road intersection associated with the object information, wherein the reference passing time is related to a historical passing time of the target road intersection; determining signal light information of a target signal light in the target road intersection, the signal light information including a correspondence between light states and times of the target signal light after a current time; and determining a target passing time of the target road intersection according to the object information, the reference passing time, and the signal light information, so as to display the target passing time.

[0005] According to another aspect of the present disclosure, a device for determining a road intersection passing time is provided, including: an obtaining module, a first determining module, a second determining module, and a third determining module. The obtaining module is configured to obtain object information of a target object, the object information including current position information of the target object and path planning information of the target object. The first determining module is configured to determine a reference passing time of a target road intersection associated with the object information, wherein the reference passing time is related to a historical passing time of the target road intersection. The second determining module is configured to determine signal light information of a target signal light in the target road intersection, the signal light information including a correspondence between light states and times of the target signal light after a current time. The third determining module is configured to determine a target passing time of the target road intersection according to the object information, the reference passing time, and the signal light information, so as to display the target passing time.

[0006] According to another aspect of the present disclosure, an electronic device is provided, comprising: at least one processor; and a memory connected with the at least one processor in communication; wherein the memory stores instructions executable 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 provided by the present disclosure.

[0007] According to another aspect of the present disclosure, a non-transitory computer readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable a computer to perform the method provided by the present disclosure.

[0008] According to another aspect of the present disclosure, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the method provided by the present disclosure.

[0009] It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of the present disclosure, nor is it used to limit the scope of the present disclosure. Other features of the present disclosure will become apparent through the following description. BRIEF DESCRIPTION OF DRAWINGS

[0010] The accompanying drawings are used to better understand the present scheme, and do not limit the present disclosure. Among them:

[0011] Figure 1 is an application scenario diagram of the method and device for determining intersection passing duration according to the embodiments of the present disclosure;

[0012] Figure 2 is a schematic flow chart of the method for determining intersection passing duration according to the embodiments of the present disclosure;

[0013] Figure 3 is a schematic flow chart of the method for determining signal light information according to the embodiments of the present disclosure;

[0014] Figure 4 is a schematic principle diagram of the method for determining intersection passing duration according to the embodiments of the present disclosure;

[0015] Figure 5 is a schematic structural block diagram of the device for determining intersection passing duration according to the embodiments of the present disclosure; and

[0016] Figure 6 is a structural block diagram of an electronic device for implementing the method for determining intersection passing duration according to the embodiments of the present disclosure. DETAILED DESCRIPTION

[0017] Exemplary embodiments of the present disclosure are described herein with reference to the accompanying drawings, which are meant to be exemplary. Therefore, various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Also, descriptions of known functions and constructions are omitted herein for clarity and conciseness.

[0018] In some embodiments, the intersection can be divided into multiple road segments, and for each road segment, the average passing speed of the road segment can be determined according to the average passing speed of the road segment in the past several time slices (for example, 10 minutes as a time slice) and the average passing speed of the intersection at the time in the historical data. The average passing speeds of the multiple road segments included in the same intersection can be used to determine the overall average passing speed and the average passing time of the intersection.

[0019] It should be noted that for intersections with traffic lights, there is a large difference in the passing time of the same intersection by different users at similar times. For example, some users do not need to wait for a red light or only need to wait for half a red light, and can pass through the intersection in a short time. However, some users need to wait for two red lights, and thus need a long time to pass through the intersection.

[0020] Therefore, the above technical solution is based on the average speed for statistics, and cannot estimate the light state and countdown of the traffic light when the user arrives at the intersection, and thus cannot accurately depict the passing time of each user, resulting in a low accuracy of the passing time.

[0021] Embodiments of the present disclosure aim to provide a method for determining the passing time of an intersection, which can use user starting, stopping, accelerating, decelerating, and other behavior data, image information of traffic lights, and the like, to mine the cycle information and switching time of the traffic light, obtain the traffic light information, and then accurately estimate the target passing time of the user based on the traffic light information.

[0022] The technical solutions provided by the present disclosure will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0023] Figure 1 is a schematic diagram of an application scenario of the method and device for determining the passing time of an intersection according to embodiments of the present disclosure.

[0024] It should be noted that, Figure 1 The system architecture shown is only an example of a system architecture to which embodiments of the present disclosure can be applied, to help those skilled in the art understand the technical content of the present disclosure, but does not mean that embodiments of the present disclosure cannot be applied to other devices, systems, environments, or scenarios.

