A method, device and related product for determining traffic signal change rule

Abstract

The application discloses a method and device for determining traffic signal change rule and related products. The type of lane that the target vehicle should enter before driving into a target intersection is determined according to the current path planning data of the target vehicle. The traffic signal at the target intersection whose change rule is to be analyzed is determined based on the lane type determined by the navigation path planning data. The stop area corresponding to the traffic signal can be determined according to the map data of the target intersection. The change rule of the traffic signal is determined based on the speed information of the sample vehicle in the stop area which is currently in or has been in the stop area before the target vehicle drives into the target intersection. When the target vehicle passes through the stop area of the traffic signal reached by navigation or is about to reach the stop area of the traffic signal, a warning prompt of the change of the traffic signal is provided for the driver of the target vehicle, so as to assist the driver to operate the target vehicle to safely wait or efficiently pass through the target intersection according to the indication state of the traffic signal.

Description

Technical Field

[0001] This application relates to the field of traffic technology, and in particular to a method, apparatus, and related products for determining the changing patterns of traffic lights. Background Technology

[0002] With the increasing number of cars in major cities and the growing traffic congestion, urban traffic problems are becoming increasingly prominent. Traffic lights play a crucial role in traffic management, enabling effective traffic control and significantly improving traffic flow, road capacity, and reducing traffic accidents. Traffic lights generally include red and green lights; red indicates a prohibition to proceed, and green indicates a permit to proceed. Pedestrians and drivers can perform corresponding actions at intersections based on the traffic light signals, such as proceeding or waiting. In vehicle navigation scenarios, to provide better navigation services and assist drivers in adhering to traffic regulations along the navigation route, it is necessary to analyze the patterns of traffic light changes at intersections. Summary of the Invention

[0003] Based on the above problems, this application provides a method, device and related products for determining the changing patterns of traffic lights. The purpose is to provide accurate traffic light prompts to drivers while providing navigation services by exploring the changing patterns of traffic lights, so as to assist drivers in obeying traffic regulations and driving safely.

[0004] The embodiments of this application disclose the following technical solutions:

[0005] The first aspect of this application provides a method for determining the changing patterns of traffic lights, the method comprising:

[0006] The type of lane the target vehicle should enter before entering the target intersection is determined based on the target vehicle's current route planning data;

[0007] Based on the aforementioned type, determine the traffic lights at the target intersection whose changing patterns need to be analyzed;

[0008] The stop area corresponding to the traffic light is determined based on the map data of the target intersection;

[0009] Based on the speed information of sample vehicles that were or had been in the stop area before the target vehicle entered the target intersection, the change pattern of the traffic lights is determined.

[0010] In an optional implementation, determining the traffic light change pattern based on the speed information of sample vehicles that were or had been in the stop area before the target vehicle entered the target intersection includes:

[0011] Based on the location information of the stop area and the vehicle's positioning data, multiple vehicles that are currently in or have been in the stop area are respectively used as sample vehicles.

[0012] The instantaneous speeds of multiple sample vehicles when they are in the stopping area and the corresponding vehicle speed acquisition time are obtained.

[0013] The instantaneous speed and the corresponding vehicle speed acquisition time are constructed into a vehicle speed time series data pair to obtain the vehicle speed time series data set of the traffic light; the vehicle speed change data set includes multiple constructed vehicle speed time series data pairs.

[0014] The changing pattern of the traffic lights is determined based on the vehicle speed time series data set.

[0015] In an optional implementation, determining the changing pattern of the traffic lights based on the vehicle speed time-series data set includes:

[0016] The instantaneous speed of each vehicle speed time-series data pair in the vehicle speed time-series data set is compared with a preset speed threshold. The vehicle speed time-series data pairs in the vehicle speed time-series data set are divided into a first type of vehicle speed time-series data pair and a second type of vehicle speed time-series data pair. The instantaneous speed of the first type of vehicle speed time-series data pair is less than the preset speed threshold. The instantaneous speed of the second type of vehicle speed time-series data pair is greater than or equal to the preset speed threshold.

[0017] A first indication state timing data pair of the traffic light is generated based on the first type of vehicle speed timing data pair, and a second indication state timing data pair of the traffic light is generated based on the second type of vehicle speed timing data pair; the first indication state timing data pair includes a first value and a vehicle speed acquisition time, and the second indication state timing data pair includes a second value and a vehicle speed acquisition time; the first value and the second value are different, the first value is used to indicate that the traffic light is in a first indication state where vehicles are prohibited from passing, and the second value is used to indicate that the traffic light is in a second indication state where vehicles are allowed to pass;

[0018] The traffic light's indication state timing data set is constructed based on multiple first indication state timing data pairs and multiple second indication state timing data pairs.

[0019] The changing patterns of the traffic lights are determined by analyzing the time-series dataset of the indicated states.

[0020] In an optional implementation, determining the changing pattern of the traffic light by analyzing the time-series dataset of the indicated state includes:

[0021] Identify data from multiple cyclic periods within the indicated state timing data set;

[0022] The data from the identified N cycle periods are placed into the same cycle period and superimposed to form the superimposed indicator state cycle change data; where N is a positive integer greater than 1.

[0023] The change pattern of the traffic light is determined based on the jump pattern between the first and second values ​​in the superimposed indicator state cycle data.

[0024] In an optional implementation, where N is an odd number, the step of superimposing the data from the identified N cycle periods into the same cycle to form superimposed indicator state cycle change data includes:

[0025] The data from N identified cycles are superimposed within the same cycle. If the values ​​used to indicate the state of the traffic light at the same target time in the N cycles are inconsistent, then the number K of cycles corresponding to the same target time using the first value and the number Q of cycles corresponding to the same target time using the second value are obtained; where N = K + Q.

[0026] Determine the relative magnitudes of K and Q. If K is greater than Q, then set the value used to indicate the state of the traffic light corresponding to the same target time within the same cycle to the first value; if K is less than Q, then set the value used to indicate the state of the traffic light corresponding to the same target time within the same cycle to the second value, thus forming superimposed indicator state change data within the cycle.

[0027] In an optional implementation, identifying data from multiple cyclic periods in the indicated state time-series data set includes: using a sliding window method to determine whether a periodic pattern is identified in the indicated state time-series data set before the sliding window has accumulated to a preset number of steps; if yes, then obtaining data from multiple cyclic periods in the indicated state time-series data set based on the identified periodic pattern; if no, then increasing the fault tolerance threshold for identifying the periodic pattern in the indicated state time-series data set using the sliding window method, and identifying the periodic pattern in the time-series change data based on the increased fault tolerance threshold to obtain data from multiple cyclic periods in the indicated state time-series data set.

