Method and device for predicting position of trajectory point, equipment and storage medium

Abstract

The invention discloses a method and device for predicting the position of a trajectory point, equipment and a storage medium. Comprising the steps of performing road network matching based on a preset number of continuous track points before a current track point to obtain a road section position of the current track point; calculating a vehicle driving distance based on the vehicle speed information and the predicted time interval; judging whether the predicted position is in the current road section range or not based on the road section position, the road section direction and the vehicle driving distance of the current track point; if the predicted position is within the current road section range, the predicted position is used as a predicted next track point, and if the predicted position is not within the current road section range, the vehicle driving-out road section is predicted based on the historical driving characteristics of the vehicle, and the predicted position of the vehicle at the driving-out road section is determined. According to the scheme of the invention, high-precision position prediction at a shorter time interval can be realized on the premise that the vehicle-mounted terminal reports the interval, and the real-time performance of vehicle track monitoring and the user experience are remarkably improved.

Description

Methods, apparatus, devices, and storage media for predicting trajectory point locations Technical Field

[0001] This application relates to the field of trajectory prediction technology, and more specifically, to a method, apparatus, device, and storage medium for predicting the location of trajectory points. Background Technology

[0002] In real-time safety monitoring and operation management of road transport vehicles, on-board terminals continuously report vehicle trajectory locations via satellite positioning technology. Currently, due to limitations in communication costs and power consumption, mainstream truck monitoring terminals typically set their trajectory location reporting interval to 30 seconds. However, in real-time monitoring scenarios such as urban delivery, hazardous materials transportation, and precision dispatching, a 30-second location update interval is too long for users. This results in an inability to promptly grasp the precise dynamics of vehicles, and excessively long periods of "disconnection" between two reports, severely impacting the real-time performance of monitoring and user experience.

[0003] The common prediction method currently used is similar to inertial navigation, which uses the current speed and travels forward. However, when passing through complex intersections, it cannot make correct judgments on the vehicle's turning behavior at the intersection, resulting in a serious deviation between the predicted trajectory and the vehicle's actual driving path. Summary of the Invention

[0004] This application provides a method, apparatus, device, and storage medium for predicting the location of trajectory points, so as to at least solve the technical problem in the related art that it is difficult to predict vehicle trajectory points in real time and accurately.

[0005] According to one aspect of the embodiments of this application, a method for predicting the location of a trajectory point is provided, comprising: performing road network matching based on a preset number of consecutive trajectory points preceding the current trajectory point to obtain the location of the road segment where the current trajectory point is located; calculating the vehicle's travel distance based on vehicle speed information and a prediction time interval; determining whether the predicted location is within the current road segment based on the location of the current trajectory point on the road segment, the road segment direction, and the vehicle's travel distance; if the predicted location is within the current road segment, using the predicted location as the next predicted trajectory point; if the predicted location is not within the current road segment, predicting the vehicle's exit from the road segment based on the vehicle's historical driving characteristics, and determining the predicted location of the vehicle in the exited road segment.

[0006] In one implementation, if the predicted location is not within the current road segment, the vehicle exits the road segment based on its historical driving characteristics, and the predicted location of the vehicle in the exit road segment is determined. This includes: if the predicted location is not within the current road segment, taking the current road segment as the entry road segment, querying the statistical results of each exit road segment at the intersection in the current time period from a preset historical traffic characteristic database according to the intersection marker at the road segment boundary to see if there is a statistical result for the vehicle; if there is an exit road segment for the vehicle in the statistical results, then the exit road segment is taken as the vehicle exit road segment, and the predicted location of the vehicle in the exit road segment is determined based on the vehicle's travel distance.

[0007] In one implementation, the method further includes: if the statistical results show that there are multiple exit segments for the vehicle, calculating the number of times the vehicle passes through each exit segment; taking the exit segment with the most passes as the vehicle's exit segment, and determining the predicted position of the vehicle in the exit segment based on the vehicle's travel distance.

[0008] In one implementation, the method further includes: if the statistical results show multiple exit segments for the vehicle, and multiple exit segments with the highest number of passages for the vehicle also exist, using the current segment as the entry segment and the multiple exit segments with the highest number of passages for the current vehicle as exit segments as query conditions, querying the statistical values ​​of each vehicle type at the intersection during the current preset time period in the preset historical traffic feature database; taking the exit segment with the highest number of passages for the same type of vehicle as the current vehicle as the vehicle exit segment, and determining the predicted position of the vehicle in the exit segment based on the vehicle's travel distance.