[0025] As Figure 1As shown, the system architecture 100 according to the embodiment can include a vehicle 110, a network 120, a server 130, and the like. The network 120 is a medium for providing a communication link between the vehicle 110 and the server 130. The network 120 can include various connection types, such as wired and / or wireless communication links, and the like.

[0026] The vehicle 110 can be integrated with sensors such as a vehicle-mounted camera, which can interact with the server 130 through the network 120 to receive or send messages, and the like.

[0027] The server 130 can be disposed at a remote end capable of establishing communication with the vehicle terminal, and can be implemented as a distributed server cluster composed of multiple servers, or as a single server.

[0028] The server 130 can be a server providing various services. The server 130 can be installed with, for example, a map application, a data processing application, and the like. Taking the case where the server 130 runs the data processing application as an example: receiving, through the network 120, data such as current position information and path planning information transmitted from the vehicle 110, and taking the data as to-be-processed data, and processing the to-be-processed data to obtain a target passage time required for the vehicle to pass through the intersection from the current position.

[0029] It should be noted that the method for determining the intersection passage time provided by the embodiments of the present disclosure can generally be executed by the server 130. Accordingly, the device for determining the intersection passage time provided by the embodiments of the present disclosure can also be disposed in the server 130.

[0030] It can be understood that the number of sensors, networks, and servers in the system architecture 100 is merely illustrative. According to the implementation needs, there can be any number of sensors, networks, and servers. Figure 1

[0031] Figure 2 FIG. 2 is a schematic flowchart of a method for determining an intersection passage time according to an embodiment of the present disclosure.

[0032] As shown in FIG. 2, the method 200 for determining the intersection passage time can include operations S210-S240. Figure 2

[0033] At operation S210, object information of a target object is obtained, and the object information includes current position information of the target object and path planning information of the target object.

[0034] For example, the target object can be a pedestrian, a vehicle, or the like in a target road and planning to pass through a target intersection.

[0035] ​​For example, the path planning information in the object information can represent route information selected from a departure location to a destination location, which can include turning information such as left turn, straight, right turn, etc. when the user passes through an intersection. For example, for a target object using a navigation application, the current location information and the path planning information of the target object can be determined through the navigation application.

[0036] For example, the object information can also include the current speed of the target object, trajectory information in the past period of time, etc.

[0037] For example, the object information of the target object can be acquired in real time, or the object information can be acquired when the distance between the target object and the intersection is within a predetermined range, which can be between 200 meters and 1000 meters from the intersection.

[0038] In operation S220, the reference passing duration of the target intersection associated with the object information is determined, wherein the reference passing duration is related to the historical passing duration of the target intersection.

[0039] For example, the target intersection can represent the first intersection that the target object needs to pass through at the current location in the process of moving along the pre-planned route.

[0040] For example, the passing duration of the objects such as vehicles, pedestrians, etc. passing through each intersection in a plurality of intersections in a predetermined period of time according to various paths (such as straight, left turn, etc.) can be counted in advance, so as to obtain the average passing duration of each path passing through the intersection, for example, the average passing duration of straight passing through the intersection is 3 minutes, and the average passing duration of left turn passing through the intersection is 5 minutes. The average passing duration, the path, and the corresponding relationship therebetween can also be stored. The stored data can also be updated periodically to ensure the timeliness of the data, for example, the average passing duration of each intersection according to various paths is updated once every 5 minutes.

[0041] For example, the target intersection associated with the object information can be determined from a plurality of intersections first, and then the average passing duration corresponding to the target intersection and the target path is found from the stored data as the reference passing duration of the target intersection.

[0042] In operation S230, the signal light information of the target signal light in the target intersection is determined, and the signal light information includes the correspondence between the light state and the time of the target signal light after the current time.

[0043] For example, the same target intersection can be provided with a plurality of signal lights, and the target signal light can represent the signal light that the target object needs to comply with when passing through the target intersection in the process of moving along the pre-planned route.

[0044] For example, the light states of the signal light can include a red light, a yellow light and a green light.

[0045] For example, the signal light information can embody the display time period of each light state of the target signal light in a future period of time, and the time of switching of the signal light from one light state to another light state. For example, the current time is 4:00:00 pm, the signal light information can include: the red light is displayed from 4:51:00 to 4:52:00, and the switching of the green light to the red light occurs at 4:58:00.

[0046] For example, for each intersection in the road, reference traffic light information such as timing rules and signal light switching time of each intersection can be obtained from a third party. For example, the data provided by the third party indicates that in the next two months, from 4 pm to 5 pm every day, the period of the straight signal light of a certain intersection is 2 minutes, that is, it takes 2 minutes to complete a round of signal light display, and the red light is displayed for 58 seconds, the green light is displayed for 58 seconds, and the yellow light is displayed for 4 seconds.