[0028] In an optional implementation, after determining the traffic light change pattern based on the speed information of sample vehicles that were or had been in the stop area before the target vehicle entered the target intersection, the method further includes:

[0029] If the traffic light is in a first indication state prohibiting vehicles from passing when the target vehicle enters the stopping area, then based on the determined change pattern of the traffic light, the driver of the target vehicle is prompted with the remaining duration of the first indication state, and the driver is prompted to prepare at a prompt time before the end of the first indication state; the prompt time is the xth second before the end of the first indication state, where x is a positive number and x is less than the duration during which the traffic light has been continuously in the first indication state;

[0030] If the traffic light is in a second indication state that allows vehicles to pass when the target vehicle is about to enter the stop area, then when there are only y seconds left in the second indication state, the driver will be notified that there are only y seconds left in the second indication state, where y is a positive number and y is less than the duration for which the traffic light has been in the second indication state continuously.

[0031] In an optional implementation, determining the stop area corresponding to the traffic light based on the map data of the target intersection includes:

[0032] If the map data achieves the accuracy of locating the stop line of the target intersection, then the stop area is determined based on the stop line corresponding to the traffic light and the lane width of the target intersection.

[0033] If the map data does not achieve the accuracy required to locate the stop line of the target intersection, then the location information of the center point of the target intersection is determined based on the map data, and the location of the stop line corresponding to the traffic light at the target intersection is estimated based on the location information of the center point and the width of the two lanes. The stop area is then determined based on the stop line corresponding to the traffic light and the lane width of the target intersection.

[0034] The first boundary of the stop area is a boundary that extends beyond the stop line by one lane width, and the second boundary of the stop area is a boundary that does not extend beyond the stop line by two lane widths. The first boundary and the second boundary are parallel.

[0035] A second aspect of this application provides an apparatus for determining the changing patterns of traffic lights, the apparatus comprising:

[0036] The lane type determination module is used to determine the type of lane that the target vehicle should enter before entering the target intersection based on the target vehicle's current path planning data;

[0037] A traffic light determination module is used to determine the traffic lights at the target intersection whose changing patterns need to be analyzed based on the type described above.

[0038] The stop area determination module is used to determine the stop area corresponding to the traffic light based on the map data of the target intersection.

[0039] The traffic light change pattern determination module is used to determine the change pattern of the traffic lights based on the speed information of sample vehicles that were or had been in the stop area before the target vehicle entered the target intersection.

[0040] A third aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements a method for determining the changing patterns of traffic lights as provided in any implementation of the first aspect.

[0041] The fourth aspect of this application provides a processor for running a computer program that, when running, executes a method for determining the changing patterns of traffic lights as provided in any implementation of the first aspect.

[0042] Compared with the prior art, this application has the following beneficial effects:

[0043] In this technical solution, the first step is to determine the lane type the target vehicle should enter before entering the target intersection based on its current route planning data. Since lane types are related to traffic lights—for example, a traffic light displaying a straight-ahead arrow corresponds to a lane type that allows straight-ahead movement—the lane type determined based on the navigation route planning data can be used to identify the traffic lights at the target intersection whose changing patterns are to be analyzed. After identifying the traffic lights whose changing patterns are to be analyzed, since traffic lights have corresponding stop areas in geographic space, the stop areas corresponding to the traffic lights can be determined based on the map data of the target intersection. After determining the stop areas, the changing patterns of the traffic lights can be determined by analyzing the speed information of sample vehicles that were or had been in the stop areas before the target vehicle entered the target intersection. In this technical solution, by analyzing the traffic light change patterns through the above steps, a warning message about traffic light changes can be provided to the driver of the target vehicle when it arrives at or is about to arrive at the stop area of ​​the traffic light via navigation. This assists the driver in safely waiting or efficiently passing through the target intersection according to the traffic light signals. Compared to existing navigation technologies, this solution also improves the quality of navigation services and enhances the driver's driving experience along the planned navigation route. Attached Figure Description

[0044] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 A flowchart illustrating a method for determining the changing patterns of traffic lights, provided in an embodiment of this application;

[0046] Figure 2 A schematic diagram of a stop area at an intersection provided in an embodiment of this application;

[0047] Figure 3 A schematic diagram of another intersection stop area provided in an embodiment of this application;

[0048] Figure 4 A schematic diagram of another intersection stop area provided in the embodiments of this application;

[0049] Figure 5A A vehicle speed timing diagram provided in an embodiment of this application;

[0050] Figure 5B A flowchart illustrating another method for determining the changing patterns of traffic lights provided in this application embodiment;

[0051] Figure 6 A timing diagram of the indication state of a traffic light is provided as an embodiment of this application;

[0052] Figure 7A A partial vehicle speed timing diagram provided for an embodiment of this application;

[0053] Figure 7B A vehicle speed timing diagram for another part of the embodiments of this application;

[0054] Figure 7C Based on Figure 7A The timing diagram of the indication state generated by the vehicle speed timing sequence;

[0055] Figure 7D Based on Figure 7B The timing diagram of the indication state generated by the vehicle speed timing sequence;

[0056] Figure 8 Another vehicle speed timing diagram provided in the embodiments of this application;

[0057] Figure 9 Another timing diagram of the indication state of a traffic light provided in an embodiment of this application;

[0058] Figure 10 This is a schematic diagram of a device for determining the changing pattern of traffic lights, provided in an embodiment of this application. Detailed Implementation

[0059] Currently, navigation systems lack warnings about traffic light status, forcing drivers to rely on visual color perception near intersections to determine whether it's safe to proceed. Impatient or novice drivers may misjudge speed and traffic light patterns, leading to failed attempts to run red lights, resulting in license violations, points deductions, and potentially accidents. In other situations, the lack of traffic light information in navigation systems, coupled with driver inattention, can also lead to low driving efficiency, slow intersection movement, and localized or widespread traffic congestion.

[0060] In view of this, this application proposes to determine the type of lane a target vehicle should enter before entering a target intersection by using navigation route planning data; then, based on this lane type, determine the traffic light whose changing patterns are to be analyzed, and further determine the stop area corresponding to the traffic light. Based on the stop area, sample vehicles can be identified. The speeds of these sample vehicles, which were in or had been in the stop area before the target vehicle entered the target intersection, can serve as the data basis for mining the changing patterns of the traffic light. Based on this, the changing patterns of the traffic light can be effectively determined, thereby assisting the driver of the target vehicle to drive safely or effectively through the intersection in accordance with traffic regulations. This solution can assist drivers and improve their driving experience, and can also help alleviate the problems of delayed vehicle starts or failure to run a green light in congested traffic conditions, thus improving road driving safety.

[0061] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0062] See Figure 1 The figure is a flowchart of a method for determining the changing pattern of traffic lights according to an embodiment of this application. Figure 1 As shown, methods for determining the changing patterns of traffic lights include:

[0063] S101. Determine the type of lane the target vehicle should enter before entering the target intersection based on the target vehicle's current route planning data.