[0009] In one implementation, the method further includes: when there are multiple exit segments with the most frequent trips for the current vehicle type, randomly selecting one exit segment as the vehicle exit segment, and determining the predicted position of the vehicle in the exit segment based on the vehicle's travel distance.

[0010] In one implementation, the method further includes: if there is no statistical result for the vehicle identifier in the statistical results of each exit segment of the intersection during the current time period, then the current segment is taken as the entry segment, and the statistical values ​​of exit segments for each vehicle type during the current preset time period of the intersection are queried; the exit segment with the most trips of the same vehicle type as the current vehicle is taken as the vehicle exit segment; if there are multiple exit segments with the most trips of the same vehicle type, then an exit segment is randomly selected as the vehicle exit segment.

[0011] In one implementation, the method further includes: numbering each intersection on the road network and recording the entry and exit segments of each intersection; performing road network matching based on historical vehicle trajectory data, counting the vehicles entering and exiting each intersection, and obtaining the number of passages of each type of vehicle in the entry and exit segments of the intersection within a preset time period; and storing the statistical data within the preset time period into a historical traffic feature database.

[0012] According to another aspect of the embodiments of this application, an apparatus for predicting the location of a trajectory point is provided, comprising: a road network matching module, configured to perform road network matching based on a preset number of consecutive trajectory points preceding the current trajectory point to obtain the location of the road segment where the current trajectory point is located; a travel distance calculation module, configured to calculate the vehicle travel distance based on vehicle speed information and a prediction time interval; a judgment module, configured to determine whether the predicted location is within the current road segment based on the location of the current trajectory point on the road segment, the direction of the road segment, and the vehicle travel distance; and a trajectory prediction module, configured to, if the predicted location is within the current road segment, use the predicted location as the next predicted trajectory point, and if the predicted location is not within the current road segment, predict the vehicle's exit from the road segment based on the vehicle's historical driving characteristics, and determine the predicted location of the vehicle in the exit road segment.

[0013] According to another aspect of the embodiments of this application, an electronic device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the method for predicting the location of trajectory points through the computer program.

[0014] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided, wherein a computer program is stored in the computer program, and the computer program is configured to execute the above-described method for predicting the location of trajectory points when it is run.

[0015] The technical solution provided in this application embodiment can include the following beneficial effects: The trajectory prediction method of this application determines whether the predicted position is within the current road segment based on the current trajectory point's location, road segment direction, and vehicle travel distance. If the predicted position is within the current road segment, the predicted position is used as the next predicted trajectory point. If the predicted position is not within the current road segment, the vehicle's exit from the road segment is predicted based on the vehicle's historical driving characteristics, and the predicted position of the vehicle in the exit segment is determined. By introducing a trajectory prediction mechanism based on road network topology and historical behavior intelligent decision-making, the shortcomings of traditional linear prediction models that are prone to failure at complex intersections are fundamentally overcome. Ultimately, without increasing hardware costs and communication frequency, the effective frequency of position updates is increased from 30 seconds to 5 seconds, greatly enhancing the real-time performance, trajectory smoothness, and prediction accuracy of the monitoring system, providing users with a near-continuous high-precision dynamic view of vehicles, and has significant practical value and commercial prospects. Attached Figure Description

[0016] The accompanying drawings, which are included to provide a further understanding of this application and constitute a part of this application, illustrate exemplary embodiments of this application and, together with the description, serve to explain this application and do not constitute an undue limitation of this application. In the drawings: FIG1 is a schematic flowchart of a method for predicting the position of trajectory points according to an embodiment of this application; FIG2 is a schematic flowchart of a data preprocessing process according to an embodiment of this application; FIG3 is a schematic flowchart of a trajectory prediction method according to an embodiment of this application; FIG4 is a schematic diagram of a front-end trajectory line according to an embodiment of this application; FIG5 is a schematic diagram of another front-end trajectory line according to an embodiment of this application; FIG6 is a schematic diagram of an apparatus for predicting the position of trajectory points according to an embodiment of this application; FIG7 is a schematic diagram of the structure of an optional electronic device according to an embodiment of this application. Detailed Implementation

[0017] 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 should fall within the scope of protection of the present application.