[0047] For another example, historical data related to the period rules and switching time of the intersection signal light in the past period of time can be counted, so as to estimate the future signal light information by using the historical data. For example, the straight signal light of a certain intersection, in the past month, from 4 pm to 5 pm every day, the signal light period is 2 minutes, and the duration of the red light and the green light is 58 seconds respectively, and the duration of the yellow light is 4 seconds, and in the past week, the switching of the red light to the yellow light occurs at 4:00:50 every day. Therefore, the above signal light information can be used as the future signal light information of the intersection.

[0048] For example, the signal light information of each intersection can be determined and stored in advance by the above-mentioned third party obtaining and / or statistical historical data. The stored data can also be updated regularly to ensure the timeliness of the data, for example, the signal light information of each intersection stored is updated once every 2 minutes.

[0049] For example, the target intersection associated with the object information can be determined from a plurality of intersections, and then the signal light information of the target intersection can be determined based on the identification of the target intersection.

[0050] In operation S240, the target passing time of the target intersection is determined according to the object information, the reference passing time and the signal light information, so as to display the target passing time.

[0051] For example, the target passing time can be determined by a deep learning model, and the training method, input and output of the deep learning model are described in detail below.

[0052] For example, the time for the target object to reach the intersection can be determined in combination with the current position information and the speed of the target object, and the length of time that the target user needs to wait at the intersection can be determined based on the time and the signal light information. Next, the target passing time can be calculated based on the waiting time and the reference passing time, for example, the target passing time can be calculated based on a weighted sum.

[0053] According to the technical scheme provided by the embodiments of the present disclosure, since the correspondence between the light state and the time of the signal light in the future is estimated first to obtain the signal light information, and then the reference passing time is corrected based on the signal light information, the target passing time with higher accuracy than the reference passing time can be obtained.

[0054] Figure 3 is a schematic flowchart of a method for determining signal light information according to an embodiment of the present disclosure.

[0055] As shown in Figure 3 The method 330 for determining the passing time of the intersection can include operations S331-S334.

[0056] Before introducing operations S331-S334, the trajectory data is described first.

[0057] For example, the trajectory data is associated with a target intersection, which can indicate that the motion path represented by the trajectory data passes through the target intersection.

[0058] For example, the same trajectory data can include multiple trajectory points, and each trajectory point corresponds to the direction information, speed information, position information, etc. of the object at the trajectory point. The behavior data of the trajectory point can include acceleration, acceleration, start, stop, etc.

[0059] For example, the trajectory data can include first historical trajectory data and second historical trajectory data.

[0060] For example, the first historical trajectory data can include trajectory data within a first predetermined period in the past, and the time difference between the first predetermined period and the current time can be small, for example, the first predetermined period is the past half hour.

[0061] For example, the second historical trajectory data can include trajectory data within a second predetermined period in the past, and the time difference between the second predetermined period and the current time can be large, for example, the second predetermined period is the past one month. In addition, the second predetermined period can also be before the first predetermined period, for example, the range of the second predetermined period is the past one month to the past half hour.

[0062] In operation S331, first switching time information in the switching time information is determined according to first historical trajectory data associated with the target intersection; the first historical trajectory data includes behavior data and time information corresponding to each of the plurality of trajectory points.

[0063] For example, for an intersection with a signal light, a parking trajectory can be extracted from the trajectory information related to the intersection, and the acceleration, deceleration, stop, start, etc. of the parking trajectory can be used to determine the time when the signal light changes the light state in the past period of time.

[0064] For example, the start behavior and acceleration behavior in the trajectory data can represent that the signal light changes from red to green, and the deceleration behavior and stop behavior in the trajectory data can represent that the signal light changes from green to red.

[0065] According to the historical data, the correlation between the behavior data of the trajectory point in the trajectory data and the light state switching time can be determined, for example, according to the historical data, it is known that the signal light changes to green at time a, and the object starts at time (a-2 seconds), so it can be determined that the time difference between the start behavior and the time when the signal light changes to green is 2 seconds. Therefore, based on the behavior data of the trajectory point and the time information corresponding to the behavior, the light state switching time of the signal light can be estimated.

[0066] The operation determines the switching time of the signal light by the behavior of the target object at the trajectory point and the time when the behavior occurs, and can accurately estimate the switching time of the signal light.

[0067] In operation S332, second switching time information in the switching time information is determined according to image data of the target signal light.