[0064] Currently, vehicle navigation technology can generally provide route guidance. For example, a vehicle starts from the starting point, continues straight along road A, turns left at the third intersection onto road B, continues along road B for 2 kilometers, enters a ramp, drives 200 meters on the ramp, exits the ramp onto main road C, drives 2 kilometers along main road C, turns right at the second intersection onto road D, and drives 200 meters on road D to reach the destination.

[0065] As illustrated in the examples above, when a driver provides their destination on the vehicle's navigation software, the software can plan a route from the starting point (e.g., the vehicle's current location) to the destination based on the vehicle's location information. This route planning data includes all the roads traversed and suggested driving lanes for each road. Lane suggestions generally correspond to the driving direction to be observed at upcoming intersections. For example, if the suggested lane is a right-turn lane, the driver is generally instructed to turn right at the upcoming intersection; similarly, if the suggested lane is a left-turn lane, the driver is generally instructed to turn left at the upcoming intersection; and if the suggested lane is a straight-ahead lane, the driver is generally instructed to go straight at the upcoming intersection. In practical applications, some lanes allow multiple operations, such as lanes that allow straight and right turns, lanes that allow straight and left turns, and lanes that allow left turns and U-turns; of course, some lanes are restricted to only one operation at intersections, such as lanes that restrict left turns only, lanes that restrict straight turns only, lanes that restrict right turns only, and lanes that restrict U-turns only.

[0066] In this technical solution, the vehicle that needs to understand the traffic light change patterns at intersections encountered along the navigation route is referred to as the target vehicle. The navigation software provides navigation services to the driver of the target vehicle, generating route planning data. This navigation software can be software on the vehicle's infotainment system or software installed on the driver's portable terminal device (e.g., a mobile phone); there is no limitation here. In this step, based on the route planned in the target vehicle's current route planning data, the type of lane the target vehicle should enter before entering the target intersection can be determined. For example, a left-turn lane, a straight-ahead lane, or a right-turn lane. Since right-turn operations at many intersections are not restricted by traffic light changes, this solution focuses on traffic light change patterns for lane types, specifically considering the situations of left-turn lanes and straight-ahead lanes. Of course, for intersections where right-turn lanes are controlled by traffic lights, the subsequent operations below can also be performed with reference to the right-turn lane type.

[0067] S102. Based on the type, determine the traffic lights at the target intersection to be analyzed for their changing patterns.

[0068] In practical applications, lanes near the target intersection are marked with the permitted operating procedures for vehicles to pass through that intersection. Examples include left-turn arrows, straight-ahead arrows, and right-turn arrows on the lane markings. In everyday traffic, drivers can refer to these signs or the arrow markings on the road surface to understand the possible actions (left turn, continue straight, or right turn) that vehicles in each lane may take upon entering the intersection, thus better planning their driving maneuvers. Visually, drivers can also generally distinguish whether to stop or proceed based on traffic lights. For example, if a driver is in a left-turn lane and the left-turn indicator light is red, it means that vehicles are prohibited from turning left at the intersection at that moment; conversely, if a driver is in a straight-ahead lane and the straight-ahead indicator light is green, it means that vehicles are permitted to proceed straight through the intersection at that moment.

[0069] As can be seen, each type of lane has a corresponding traffic light with a specific pattern of change to be analyzed. For example, in the above example, the traffic light with the specific pattern of change to be analyzed for the left-turn lane is the left-turn indicator light, and the traffic light with the specific pattern of change to be analyzed for the straight-ahead lane is the straight-ahead indicator light. In some possible implementation scenarios, multiple types of lanes may also correspond to the same traffic light with the same pattern of change to be analyzed. For example, on a traffic light, if the red light is on, neither left turn nor straight-ahead is permitted; if the green light is on, left turn, straight-ahead, and right turn are all permitted. The correspondence between lane types and traffic lights with specific patterns of change to be analyzed depends on the type of traffic lights at the intersection. Different types of traffic lights may have slightly different correspondences, which are not limited here.

[0070] S103. Determine the stop area corresponding to the traffic lights based on the map data of the target intersection.

[0071] Figure 2 This is a schematic diagram of a stop area at an intersection, provided as an embodiment of this application. Figure 2 The overhead view of the intersection shown indicates that there are 8 traffic lights at the intersection, with one traffic light that pedestrians need to pay attention to in each viewing direction (e.g., ...). Figure 2 (As shown in circle 201 from a top-down view) and traffic lights that drivers need to pay attention to (such as...) Figure 2 (As shown in the top-down view, ellipse 202). This design focuses on traffic lights 202 that drivers need to pay attention to, attempting to uncover the patterns of change in these traffic lights. For example... Figure 2 As shown, opposite the traffic light 202 that the driver needs to pay attention to, there is a corresponding stop line 203. The stop area is an area located near the stop line, and it is not necessarily actually drawn on the road surface. In this scheme, the stop area corresponding to the traffic light is regarded as the area where vehicles would generally stop when they encounter a traffic light indicating that passage is prohibited. Figure 2 The dashed rectangle in the center shows the parking area 204 near parking line 203.

[0072] In the optional implementation, the stop area corresponding to the traffic light is determined based on the map data of the target intersection, including:

[0073] If the map data achieves the accuracy required to locate the stop line at the target intersection (e.g., a high-precision map), the stop line's location and lane width can be obtained from the high-precision map. The stopping area can be determined based on the stop line corresponding to the traffic lights and the lane width of the target intersection. For example, the boundary extending one lane width beyond the stop line can be designated as the first boundary, the boundary extending two lane widths within the stop line as the second boundary, and the other two boundaries can be the third and fourth boundaries. The third boundary can be the dividing line between two traffic flow directions on the road. The fourth boundary can be the right boundary of the road's traffic direction. The first and second boundaries are parallel, and the third and fourth boundaries are parallel. Figure 3 This is a schematic diagram of another intersection stop area provided in an embodiment of this application. Figure 3 In the diagram, lines a, b, c, and d represent the first, second, third, and fourth boundaries of the parking area, respectively.

[0074] If the map data is not accurate enough to pinpoint the stop line at the target intersection, the location of the center point of the intersection can be determined based on the map data. Then, the location of the stop line corresponding to the traffic lights at the target intersection can be estimated based on the center point's location and the widths of the two lanes. Next, the stopping area is determined based on the stop line corresponding to the traffic lights and the lane widths of the target intersection. For example, the boundary extending one lane width beyond the stop line can be used as the first boundary, and the boundary not extending two lane widths beyond the stop line can be used as the second boundary.