[0018] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0019] In one implementation scenario, such as logistics fleet management and hazardous materials transportation monitoring, the monitoring terminals of the trucks or commercial vehicles that these users rely on are limited by communication protocols and power consumption, typically reporting their actual trajectory location only every 30 seconds. This interval is too long in complex urban road networks, resulting in noticeable jumps in vehicle movement on the monitoring screen, failing to meet users' management needs for continuous, smooth, and accurate vehicle tracking.

[0020] This solution operates within this context: the system is activated during the 30-second "window" between two actual location reports from the terminal. Whenever a new real location report arrives, the system doesn't passively wait for the next real location; instead, it activates a high-frequency prediction engine, dynamically calculating and publishing the vehicle's future location every 5 seconds. This process continues until the next real location report arrives, forming a virtuous cycle of "real location calibration, high-frequency prediction filling." Ultimately, without changing hardware or increasing communication costs, it presents the user with a nearly continuous and smooth vehicle trajectory, greatly optimizing the visual effects of the monitoring screen and the real-time performance of dispatching decisions.

[0021] The method for predicting the location of trajectory points according to embodiments of this application will be described in detail below with reference to the accompanying drawings. As shown in Figure 1, the method mainly includes the following steps: S101 Based on a preset number of consecutive trajectory points before the current trajectory point, road network matching is performed to obtain the location of the road segment where the current trajectory point is located.

[0022] In one implementation, for a vehicle-mounted terminal that reports its location every 30 seconds, the vehicle's position is predicted every 5 seconds after each report. The specific prediction time interval is not limited in this embodiment and can be set according to actual conditions.

[0023] Specifically, during prediction, road network matching is performed based on a preset number of consecutive trajectory points preceding the current trajectory point to obtain the location of the road segment where the current trajectory point is located.

[0024] For example, after receiving the latest trajectory point or the latest predicted point reported by the vehicle terminal, the system does not process the single point in isolation. Instead, it selects five consecutive historical trajectory points, including the current point, to form a trajectory sequence. Using a map matching algorithm, this trajectory sequence is matched to possible candidate road segments in the road network. This accurately determines the correct road segment to which the current trajectory point most likely belongs and its specific location on that road segment.

[0025] S102 calculates the vehicle's travel distance based on vehicle speed information and the predicted time interval.

[0026] In one implementation, after obtaining the last trajectory point accurately located using multi-point road network matching technology, the system uses it as the starting reference point for prediction and records the corresponding timestamp (T0) and instantaneous speed (V0). When prediction is needed (e.g., 5 seconds later, at time T1), the travel distance can be calculated using V0 (S1 = V0 * (T1 - T0)). This yields the vehicle's travel distance within the prediction time period.

[0027] S103 determines whether the predicted location is within the current road segment based on the current trajectory point's location, road segment direction, and vehicle travel distance.

[0028] In one implementation, after obtaining the current trajectory point road segment position after precise matching in step S101, the system first calculates the remaining drivable length from the point to the downstream intersection along the road segment direction.

[0029] Further, the vehicle's travel distance is obtained based on step S102. The vehicle's travel distance is compared with the remaining drivable length. If the vehicle's travel distance is less than or equal to the remaining drivable length, it is determined that the predicted location is still within the geometric range of the current road segment; if the vehicle's travel distance is greater than the remaining drivable length, it is determined that the vehicle will leave the current road segment within the predicted time period.

[0030] S104 If the predicted location is within the current road segment, the predicted location will be used as the next predicted trajectory point. If the predicted location is not within the current road segment, the vehicle will exit the road segment based on its historical driving characteristics, and the predicted location of the vehicle in the exiting road segment will be determined.

[0031] In one implementation, if the predicted location is within the current road segment, the predicted location is used as the next predicted trajectory point. If the predicted location is still within the geometric range of the current road segment, the vehicle's travel distance can be directly extended along the road segment direction to output the predicted coordinates.