[0068] For example, a scheduling task can be sent to a vehicle, and after receiving the scheduling task, the vehicle-mounted camera of the vehicle can collect images of the signal light at the intersection, and then the collected images can be recognized and target detected to determine the light state of the signal light in the image. In addition, the image can have a timestamp corresponding to the collection process, and based on the timestamp, the correspondence between the light state of the signal light and the time can be determined.

[0069] For example, only one image of the signal light can be collected, and the signal light can display the remaining time of the current light state, so based on the collection time of the image, the light state obtained by processing the image and the remaining time, the light state switching time of the signal light can be determined. For example, the image is collected at 5:00:00, the light state in the image is red, and the countdown is 10 seconds, so it can be determined that the switching from red to yellow occurs at 5:00:10.

[0070] For example, a plurality of images can be continuously captured for a signal light, and it can be determined that a red light to yellow light switching occurs between the capture time of the seventh image and the capture time of the eighth image, or at the capture time of the eighth image, based on the fact that the light state of the signal light in the seventh image is red and the light state of the signal light in the eighth image is yellow after the images are processed.

[0071] The operation can accurately obtain the switching time of the signal light based on the image data of the signal light captured before the current time, and has high real-time performance.

[0072] In operation S333, period information of the target signal light is determined according to second historical trajectory data associated with the target intersection.

[0073] For example, a plurality of second historical trajectory data can be counted, for example, whether the period of the signal light has a stable state in the same period of the same day is counted. For example, the duration of the red light of a certain traffic light is 60 seconds, the duration of the green light is 55 seconds, and the duration of the yellow light is 5 seconds between 8:00 and 9:00 in most days in the past month, so it can be estimated that the period information of the signal light is 2 minutes between 8:00 and 9:00 in the future, and the first 60 seconds are red light, the middle 5 seconds are yellow light, and the last 55 seconds are green light. Through the above method, the duration of each light state in the signal light in each period of the same day in the future can be obtained.

[0074] The operation can obtain relatively accurate signal light period information by mining the period rule of the signal light through historical trajectories and estimating the future period information using historical trajectories.

[0075] In operation S334, signal light information is estimated according to the period information of the target signal light and the switching time information of the target signal light; wherein the period information includes the duration of each light state in the signal light.

[0076] For example, the period information can include the duration of the red light, the duration of the yellow light, and the duration of the green light, and the sum of the durations of the three color signal lights is a signal light period.

[0077] For example, the switching time information can represent the time information of the switching between adjacent two light states.

[0078] Based on the period information of the target signal light and the switching time information of the target signal light, the signal light information can be estimated. For example, the signal light switches from green light to red light at 4:10:00, and during the period of 4:00 to 5:00, the period of a signal light is 2 minutes, and the durations of the red light, the yellow light, and the green light are 57 seconds, 3 seconds, and 60 seconds, respectively. Therefore, it can be determined that the green light to red light switching occurs at 4:12, 4:14, 4:16, and so on.

[0079] The periodic information and the switching time of the signal light are used to determine the switching time of the signal light in the future, the light state at each time, and the like. On the premise that the periodic information and the switching time are accurate, the signal light information at each time in the future can be accurately obtained, so as to ensure accurate estimation of the target passing time.

[0080] According to another embodiment of the present disclosure, a deep learning model can be trained in advance, and then the deep learning model is used to determine the target passing time.

[0081] It can be understood that before training the deep learning model, a training sample can be constructed first. The process of constructing the sample can include determining the features and determining the label of the sample.

[0082] In the process of determining the features, for example, a plurality of categories of features can be extracted based on at least one historical trajectory passing through the intersection, such as object information features, reference passing time features, signal light information features, and the like.

[0083] In addition, other target features can also be extracted based on other information. For example, attribute features corresponding to the road where the object is located can be extracted from the road information, such as road length, width, speed limit level, ramp, highway, two-way road, one-way road, and the like. For example, intersection features such as smooth, slow, and congested can also be extracted based on the road condition information of the road where the object is located. For example, reference signal light features can also be extracted from the reference signal light related data provided by a third party, and the reference signal light features can represent the period of the signal light and the switching time between each light state. Through the above other target features, the target passing time can be more accurately described, the training accuracy of the deep learning model can be improved, and the accuracy of the target passing time can be improved.

[0084] The label of the training sample can be the deviation information between the reference passing time and the passing time output by the deep learning model. For example, the reference passing time is 60 seconds, the passing time output by the deep learning model is 78 seconds, and the deviation information can be the ratio between the passing time output by the deep learning model and the reference passing time, that is, the label is 1.3. Or the deviation information is the difference between the reference passing time and the passing time output by the deep learning model, for example, the label is 18.