[0075] exist Figure 3 The stop area shown does not have fine-grained divisions for different lanes facing the same traffic flow direction. Understandably, the stop area could be divided more precisely for different lanes. Figure 4 This is a schematic diagram of another intersection stop area provided in an embodiment of this application. Figure 4 Stop areas 41, 42, and 43 correspond to lanes 1, 2, and 3, respectively. Vehicles traveling in lanes 1, 2, and 3 all need to refer to the same traffic light; therefore, stop areas 41, 42, and 43 all correspond to the same traffic light.

[0076] S104. Based on the speed information of sample vehicles that were or had been in the stop area before the target vehicle entered the target intersection, determine the changing pattern of traffic lights.

[0077] In real-world traffic scenarios, when a vehicle approaches the stopping area at an intersection, if the traffic light for the corresponding lane indicates "allowed passage," the vehicle speed generally won't drop too low to ensure traffic efficiency; for example, it will pass through the intersection at a speed exceeding 20 km / h. However, if the traffic light for the corresponding lane indicates "prohibited passage," due to traffic regulations and safety considerations, the vehicle speed will generally be reduced to close to 10 km / h or even below 5 km / h. Therefore, based on the vehicle's speed, the pattern of the traffic light's red-green changes can be determined. This includes factors such as the traffic light cycle, the duration of the red light, and the duration of the green light.

[0078] In this scheme, in order to provide better navigation services for the driver of the target vehicle in the traffic navigation scenario, the speed information of sample vehicles that are currently or were previously in the stop area before the target vehicle enters the target intersection is used as a reference to determine the change pattern of the traffic lights to be analyzed. Figure 5A This application provides a vehicle speed time-series diagram. By analyzing this diagram, it's easy to see that by examining the correspondence between the instantaneous vehicle speeds and the speed collection times of multiple sample vehicles, a pattern of change can be identified. Since changes in vehicle speed also reflect the indication status of traffic lights, therefore, combining... Figure 5AThe speed time sequence diagram shown can also be used to analyze the changing patterns of traffic lights.

[0079] In this technical solution, by analyzing the traffic light change patterns through the above steps, a warning message about traffic light changes can be provided to the driver of the target vehicle when it arrives at or is about to arrive at the stop area of ​​the traffic light via navigation. This assists the driver in safely waiting or efficiently passing through the target intersection according to the traffic light signals. Compared to existing navigation technologies, this solution also improves the quality of navigation services and enhances the driver's driving experience along the planned navigation route.

[0080] Figure 5B This is a flowchart illustrating another method for determining the changing pattern of traffic lights, provided in an embodiment of this application. The embodiment described in this figure provides a more detailed explanation of how to determine the changing pattern of traffic lights.

[0081] like Figure 5B The method for determining traffic lights shown includes the following steps:

[0082] S501. Determine the type of lane the target vehicle should enter before entering the target intersection based on the target vehicle's current route planning data.

[0083] S502, Traffic lights at the target intersection whose change patterns are to be analyzed are determined based on their type.

[0084] S503. Determine the stop area corresponding to the traffic lights based on the map data of the target intersection.

[0085] The above S501 to S503 are basically the same as the implementation of S101 to S103 in the method embodiment described above, and will not be repeated here. The relevant technical implementation can be referred to the above description of S101 to S103.

[0086] S504. Based on the location information of the stopping area and the vehicle positioning data, multiple vehicles that are currently in or have been in the stopping area are respectively used as sample vehicles.

[0087] Since many vehicles now have positioning capabilities, real-time vehicle location data can be obtained through vehicle navigation software. Furthermore, parking areas also have corresponding location information. Therefore, based on the location information of the parking area and the vehicle's positioning data, it can be determined whether a vehicle is within the stop area corresponding to the traffic light whose changing patterns are to be analyzed at a given location time. In this embodiment, to uncover the changing patterns of traffic lights, multiple vehicles currently or previously in the stop area are considered as sample vehicles. Because these sample vehicles have stopped or are currently stopping in the stop area or have briefly stopped before passing through, their driving behavior in the stop area can reflect the changing patterns of the traffic lights. For example, a decrease in vehicle speed to a certain extent indicates that the traffic light is in a "no passage" indication; no decrease in speed or a slight decrease in speed indicates that the traffic light is in a "pass" indication.

[0088] S505. Obtain the instantaneous speed of multiple sample vehicles when they are in the stationary area and the corresponding vehicle speed acquisition time.

[0089] For these sample vehicles, this solution proposes to extract the instantaneous speed of the sample vehicle when it is in the stationary area, as well as the corresponding vehicle speed acquisition time. If the vehicle speed is acquired simultaneously with the positioning, the vehicle speed acquisition time is also the positioning time mentioned above.

[0090] S506. The instantaneous speed and the corresponding vehicle speed acquisition time are constructed into a vehicle speed time series data pair to obtain the vehicle speed time series data set of traffic lights.

[0091] The vehicle speed change dataset consists of multiple constructed vehicle speed time-series data pairs. Each vehicle speed time-series data pair includes two pieces of information: instantaneous speed and the vehicle speed acquisition time.

[0092] For example, the format of a vehicle speed time-series data pair is (v, t), where v represents the instantaneous speed and t represents the vehicle speed acquisition time. The vehicle speed change data set, constructed from the instantaneous speeds and acquisition times of the sample vehicles when they are in the stationary area, contains vehicle speed time-series data pairs from different vehicles. Combining this vehicle speed change data set, the variation pattern of traffic signals can be further determined. The following sections (S507-S510) describe the implementation process of determining the variation pattern of traffic lights based on the vehicle speed time-series data set in the technical solution of this application.

[0093] S507. Compare the instantaneous speed of each vehicle speed time series data pair in the vehicle speed time series data set with the preset speed threshold, and divide each vehicle speed time series data pair in the vehicle speed time series data set into a first type of vehicle speed time series data pair and a second type of vehicle speed time series data pair.

[0094] In this application's technical solution, for each vehicle speed time-series data pair in the vehicle speed time-series data set, the instantaneous speed is compared with a preset speed threshold to determine whether it is greater or less than the threshold. The data pairs are then classified into either a first-class or a second-class vehicle speed time-series data pair. For example, if a vehicle speed time-series data pair has an instantaneous speed of 20 km / h and a preset speed threshold of 5 km / h, then this vehicle speed time-series data pair is classified as a second-class pair; if a vehicle speed time-series data pair has an instantaneous speed of 4 km / h and a preset speed threshold of 5 km / h, then this vehicle speed time-series data pair is classified as a first-class pair. In this solution, the instantaneous speed in the first-class vehicle speed time-series data pairs is less than the preset speed threshold, while the instantaneous speed in the second-class vehicle speed time-series data pairs is greater than or equal to the preset speed threshold. The preset speed threshold can be set according to actual needs, and there is no limitation here. In this solution, if the instantaneous speed is less than the preset speed threshold, it means that the vehicle is equivalent to being stationary or nearly stationary; otherwise, it is considered that the vehicle has a tendency to move.