[0032] In another implementation, if the predicted location is not within the current road segment, a road segment switching strategy is immediately triggered. A preset historical traffic feature database is invoked, and based on historical statistics of the vehicle's or similar vehicles' probability distribution for left turns, right turns, and straight-ahead maneuvers at the intersection, the exit road segment with the highest probability is selected. The distance exceeding this limit is then converted to the starting point of the new vehicle segment for continued positioning.

[0033] Specifically, if the predicted location is not within the current road segment, the current road segment is taken as the entry segment. Based on the intersection markers at the road segment boundary, the system queries the pre-set historical traffic characteristic database to check if the vehicle's statistics exist in the statistical results of each exit segment for the current time period. If the statistical results show an exit segment for the vehicle, that exit segment is taken as the vehicle's exit segment, and the predicted location of the vehicle in the exit segment is determined based on the vehicle's travel distance.

[0034] Understandably, the system accesses a pre-set historical traffic feature database (such as Redis), generates a query key with a format such as "intersection ID:entry segment:exit segment:current hour", and searches for statistical records with the target vehicle ID as the field under this key.

[0035] If the query results show that there is one and only one exit segment associated with the vehicle ID in the historical records of the current time period at the intersection, the system interprets this result as a strong steering habit of the vehicle at this location. Based on this, the system identifies the exit segment as the predicted path and enters the location calculation stage. The remaining travel distance beyond the current segment's starting point is mapped from the starting point of the exit segment along its direction, thereby calculating the vehicle's precise predicted position on that exit segment. This mechanism effectively transforms the historical behavior patterns of individual vehicles into a reliable basis for predicting future paths, significantly improving the accuracy of trajectory prediction in intersection areas.

[0036] In one implementation, if the statistical results show that there are multiple exit segments for the vehicle, the number of times the vehicle passes through each exit segment is calculated; the exit segment with the most passes is taken as the vehicle's exit segment, and the predicted position of the vehicle in the exit segment is determined based on the vehicle's travel distance.

[0037] Understandably, if the query results show that the vehicle has exited multiple road segments in the current time period at this intersection, the system will initiate the frequency priority decision logic.

[0038] Specifically, the system calculates and compares the vehicle's historical passage counts on various exit segments. Then, applying the "maximum frequency principle," it selects the exit segment with the highest passage count as the predicted exit segment for the current vehicle journey. In the absence of unique certainty, the path most frequently chosen by the vehicle is its most likely path for the next trip. This implementation effectively solves the decision-making challenge of multiple path selection by mining the high-frequency behavioral patterns of individual vehicles, enhancing the adaptability and reliability of the prediction system.

[0039] In one implementation, if the statistical results show multiple exit segments for the vehicle, and multiple exit segments with the highest number of passages for the vehicle also exist, the method is to use the current segment as the entry segment and the multiple exit segments with the highest number of passages for the current vehicle as exit segments as query conditions, and query the statistical values ​​of each vehicle type at the intersection during the current preset time period in the preset historical traffic feature database; the exit segment with the highest number of passages for the same type of vehicle as the current vehicle is taken as the vehicle exit segment, and the predicted position of the vehicle in the exit segment is determined based on the vehicle's travel distance.

[0040] Understandably, the system also includes a supplementary decision-making mechanism for situations involving "the highest frequency of ties". When the system detects that a vehicle has exited multiple road segments, and two or more of these road segments have the same number of historical passages, the "individual frequency priority" principle fails, and the system will automatically activate group behavior analysis logic as a supplementary decision-making basis.

[0041] At this point, the system uses the current road segment as the entry segment and the exit segments with the highest number of consecutive passages as filtering criteria to initiate a new round of queries in the preset historical traffic feature database. The goal of this query is to obtain the traffic statistics of vehicle groups belonging to the same vehicle type (e.g., all "heavy trucks") as the currently predicted vehicle at this intersection within the current preset time period. The system will select the exit segment with the highest historical passage frequency under this vehicle type dimension and ultimately determine it as the driving segment for this prediction. The core of this mechanism is that when individual historical behavior cannot provide a unique decision, it refers to the general selection patterns of similar vehicle groups with similar operating characteristics, thereby making a more reasonable prediction that conforms to the macro traffic flow characteristics and ensuring the robustness of the decision-making logic.

[0042] In one implementation, when there are multiple exit segments with the most frequent trips for the current vehicle type, a random exit segment is selected as the vehicle's exit segment, and the predicted position of the vehicle in the exit segment is determined based on the vehicle's travel distance.