[0085] According to the features and labels extracted above, the training sample can be obtained.

[0086] Next, the deep learning model can be trained using the training samples. In the model training stage, the focal-loss method can be used to reduce the weight of the samples with excessive deviation in the loss function, thereby weakening the influence of the training samples with less influence on signal light switching on the model parameters. In addition, methods such as Dropout and Early Stopping can be used to prevent overfitting, thereby improving the generalization ability of the model in the case of sparse trajectories and trajectory noise.

[0087] In the model application stage, the object information, the reference passing time and the signal light information can be input into the deep learning model, and the target features (such as the road network features of the target road, the road condition features, the reference signal light features of the target signal light, etc.) can also be input into the deep learning model, and the deep learning model outputs the target passing time.

[0088] The technical solution provided by the embodiments of the present disclosure uses a deep learning model to process input data and obtain a target passing time. Compared with a technical solution that determines a target passing time through simple mathematical calculation, the accuracy of the target passing time can be further improved.

[0089] Figure 4 is a schematic diagram of a method for determining a passing time at an intersection according to an embodiment of the present disclosure.

[0090] The method for determining a passing time at an intersection provided by the embodiments of the present disclosure will be described in detail below. Figure 4 The method for determining a passing time at an intersection provided by the embodiments of the present disclosure will be described in detail below. In this embodiment, the method for determining a passing time at an intersection includes the following stages: model training, basic data determination and target passing time calculation.

[0091] In the model training stage, the model to be used can be trained based on the posterior trajectory data.

[0092] For example, taking the deep learning model 415 mentioned above as an example, the real passing time of a plurality of objects passing through each intersection can be determined based on the posterior trajectory data, and the real passing time is the sample true value. Various types of features can also be extracted from the posterior trajectory data, such as object information features, reference passing time features, signal light information features, etc. Other target features can also be extracted based on other information, such as the attribute features of the road where the object is located, the road condition information, the reference signal light features provided by a third party, etc. The label of the sample is the deviation information between the reference passing time output by the intersection ETA general model 409 and the passing time output by the deep learning model 415. Based on the above features and labels, a training sample can be constructed. The deep learning model 415 can be trained using the training sample.

[0093] In the basic data determination stage, the required basic data can be determined based on the second historical trajectory data 402 generated based on the historical navigation behavior of the user, and the data generated based on the real-time navigation behavior of the user (for example, the first historical trajectory data 401 and the real-time image 403).

[0094] For example, based on the historical navigation behavior of a plurality of users, the trajectory data in a period far from the current time (for example, in the past month before the current time) can be referred to as the second historical trajectory data 402, and the second historical trajectory data 402 is used to reflect the historical rule.

[0095] For example, based on the real-time navigation behavior of a plurality of users, the trajectory data in a period close to the current time (for example, in the past half hour before the current time) can be referred to as the first historical trajectory data 401, and the first historical trajectory data 401 is used to reflect the real-time information.

[0096] For example, the required data can include a reference calculation duration. The reference passing duration can be obtained by the following scheme: the real-time passing rule 405 can be mined from the first historical trajectory data 401, and the historical passing rule 404 can be mined from the second historical trajectory data 402. Next, the real-time passing rule 405 and the historical passing rule 404 can be processed by using the intersection ETA general model 409, so as to obtain the average passing duration of a plurality of users passing through each intersection, which can be used as the reference passing duration. The structure and working principle of the intersection ETA general model 409 are not limited in the embodiments of the present disclosure.

[0097] For example, the required data can also include signal light information. The signal light information can be obtained by the following scheme:

[0098] The second switching time 408 of the light state of the signal light can be determined based on the real-time image 403. For example, reference can be made to the operation S332 described above, and details are not described herein again.

[0099] The first switching time 407 information of the signal light can be determined based on the first historical trajectory, extracting the data related to the behaviors such as acceleration, deceleration, start, and stop, and then based on the behavior data and the time information of the behaviors.

[0100] Based on the second historical trajectory data 402, it can be determined whether the signal light has a stable state at each time period, for example, the duration of the red light is 50 seconds and the duration of the green light is 40 seconds between 8:00 and 9:00 in the morning, so as to obtain the cycle information 406 of the signal light at each time period in a day. The cycle information 406 can include the respective durations of the red light, yellow light and green light in each time period in a day. For example, reference can be made to the operation S333 described above, and details are not repeated here.