[0095] S508. Generate a first indication state timing data pair for the traffic light based on the first type of vehicle speed timing data, and generate a second indication state timing data pair for the traffic light based on the second type of vehicle speed timing data.

[0096] In the technical solution of this application, by executing step S508, based on the aforementioned first type of vehicle speed time-series data pairs and second type of vehicle speed time-series data pairs, a first indication state time-series data pair and a second indication state time-series data pair regarding the traffic light's changing pattern to be analyzed are generated. The first type of vehicle speed time-series data pair and the second type of vehicle speed time-series data pair reflect the relationship between vehicle speed and time; the first indication state time-series data pair and the second indication state time-series data pair reflect the relationship between the traffic light's indication state and time. It can be seen that step S508 completes the conversion from the relationship between vehicle speed and time to the relationship between indication state and time.

[0097] In one possible timing method, the execution process of S508 is as follows: The instantaneous speed in the first type of vehicle speed timing data pair is set to a first value (e.g., the first value is 0), while the time information remains unchanged, thus generating a first indication state timing data pair; the first indication state timing data pair includes the first value and the vehicle speed acquisition time. Similarly, the instantaneous speed in the second type of vehicle speed timing data pair is set to a second value (e.g., the second value is 1), while the time information remains unchanged, thus generating a second indication state timing data pair; the second indication state timing data pair includes the second value and the vehicle speed acquisition time. The first value and the second value are different; the first value indicates that the traffic light is in a first indication state where vehicles are prohibited from passing, while the second value indicates that the traffic light is in a second indication state where vehicles are permitted to pass. Both the first value and the second value can be customized.

[0098] S509. Construct a set of traffic signal indicator timing data based on the generated multiple first indicator state timing data pairs and multiple second indicator state timing data pairs.

[0099] The multiple first indication state timing data pairs and multiple second indication state timing data pairs generated after step S508 above may be continuously or interleaved in the time dimension. To more accurately uncover the changing patterns of the traffic lights, this step unifies these first and second indication state timing data pairs into a single set, forming an indication state timing data set. This set contains multiple indication state timing data pairs with different indication states. Furthermore, this forms the indication state timing diagram of the traffic lights. Figure 6 This application provides a timing diagram of the indication state of a traffic light, which is related to... Figure 5A Corresponding. Combining Figure 6 It is not difficult to see that by analyzing the timing of the indication states, the duration and cyclical changes of traffic lights under two different indication states can be identified. It should be noted that step S509 is optional; either the aforementioned timing data set of indication states can be constructed, or multiple pairs of first and second indication state timing data can be directly used to mine the patterns of change.

[0100] S510. Determine the changing patterns of traffic lights by analyzing the timing dataset of the indicated states.

[0101] In practice, this step arranges the data pairs in the indicated state time-series data set according to time sequence. In addition to time, there are two values ​​on another dimension (representing the indicated state): a first value and a second value. Based on this, data from multiple cyclical periods in the indicated state time-series data set are identified. The cycle period reflects the changing pattern of the traffic lights.

[0102] In practical applications, data sparsity may exist. For example, only six pairs of vehicle speed time-series data might be collected within 20 seconds, resulting in only six pairs of indicator status time-series data generated within that 20-second period. Understandably, excessively sparse data may lead to inaccurate patterns of change. Increasing the amount of data reflects more reliable and credible patterns of change. To address the issue of data coefficients, this solution can superimpose data from N identified cycles into a single cycle to form superimposed indicator status cycle change data. This enhances the algorithm. Here, N is a positive integer greater than 1. For example, initially identifying 20 cycles of data, superimposing five cycles' data allows us to reflect the data volume and characteristics of five cycles within a single cycle. Then, based on the jump pattern between the first and second values ​​in the superimposed indicator status cycle change data, the traffic light's change pattern is determined.

[0103] Data points from multiple cycles are merged into a single cycle, i.e., superimposed according to the relative time offset position of each cycle. The same offset time points are superimposed together, allowing for time granularity accurate to the second. If data from different cycles participating in the fusion (superposition) have different indication states at the same superimposed time point, a voting method can be used to determine the indication state. Therefore, an odd number of superimposed cycles is preferable, reducing the probability of a tie. In an optional implementation, N is odd. The data from N identified cycles are placed within the same cycle and superimposed to form the superimposed indication state cycle change data. This includes: placing the data from N identified cycles within the same cycle and superimposing them; if the values ​​used to indicate the traffic light state at the same target time within the N cycles are inconsistent, then the number K of cycles corresponding to the same target time indicating the traffic light state with the first value and the number Q of cycles corresponding to the same target time indicating the traffic light state with the second value are obtained. Where N = K + Q. The relative magnitudes of K and Q are determined. If K is greater than Q, the value indicating the traffic light's state at the same target time within the same cycle is set to the first value. If K is less than Q, the value indicating the traffic light's state at the same target time within the same cycle is set to the second value, forming a superimposed data set of changes in the indicated state within the cycle. This approach results in denser time-series points, making it easier to uncover real patterns.

[0104] The above overlay operation is mainly performed on data from multiple cycles within the indicator state time series dataset. Additionally, the overlay operation can also be performed on vehicle speed time series datasets. The resulting overlay is then converted into an indicator state time series diagram. Figure 7A and Figure 7B The two speed time-series diagrams are shown separately, each corresponding to one of the two cycles. Figure 7C and Figure 7D Corresponding to Figure 7A and Figure 7B , for in Figure 7A and Figure 7B The resulting timing diagram of the indicator states. Combined with... Figure 7A and Figure 7B Perform data overlay (data augmentation algorithm processing) to obtain Figure 8 The vehicle speed timing diagram shown. Combined with... Figure 7A , Figure 7B and Figure 8 Analysis makes it easy to see Figure 8 The vehicle speed time series data shown appears to be different at some data-sparse time points compared to Figure 7A and Figure 7B It has improved somewhat. Figure 8 Generate an indication state timing diagram based on this, such as Figure 9 As shown. Comparison Figure 7C , Figure 7D and Figure 9 As shown in the timing diagram, it is not difficult to see that by superimposing the vehicle speed timing data, the final timing diagram of the indicated state is different from that before superposition. Figure 9 The displayed indicator status timing compared to Figure 7C and Figure 7D More accurate.

[0105] The above steps describe a complete enhancement algorithm. Subsequently, the overall cycle period can be determined based on the newly identified pattern, and then superimposed according to the new overall cycle. After multiple rounds of the above cycle and correction, the discovered change pattern is closer to the real change pattern.