[0043] Understandably, even after applying the "group behavior analysis" logic, if a group of vehicles of the same type as the current vehicle still has the highest number of historical passages across multiple candidate exit segments, the system will initiate a random selection mechanism. This mechanism will randomly select one exit segment from among those with the highest number of passages as the final exit segment for the vehicle. Subsequently, the system will determine the vehicle's predicted position on this randomly selected segment based on the previously calculated remaining travel distance. This mechanism, serving as the endpoint of the entire multi-level decision-making process, ensures the completeness of the prediction algorithm and avoids process interruptions due to inability to make a decision in extreme cases, thereby maintaining the continuity and reliability of the system service.

[0044] In one implementation, if the statistical results of each exit segment at the intersection during the current time period do not include the statistical results of the vehicle identifier, then the current segment is taken as the entry segment, and the statistical values ​​of exit segments for each vehicle type at the intersection during the current preset time period are queried; the exit segment with the most trips of the same vehicle type as the current vehicle is taken as the vehicle exit segment; if there are multiple exit segments with the most trips of the same vehicle type, then an exit segment is randomly selected as the vehicle exit segment.

[0045] Understandably, the system also includes comprehensive logic for handling scenarios where a vehicle is "visiting for the first time" or has "no historical records." When a query reveals that there are no historical passage records for the target vehicle in the current time period's statistics for that intersection, the system will be unable to employ a prediction strategy based on individual behavior. In this case, the system immediately switches to a prediction mode dominated by group behavior.

[0046] Specifically, the system uses the current road segment as the entry point and queries the historical traffic characteristic database to retrieve the exit route statistics for all vehicle types within the current preset time period at that intersection. The system then filters out statistical groups that match the currently predicted vehicle type and applies the "maximum frequency principle," selecting the exit route with the highest group passage frequency as the predicted route. The rationale behind this strategy is that, in the absence of individual data, following the common selection patterns of vehicles of the same type at that location is the optimal estimate. Furthermore, if multiple exit routes with the highest number of passages exist within this vehicle type dimension, the system will again employ the aforementioned random selection mechanism, randomly selecting one from these tied routes as the final result, thus ensuring the entire prediction process can be completed smoothly and a definite route selection is output.

[0047] In one implementation, the historical data is preprocessed to construct a historical access feature database.

[0048] Each intersection is numbered on the road network, and the entry and exit segments of each intersection are recorded. Based on historical vehicle trajectory data, road network matching is performed to count the number of vehicles entering and exiting each intersection, and the number of times each type of vehicle passes through the intersection's entry and exit segments within a preset time period. The statistical data within the preset time period is stored in the historical traffic feature database.

[0049] To facilitate understanding of the method for constructing the historical access feature database in this application embodiment, it will be further described below with reference to Figure 2.

[0050] As shown in Figure 2, each intersection is first numbered on the road network, and the entry and exit segments of that intersection are recorded in Redis. The road network is then matched with the daily vehicle trajectories to count the vehicles entering and exiting each intersection.

[0051] Furthermore, based on vehicle type and vehicle ID, statistics such as vehicle entry segments, exit segments, and times at each intersection are stored in Redis. For vehicle ID, using "Intersection ID:Entry Segment:Exit Segment:Hour:Date" as the key, RedisHash is used to store the number of times each vehicle passes through the intersection within a certain hour of a given day, with the vehicle ID as the hash key.

[0052] Furthermore, for vehicle type, using "Intersection ID:Entry Segment:Exit Segment:Hour:Date:Vehicle Type" as the key, RedisHash is used to store the number of times each type of vehicle passes through the intersection's passage segment within a specific hour of a given day, with the vehicle type as the hash key. Redis stores 15 days of statistical data. The specific data update time is not specifically limited in this application.

[0053] By constructing a historical traffic feature database, the historical behavior of individual vehicles and the group patterns of similar vehicles are transformed into predictive knowledge. This enables the system to intelligently predict turns at complex intersections based on "memory" and "experience," rather than blindly extrapolating linearly. As a result, the accuracy and rationality of low-frequency positioning data in high-frequency trajectory prediction are significantly improved without increasing hardware costs.

[0054] To facilitate understanding of the trajectory prediction method in the embodiments of this application, it will be further described below with reference to Figure 3.