[0101] Next, according to the first switching time 407 and the second switching time 408 of each signal light and the cycle information 406 of the signal light, the correspondence between the light state and the time of the signal light in the future time period can be determined. In actual application, for the same signal light, when the first switching time 407 and the second switching time 408 corresponding to the signal light exist at the same time, the one with lower confidence can be discarded.

[0102] It should be noted that the above reference calculation duration and signal light information are mainly obtained based on the first historical trajectory data 401 and the second historical trajectory data 402, and the first historical trajectory data 401 and the second historical trajectory data 402 can be updated regularly to ensure the timeliness of the data. For example, the first historical trajectory data 401 can be updated every few minutes, and the second historical trajectory data 402 can be updated every day.

[0103] In the target passing time calculation stage, the pre-trained deep learning model 415 and the above pre-determined or real-time determined reference calculation duration, signal light information, etc. can be used to calculate the target passing time 416.

[0104] For example, when the distance between the target object and the intersection is within a predetermined range, which can be 200-1000 meters from the intersection, the terminal can send a message to the server, and the message sent includes the target object information 413, such as the current position information, path planning information, current speed, trajectory information in the past time, etc. of the target object.

[0105] Next, the server can determine the target intersection based on the object information 413 of the target object, and then query the reference passing time corresponding to the target path. Similarly, the target signal light can also be determined from the plurality of signal lights, and the signal light information corresponding to the target signal light can be queried.

[0106] In addition, the server can also determine other target features based on the object information 413 of the target object. For example, the road network features 411 of the target road where the target object is located, the real-time traffic features 412, etc. can be determined, and the third-party provided reference signal light information 414 can also be determined.

[0107] After obtaining the above information, the server can input the road network features 411, real-time road condition features 412, reference signal light information 414, reference passing time, signal light information, and the like into the trained deep learning model 415, and the deep learning model 415 outputs the target passing time 416. The target passing time 416 represents the time required for the target object to pass through the target intersection at the current location. In addition, the number of rounds of red light that the target object needs to wait can also be calculated based on the target passing time 416 and the cycle information 406 of the signal light.

[0108] After obtaining the target passing time 416, the server can send the target passing time 416, signal light information, and the like to the terminal, and the terminal can display the target passing time 416, signal light information, and the number of rounds of waiting for the signal light, so as to prompt the target object of the actual distance from the target intersection, the time required through the target path in front, the remaining time of the signal light switching, and the like. information, and the number of rounds of waiting for the signal light at the intersection.

[0109] The method provided by the embodiments of the present disclosure is suitable for various scenes such as congestion, slow driving, and smooth driving. By prompting the target passing time 416 and the like during real-time navigation of the user, the accuracy of the target passing time 416 can reach seconds, thereby improving the driving experience of the user.

[0110] Figure 5 is a schematic structural block diagram of a device for determining an intersection passing time according to an embodiment of the present disclosure.

[0111] As shown in Figure 5 , the device 500 for determining an intersection passing time can include an acquisition module 510, a first determination module 520, a second determination module 530, and a third determination module 540.

[0112] The acquisition module 510 is configured to acquire object information of a target object, the object information including current location information of the target object and path planning information of the target object.

[0113] The first determination module 520 is configured to determine a reference passing time of a target intersection associated with the object information, wherein the reference passing time is related to a historical passing time of the target intersection.

[0114] The second determination module 530 is configured to determine signal light information of a target signal light in the target intersection, the signal light information including a correspondence between light states and times of the target signal light after a current time.

[0115] The third determination module 540 is configured to determine a target passing time of the target intersection according to the object information, the reference passing time, and the signal light information, so as to display the target passing time.

[0116] According to another embodiment of the present disclosure, the second determining module comprises: an estimating submodule, configured to estimate the signal light information according to the period information of the target signal light and the switching time information of the target signal light; wherein the period information comprises the duration of each light state of the signal light.

[0117] According to another embodiment of the present disclosure, the second determining module further comprises at least one of a first determining submodule and a second determining submodule. The first determining submodule is configured to determine first switching time information in the switching time information according to first historical trajectory data associated with the target intersection. The first historical trajectory data comprises behavior data and time information corresponding to each trajectory point. The second determining submodule is configured to determine second switching time information in the switching time information according to image data of the target signal light.

[0118] According to another embodiment of the present disclosure, the second determining module further comprises: a third determining submodule, configured to determine the period information of the target signal light according to second historical trajectory data associated with the target intersection; wherein the generation time of the second historical trajectory data is prior to the generation time of the first historical trajectory data.