[0106] The above outlines the overall approach to the algorithm enhancement. A key challenge is determining the traffic light cycle. Once the cycle is determined, overlaying data with the same offset across multiple cycles is relatively straightforward. However, in some possible implementations, if the requirements for identifying periodic patterns in the indicator state time-series data are strictly controlled, it may lead to difficulties in effectively identifying the changing patterns of the traffic lights. To avoid this problem, this solution proposes adjusting the fault tolerance threshold for identifying periodic patterns in the indicator state time-series data. Therefore, identifying data across multiple cycles in the indicator state time-series data can include: using a sliding window method to determine whether a periodic pattern has been identified before the sliding window reaches a preset number of steps (e.g., 100 steps). If so, the data from multiple cycles in the indicator state time-series data is obtained based on the identified periodic pattern. If not, the fault tolerance threshold for identifying periodic patterns in the time series data set of the indication state using the sliding window method is increased, and the periodic patterns in the time series change data are identified based on the increased fault tolerance threshold to obtain data from multiple cyclic periods in the time series data set of the indication state.

[0107] The following is a detailed explanation of the period determination algorithm:

[0108] Traffic light cycles can be simple single cycles, such as: 10, 12; 10, 12; 10, 12 (10 is the red light duration, 12 is the green light duration, red and green lights alternate, in seconds, clearly 10, 12 form a cycle). Alternatively, traffic lights may have more complex multi-cycle cycles, such as: 10, 12, 13, 11, 12, 11; 10, 12, 13, 11, 12, 11; 10, 12, 13… (the red and green light alternation rules are the same, where 10, 12, 13, 11, 12, 11 are multiple cycles combined into a single cycle).

[0109] The sliding window method can be used to determine a cycle. This method first divides the sequence into fixed-length windows at the beginning, then slides the window forward one step at a time until the window repeats the data in the first window, indicating a cycle in the sequence. Taking a more complex sequence as an example, let the window length be 8. The first window is 10, 12, 13, 11, 12, 11, 10, 12 (in seconds). Moving the window forward one step results in 12, 13, 11, 12, 11, 10, 12, 13, and so on. After six moves, the window is 10, 12, 13, 11, 12, 11, 10, 12. At this point, the window repeats the first one, indicating a cycle in the sixth position. Therefore, the first six moves constitute a cycle, namely 10, 12, 13, 11, 12, 11. This method is effective for both single and multiple cycles, and the effectiveness of cycle detection can be controlled by adjusting the window size. Setting a larger window helps improve the accuracy of the data mining; setting a smaller window can improve the efficiency of mining periodic patterns.

[0110] The initially determined traffic light durations may be inaccurate, making it impossible to find cyclical patterns. To address this, if the window moves forward a specified number of steps (e.g., 100 steps) and still fails to find a cycle, the window matching criteria can be relaxed. Each traffic light duration within a certain threshold range can be considered identical. For example, if the first window has durations of 10, 12, and 13, and a subsequent window also has durations of 10, 10, and 13, with a threshold set to 2, then because the second duration of 10 differs from the second duration of 12 in the first window within the threshold, these two windows can be considered duplicates, thus determining the cycle. In this application's technical solution, by setting a tolerance threshold for the window, the recognition efficiency is improved, enhancing the efficiency of uncovering the periodic change patterns of traffic lights.

[0111] In an optional implementation of the technical solution of this application, after determining the traffic light change pattern based on the speed information of sample vehicles that were or had been in the stop area before the target vehicle entered the target intersection, the method further includes:

[0112] If the traffic light is in the first indication state (prohibiting vehicles) when the target vehicle enters the stop area, the driver of the target vehicle will be informed of the remaining duration of the first indication state based on the determined traffic light change pattern. The driver will be prompted to prepare at the last second before the end of the first indication state. The prompt time is the x-second countdown before the end of the first indication state, where x is a positive number and less than the duration the traffic light has been in the first indication state continuously. For example, if x is 3, the driver will be prompted to prepare at the third-to-last second.

[0113] If the traffic light is in the second indication state (allowing vehicles to proceed) when the target vehicle is about to enter the stop area, the driver will be notified when only y seconds remain in the second indication state. y is a positive number and is less than the duration the traffic light has been in the second indication state continuously. For example, if y is 10, the driver will be notified of the remaining time of the green light when there are only 10 seconds left before it turns red. This allows the driver to consider their vehicle speed, distance from the intersection, and the situation of pedestrians and other vehicles on the road to determine whether to adjust their current speed.

[0114] By executing the steps described above, navigation software can analyze the traffic light status and duration at intersections, providing drivers with early warnings and improving the following two issues: ① the slow start caused by drivers' inability to promptly perceive traffic light changes while waiting at red lights due to their eyes being off-road; ② insufficient estimation of remaining time at green lights, leading to failed green light activation and necessitating sudden braking at intersections (this problem is more pronounced at intersections with short yellow lights). This improves overall traffic efficiency and safety.

[0115] Based on the method for determining the changing patterns of traffic lights provided in the foregoing embodiments, this application also provides an apparatus for determining the changing patterns of traffic lights. For example... Figure 10 This is a schematic diagram of a device for determining the changing pattern of traffic lights, provided as an embodiment of this application. Figure 10 As shown, the device for determining the traffic light change pattern includes the following components: lane type determination module W1, traffic light determination module W2, stop area determination module W3, and traffic light change pattern determination module W4.

[0116] Among them, the lane type determination module W1 is used to determine the type of lane that the target vehicle should enter before entering the target intersection based on the target vehicle's current path planning data;

[0117] Traffic light determination module W2 is used to determine the traffic lights at the target intersection whose change patterns need to be analyzed based on the type described above.

[0118] The stop area determination module W3 is used to determine the stop area corresponding to the traffic light based on the map data of the target intersection.

[0119] The traffic light change pattern determination module W4 is used to determine the change pattern of the traffic lights based on the speed information of sample vehicles that were or had been in the stop area before the target vehicle entered the target intersection.

[0120] Optionally, the signal light change pattern determination module includes:

[0121] The sample vehicle determination unit is used to identify multiple vehicles that are currently in or have been in the stop area as sample vehicles based on the location information of the stop area and the positioning data of the vehicles.

[0122] The data acquisition unit is used to acquire the instantaneous speeds of multiple sample vehicles when they are in the stopping area and the vehicle speed acquisition time corresponding to the instantaneous speeds;

[0123] The vehicle speed time-series data set acquisition unit is used to construct a vehicle speed time-series data pair by acquiring the instantaneous speed and the corresponding vehicle speed acquisition time, thereby obtaining the vehicle speed time-series data set of the traffic light; the vehicle speed change data set includes multiple constructed vehicle speed time-series data pairs;

[0124] The change pattern determination unit is used to determine the change pattern of the traffic lights based on the vehicle speed time series data set.