[0119] According to another embodiment of the present disclosure, the third determining module comprises: a fourth determining submodule, configured to determine the target passing time according to the target feature, the object information, the signal light information and the reference passing time; wherein the target feature comprises at least one of the following: attribute feature of the target road corresponding to the current position information, road condition feature of the target road, reference signal light information for the target signal light.

[0120] According to another embodiment of the present disclosure, the third determining module comprises: an input submodule, configured to input the object information, the reference passing time and the signal light information into a deep learning model to obtain the target passing time of the target intersection; wherein the deep learning model is obtained by training a plurality of training samples, each training sample comprises object information feature, reference passing time feature and signal light information feature, and the label of the training sample is the deviation information between the reference passing time and the passing time output by the deep learning model.

[0121] In the technical solution of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved in the technical solution of the present disclosure all comply with the relevant laws and regulations and do not violate public order and good customs.

[0122] In the technical solution of the present disclosure, the authorization or consent of the user is obtained before the user's personal information is acquired or collected.

[0123] According to an embodiment of the present disclosure, the present disclosure further provides an electronic device, comprising at least one processor; and a memory connected with the at least one processor in communication; the memory stores instructions executable 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 for determining the intersection passing duration.

[0124] According to an embodiment of the present disclosure, the present disclosure further provides a non-transitory computer readable storage medium storing computer instructions, wherein the computer instructions are used to enable a computer to perform the method for determining the intersection passing duration.

[0125] According to an embodiment of the present disclosure, the present disclosure further provides a computer program product comprising a computer program, wherein the computer program, when executed by a processor, implements the method for determining the intersection passing duration.

[0126] Figure 6 A schematic block diagram of an example electronic device 600 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the present disclosure described and / or claimed in this document.

[0127] As shown in Figure 6 The device 600 includes a computing unit 601 that can perform various appropriate actions and processes in accordance with a computer program stored in a read-only memory (ROM) 602 or a computer program loaded into a random access memory (RAM) 603 from a storage unit 608. Various programs and data required for the operation of the device 600 can also be stored in the RAM 603. The computing unit 601, the ROM 602, and the RAM 603 are connected to each other through a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0128] Various components in the device 600 are connected to the I / O interface 605, including an input unit 606, such as a keyboard, a mouse, etc., an output unit 607, such as various types of displays, speakers, etc., a storage unit 608, such as a magnetic disk, an optical disk, etc., and a communication unit 609, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 609 allows the device 600 to exchange information / data with other devices through a computer network, such as the Internet, and / or various telecommunication networks.

[0129] The computing unit 601 can be various general and / or special purpose processing components with processing and computing capabilities. Some examples of the computing unit 601 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 601 performs various methods and processes described above, such as the method of determining intersection passage duration. For example, in some embodiments, the method of determining intersection passage duration can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 608. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 600 via the ROM 602 and / or the communication unit 609. When the computer program is loaded onto the RAM 603 and executed by the computing unit 601, one or more steps of the method of determining intersection passage duration described above can be performed. Alternatively, in other embodiments, the computing unit 601 can be configured to perform the method of determining intersection passage duration by any other suitable means, such as by means of firmware.

[0130] Various implementations of the systems and techniques described above can be realized in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on a chip (SOC), a complex programmable logic device (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0131] Program code for carrying out methods of the present disclosure can be written in any combination of one or more programming languages. The program code can be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the program code, when executed by the processor or controller, produces a means for implementing the functions / acts specified in the flowcharts and / or block diagrams. The program code can be executed entirely on a machine, partially on a machine, partially on a machine as a stand-alone software package, partially on a machine and partially on a remote machine or entirely on a remote machine or server.

[0132] In the context of this disclosure, a machine-readable medium can be a tangible medium that contains or stores a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include but is not limited to an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, 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 disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0133] To provide for interaction with a user, the systems and techniques described here 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 a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, 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, speech, or tactile input.

[0134] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can 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.

[0135] The computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.

[0136] It should be understood that the various forms of flow shown above can be used to reorder, add, or remove steps. For example, the steps described in the present disclosure can be performed in parallel, in series, or in a different order, as long as the desired results of the technology disclosed in the present disclosure are achieved, which is not limited herein.