[0125] Optionally, the unit for determining the pattern of change is specifically used for:

[0126] The instantaneous speed of each vehicle speed time-series data pair in the vehicle speed time-series data set is compared with a preset speed threshold. The vehicle speed time-series data pairs in the vehicle speed time-series data set are divided into a first type of vehicle speed time-series data pair and a second type of vehicle speed time-series data pair. The instantaneous speed of the first type of vehicle speed time-series data pair is less than the preset speed threshold. The instantaneous speed of the second type of vehicle speed time-series data pair is greater than or equal to the preset speed threshold.

[0127] A first indication state timing data pair of the traffic light is generated based on the first type of vehicle speed timing data pair, and a second indication state timing data pair of the traffic light is generated based on the second type of vehicle speed timing data pair; the first indication state timing data pair includes a first value and a vehicle speed acquisition time, and the second indication state timing data pair includes a second value and a vehicle speed acquisition time; the first value and the second value are different, the first value is used to indicate that the traffic light is in a first indication state where vehicles are prohibited from passing, and the second value is used to indicate that the traffic light is in a second indication state where vehicles are allowed to pass;

[0128] The traffic light's indication state timing data set is constructed based on multiple first indication state timing data pairs and multiple second indication state timing data pairs.

[0129] The changing patterns of the traffic lights are determined by analyzing the time-series dataset of the indicated states.

[0130] Optionally, the unit for determining the pattern of change is specifically used for:

[0131] Identify data from multiple cyclic periods within the indicated state timing data set;

[0132] The data from the identified N cycle periods are placed into the same cycle period and superimposed to form the superimposed indicator state cycle change data; where N is a positive integer greater than 1.

[0133] The change pattern of the traffic light is determined based on the jump pattern between the first and second values ​​in the superimposed indicator state cycle change data.

[0134] Optionally, N is an odd number, and the variation pattern determination unit is specifically used for:

[0135] The data from N identified cycles are superimposed within the same cycle. If the values ​​used to indicate the state of the traffic light at the same target time in the N cycles are inconsistent, then the number K of cycles corresponding to the same target time using the first value and the number Q of cycles corresponding to the same target time using the second value are obtained; where N = K + Q.

[0136] Determine the relative magnitudes of K and Q. If K is greater than Q, then set the value used to indicate the state of the traffic light corresponding to the same target time within the same cycle to the first value; if K is less than Q, then set the value used to indicate the state of the traffic light corresponding to the same target time within the same cycle to the second value, thus forming superimposed indicator state change data within the cycle.

[0137] Optionally, the unit for determining the pattern of change is specifically used for:

[0138] In the indicated state time-series data set, a sliding window method is used to determine whether a periodic pattern in the indicated state time-series data set is identified before the sliding window has accumulated to a preset number of steps. If yes, data for multiple cycles in the indicated state time-series data set is obtained based on the identified periodic pattern. If no, the fault tolerance threshold for identifying the periodic pattern in the indicated state time-series data set using the sliding window method is increased, and the periodic pattern in the time-series change data is identified based on the increased fault tolerance threshold to obtain data for multiple cycles in the indicated state time-series data set.

[0139] Optionally, the device for determining the changing pattern of traffic lights also includes:

[0140] The enhancement module is used to, if the traffic light is in a first indication state prohibiting vehicles from passing when the target vehicle enters the stop area, based on the determined change pattern of the traffic light, indicate the remaining duration of the first indication state to the driver of the target vehicle, and prompt the driver to prepare at a prompt time before the end of the first indication state; the prompt time is the xth second before the end of the first indication state, where x is a positive number and x is less than the duration during which the traffic light has been continuously in the first indication state;

[0141] The prompting module is further configured to prompt the driver that the second indication state ends when the traffic light is in a second indication state that allows vehicles to pass, and when the target vehicle is about to enter the stop area. The second indication state ends when there are only y seconds left, where y is a positive number and y is less than the duration for which the traffic light has been in the second indication state continuously.

[0142] Optionally, the stop area determination module includes:

[0143] The first determining unit is used to determine the stopping area based on the stop line corresponding to the traffic light and the lane width of the target intersection if the map data reaches the accuracy of locating the stop line of the target intersection.

[0144] The second determining unit is used to determine the position information of the center point of the target intersection based on the map data if the map data does not reach the accuracy of locating the stop line of the target intersection, and to estimate the position of the stop line corresponding to the traffic light at the target intersection based on the position information of the center point and the width of the two lanes, and to determine the stopping area according to the stop line corresponding to the traffic light and the lane width of the target intersection.

[0145] The first boundary of the stop area is a boundary that extends beyond the stop line by one lane width, and the second boundary of the stop area is a boundary that does not extend beyond the stop line by two lane widths. The first boundary and the second boundary are parallel.

[0146] Based on the method and apparatus for determining the changing patterns of traffic lights provided in the foregoing embodiments, this application also provides a computer-readable storage medium. This storage medium stores a program, which, when executed by a processor, implements some or all of the steps in the method for determining the changing patterns of traffic lights protected in the foregoing method embodiments of this application. The storage medium can be any medium capable of storing program code, such as a USB flash drive, portable hard drive, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0147] Based on the method, apparatus, and storage medium for determining traffic light change patterns provided in the foregoing embodiments, this application provides a processor. This processor is used to run a program, wherein the program, when running, executes some or all of the steps in the method for determining traffic light change patterns protected by the foregoing method embodiments.

[0148] Based on the storage medium and processor provided in the foregoing embodiments, this application also provides a device for determining the changing patterns of traffic lights. The device for determining the changing patterns of traffic lights includes: a memory, a processor, a communication bus, and a communication interface. The memory stores a program that can run on the processor, and when the program is executed, it implements some or all of the steps in the method for determining the changing patterns of traffic lights provided in the foregoing method embodiments of this application. The memory may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. In this device, the processor and memory transmit signaling, logic instructions, etc., through the communication bus. This device can communicate and interact with other devices through the communication interface.