[0137] The specific implementation described above does not constitute a limitation on the protection scope of the present disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements, and improvements made within the spirit and principles of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for determining a duration of a traffic light at an intersection, comprising: obtaining object information of a target object, the object information comprising current position information of the target object and path planning information of the target object; determining a reference duration of a target intersection associated with the object information, wherein the reference duration is related to a historical duration of the target intersection; determining signal light information of a target signal light at the target intersection, the signal light information comprising a correspondence between light states and time of the target signal light after a current time; and determining a target duration of the target intersection according to the object information, the reference duration and the signal light information, so as to display the target duration; the determining of the signal light information of the target signal light at the target intersection comprises: estimating the signal light information according to cycle information of the target signal light and switching time information of the target signal light, wherein the cycle information comprises a duration of each light state in the signal light; the determining of the signal light information of the target signal light at the target intersection further comprises: determining first switching time information in the switching time information according to first historical trajectory data associated with the target intersection, wherein the first historical trajectory data comprises behavior data and time information corresponding to each trajectory point; determining cycle information of the target signal light according to second historical trajectory data associated with the target intersection, wherein a generation time of the second historical trajectory data is earlier than a generation time of the first historical trajectory data; the determining of the signal light information of the target signal light at the target intersection further comprises at least one of: determining first switching time information in the switching time information according to first historical trajectory data associated with the target intersection, wherein the first historical trajectory data comprises behavior data and time information corresponding to each trajectory point; and determining second switching time information in the switching time information according to image data of the target signal light; the determining of the target duration of the target intersection according to the object information, the reference duration and the signal light information comprises: determining the target duration according to target features, the object information, the signal light information and the reference duration; wherein the target features comprise at least one of: attribute features of a target road corresponding to the current position information, road condition features of the target road, reference signal light information for the target signal light; the determining of the target duration of the target intersection according to the object information, the reference duration and the signal light information comprises: inputting the object information, the reference duration and the signal light information into a deep learning model to obtain the target duration of the target intersection. ​ ​ ​ ​ ​ ​ ​ ​ 2. The method of claim 1, wherein, ​ ​ ​ ​ 3. The method of claim 1 or 2, wherein, ​ ​ ​ 4. The method of claim 1 or 2, wherein, ​ ​ The deep learning model is trained by using a plurality of training samples. Each training sample includes object information features, reference passing time features, and signal light information features. A label of the training sample is deviation information between the reference passing time and a passing time output by the deep learning model.

5. An apparatus for determining a passing time of an intersection, comprising: an obtaining module configured to obtain object information of a target object, the object information comprising current position information of the target object and path planning information of the target object; a first determining module configured to determine a reference passing time of a target intersection associated with the object information, wherein the reference passing time is related to a historical passing time of the target intersection; a second determining module configured to determine signal light information of a target signal light in the target intersection, the signal light information comprising a correspondence between light states and times of the target signal light after a current time; and a third determining module configured to determine a target passing time of the target intersection according to the object information, the reference passing time, and the signal light information, so as to display the target passing time. The second determining module comprises: an estimating submodule configured to estimate the signal light information according to cycle information of the target signal light and switching time information of the target signal light, wherein the cycle information comprises a duration of each light state in the signal light; a first determining submodule configured to determine first switching time information in the switching time information according to first historical trajectory data associated with the target intersection, wherein the first historical trajectory data comprises behavior data and time information corresponding to each trajectory point; a third determining submodule configured to determine the cycle information of the target signal light according to second historical trajectory data associated with the target intersection, wherein a generation time of the second historical trajectory data is earlier than a generation time of the first historical trajectory data.

6. The apparatus of claim 5, wherein, The second determining module further comprises at least one of the following: a first determining submodule configured to determine first switching time information in the switching time information according to first historical trajectory data associated with the target intersection, wherein the first historical trajectory data comprises behavior data and time information corresponding to each trajectory point; and a second determining submodule configured to determine second switching time information in the switching time information according to image data of the target signal light. The third determining module comprises:

7. The apparatus of claim 5 or 6, wherein, a fourth determining submodule configured to determine the target passing time according to target features, the object information, the signal light information, and the reference passing time, wherein the target features comprise at least one of the following: attribute features of a target road corresponding to the current position information, road condition features of the target road, and reference signal light information of the target signal light. The third determining module comprises: an input submodule configured to input the object information, the reference passing time, and the signal light information into a deep learning model to obtain the target passing time of the target intersection.

8. The apparatus of claim 5 or 6, wherein, ​ ​ The deep learning model is obtained by training a plurality of training samples. Each training sample includes object information features, reference passing time length features, and signal light information features. A label of the training sample is deviation information between the reference passing time length and a passing time length output by the deep learning model. 9.An electronic device, comprising: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable 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 to 4.

10. A non-transitory computer readable storage medium having stored thereon computer instructions, wherein, The computer instructions are used to enable the computer to perform the method of any one of claims 1 to 4. 11.A computer program product comprising a computer program which, when executed by a processor, implements the method of any one of claims 1 to 4.

Images