[0149] It should be noted that the various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for the device and apparatus embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the description of the method embodiments. The device and apparatus embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components indicated as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of the solution in this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0150] The above description is merely one specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for determining the changing patterns of traffic lights, characterized in that, include: The type of lane the target vehicle should enter before entering the target intersection is determined based on the target vehicle's current route planning data; Based on the aforementioned type, determine the traffic lights at the target intersection whose changing patterns need to be analyzed; The stop area corresponding to the traffic light is determined based on the map data of the target intersection; The instantaneous speed of each speed time series data pair in the sample vehicle's speed time series data set in the stopping area is numerically compared with a preset speed threshold. The speed time series data pairs in the set are then divided into a first type of speed time series data pair and a second type of speed time series data pair. The instantaneous speed in the first type of speed time series data pair is less than the preset speed threshold, while the instantaneous speed in the second type of speed time series data pair is greater than or equal to the preset speed threshold. A first indication state timing data pair of the traffic light is generated based on the first type of vehicle speed timing data pair, and a second indication state timing data pair of the traffic light is generated based on the second type of vehicle speed timing data pair; the first indication state timing data pair includes a first value and a vehicle speed acquisition time, and the second indication state timing data pair includes a second value and a vehicle speed acquisition time; the first value and the second value are different, the first value is used to indicate that the traffic light is in a first indication state where vehicles are prohibited from passing, and the second value is used to indicate that the traffic light is in a second indication state where vehicles are allowed to pass; The traffic light's indication state timing data set is constructed based on multiple first indication state timing data pairs and multiple second indication state timing data pairs. The changing patterns of the traffic lights are determined by analyzing the time-series dataset of the indicated states; When the amount of data in the indicated state time series data pairs does not meet the accuracy requirements for mining change patterns, the step of determining the change patterns of the traffic lights through analysis of the indicated state time series dataset includes: Identify data from multiple cyclic periods within the indicated state timing data set; The data from N identified cycles are superimposed within the same cycle. If the values ​​used to indicate the state of the traffic light at the same target time in the N cycles are inconsistent, then the number K of cycles corresponding to the same target time using the first value and the number Q of cycles corresponding to the same target time using the second value are obtained; where N = K + Q; N is an odd number greater than 1. Determine the relative magnitudes of K and Q. If K is greater than Q, then set the value used to indicate the state of the traffic light corresponding to the same target time within the same cycle to the first value; if K is less than Q, then set the value used to indicate the state of the traffic light corresponding to the same target time within the same cycle to the second value, thus forming superimposed data on the changes in the indicated state within the cycle. The change pattern of the traffic light is determined based on the jump pattern between the first and second values ​​in the superimposed indicator state cycle data.

2. The method for determining the changing pattern of traffic lights according to claim 1, characterized in that, The identification of data from multiple cyclic periods in the indicated state time-series data set includes: In the indicated state time series data set, the sliding window method is used to determine whether a periodic pattern in the indicated state time series data set is identified before the cumulative sliding of the sliding window reaches a preset number of steps; If so, then data from multiple cyclic periods in the indicated state time series data set are obtained based on the identified periodic patterns; If not, the fault tolerance threshold for identifying periodic patterns in the indicated state time series data set using the sliding window method is increased, and the periodic patterns in the time series change data are identified based on the increased fault tolerance threshold to obtain data from multiple cyclic periods in the indicated state time series data set.

3. The method for determining the changing pattern of traffic lights according to claim 1 or 2, characterized in that, After determining the changing pattern of the traffic lights, the method further includes: If the traffic light is in a first indication state prohibiting vehicles from passing when the target vehicle enters the stopping area, then based on the determined change pattern of the traffic light, the driver of the target vehicle is prompted with the remaining duration of the first indication state, and the driver is prompted to prepare at a prompt time before the end of the first indication state; the prompt time is the xth second before the end of the first indication state, where x is a positive number and x is less than the duration during which the traffic light has been continuously in the first indication state; If the traffic light is in a second indication state that allows vehicles to pass when the target vehicle is about to enter the stop area, then when there are only y seconds left in the second indication state, the driver will be notified that there are only y seconds left in the second indication state, where y is a positive number and y is less than the duration for which the traffic light has been in the second indication state continuously.

4. The method for determining the changing pattern of traffic lights according to claim 1 or 2, characterized in that, Determining the stop area corresponding to the traffic light based on the map data of the target intersection includes: If the map data achieves the accuracy of locating the stop line of the target intersection, then the stop area is determined based on the stop line corresponding to the traffic light and the lane width of the target intersection. If the map data does not achieve the accuracy required to locate the stop line of the target intersection, then the location information of the center point of the target intersection is determined based on the map data, and the location of the stop line corresponding to the traffic light at the target intersection is estimated based on the location information of the center point and the width of the two lanes. The stop area is then determined based on the stop line corresponding to the traffic light and the lane width of the target intersection. The first boundary of the stop area is a boundary that extends beyond the stop line by one lane width, and the second boundary of the stop area is a boundary that does not extend beyond the stop line by two lane widths. The first boundary and the second boundary are parallel.

5. A device for determining the changing pattern of traffic lights, characterized in that, include: The lane type determination module is used to determine the type of lane that the target vehicle should enter before entering the target intersection based on the target vehicle's current path planning data; A traffic light determination module is used to determine the traffic lights at the target intersection whose changing patterns need to be analyzed based on the type described above. The stop area determination module is used to determine the stop area corresponding to the traffic light based on the map data of the target intersection. The traffic light change pattern determination module is used for: The instantaneous speed of each speed time series data pair in the sample vehicle's speed time series data set in the stopping area is numerically compared with a preset speed threshold. The speed time series data pairs in the set are then divided into a first type of speed time series data pair and a second type of speed time series data pair. The instantaneous speed in the first type of speed time series data pair is less than the preset speed threshold, while the instantaneous speed in the second type of speed time series data pair is greater than or equal to the preset speed threshold. A first indication state timing data pair of the traffic light is generated based on the first type of vehicle speed timing data pair, and a second indication state timing data pair of the traffic light is generated based on the second type of vehicle speed timing data pair; the first indication state timing data pair includes a first value and a vehicle speed acquisition time, and the second indication state timing data pair includes a second value and a vehicle speed acquisition time; the first value and the second value are different, the first value is used to indicate that the traffic light is in a first indication state where vehicles are prohibited from passing, and the second value is used to indicate that the traffic light is in a second indication state where vehicles are allowed to pass; The traffic light's indication state timing data set is constructed based on multiple first indication state timing data pairs and multiple second indication state timing data pairs. The changing patterns of the traffic lights are determined by analyzing the time-series dataset of the indicated states; When the amount of data in the indicated state time series data pair does not meet the accuracy requirements for mining the change pattern, the signal light change pattern determination module is specifically used for: Identify data from multiple cyclic periods within the indicated state timing data set; The data from N identified cycles are superimposed within the same cycle. If the values ​​used to indicate the state of the traffic light at the same target time in the N cycles are inconsistent, then the number K of cycles corresponding to the same target time using the first value and the number Q of cycles corresponding to the same target time using the second value are obtained; where N = K + Q; N is an odd number greater than 1. Determine the relative magnitudes of K and Q. If K is greater than Q, then set the value used to indicate the state of the traffic light corresponding to the same target time within the same cycle to the first value; if K is less than Q, then set the value used to indicate the state of the traffic light corresponding to the same target time within the same cycle to the second value, thus forming superimposed data on the changes in the indicated state within the cycle. The change pattern of the traffic light is determined based on the jump pattern between the first and second values ​​in the superimposed indicator state cycle data.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method for determining the changing patterns of traffic lights as described in any one of claims 1-4.

7. A processor, characterized in that, Used to run a computer program, which, when running, executes the method for determining the changing patterns of traffic lights as described in any one of claims 1-4.

Images