[0055] As shown in Figure 3, for a vehicle-mounted terminal that reports its location every 30 seconds, the vehicle's position is predicted every 5 seconds after each report. During prediction, the previous five consecutive trajectory points are selected, and the location of the last trajectory point on the road segment is accurately determined using multi-point road network matching technology.

[0056] Furthermore, based on the vehicle speed, the distance the vehicle travels within the predicted time is determined. Based on the location of the vehicle's last trajectory point on the road segment, the direction of the road segment, and the position of that travel distance within the road segment, it is determined whether the location is within the road segment.

[0057] If the location is within a road segment, select the predicted road segment as the current road segment.

[0058] If the location is outside the road segment, the road segment is used as the entry road segment. Based on the intersection ID of the road segment boundary, the statistical results of the vehicle ID for the current hour at that intersection are retrieved from Redis.

[0059] Determine if there are any query results. If there are, and the query results are unique, then select the predicted road segment as the current road segment, that is, take the query exit road segment as the vehicle's next exit road segment. If the query results are not unique, determine if the road segment with the highest number of statistics is unique. If there is only one road segment with the highest number of statistics, then take the road segment with the highest number of statistics as the exit road segment.

[0060] Furthermore, if the road segment with the highest number of statistics is not unique, the current road segment is taken as the entry road segment, and the vehicle ID with the highest statistical value is taken as the exit road segment. The statistical values ​​of each vehicle type at the intersection for the current hour are queried, and the road segment with the highest number of trips of the same type as the current vehicle is selected as the predicted driving road segment. If there is more than one road segment with the highest number of trips, one is randomly selected.

[0061] Furthermore, if no query result is found for the current vehicle ID, the current road segment is taken as the entry segment. The statistical values ​​for each type of exit segment for the current hour are queried, and the road segment with the highest number of trips of the same type as the current vehicle is selected as the predicted driving segment. If there are multiple road segments with the highest number of trips, one is randomly selected. If the cumulative travel distance on the predicted road segment equals the product of time and speed, then the point on the predicted road segment is used as the predicted location, and the loop terminates.

[0062] At the user interaction level, this application integrates the predicted trajectory with the actual reported trajectory through front-end visualization technology. In specific implementation, the system receives and processes two data streams from the backend: one is the actual trajectory points reported periodically by the vehicle terminal, and the other is the predicted trajectory points generated by the algorithm of this application. The front-end interface employs a differentiated rendering strategy. For example, solid lines or highlighted colors are used to connect actual trajectory points to represent confirmed paths, while dashed lines, semi-transparent lines, or a different color scheme are used to connect predicted trajectory points, forming a predicted trajectory line that smoothly connects with the actual trajectory.

[0063] The advantage of this display method is that it provides users with a continuous and uninterrupted view of the vehicle's trajectory. Even during the reporting intervals at the terminal, users can clearly observe the vehicle's estimated path and direction, effectively eliminating the "jumping" feeling caused by sparse data points. Simultaneously, the differentiated display ensures that users can clearly distinguish between real data and predicted results, preventing misunderstandings. When new real data points arrive, the predicted trajectory is calibrated and updated accordingly, dynamically optimizing the display effect and greatly improving the intuitiveness, smoothness, and user experience of the monitoring screen.

[0064] As shown in Figures 4 and 5, the thickened trajectory lines are the predicted reporting points, while the thinner lines are the actual trajectory reporting points, which effectively supplements the trajectory.

[0065] In the scenario of a vehicle passing through an intersection, when the predicted point exceeds the current road segment (i.e., it is determined that the vehicle is about to or has already passed through the intersection), the system does not blindly extend in a straight line as in traditional methods, but intelligently triggers a multi-level decision-making process.

[0066] First, the vehicle's historical turning preferences at this intersection are queried (indexed by vehicle ID). If there is no ambiguity, the preference is adopted directly. If multiple high-frequency choices exist, the group behavioral characteristics of the vehicle type are further incorporated into the decision. This decision-making strategy of "prioritizing individual habits and supplementing with group patterns" allows the predictive model to deeply understand and learn the personalized driving habits and general patterns of drivers or vehicles at specific intersections. This enables it to make the intelligent choice that best reflects the vehicle's actual driving intentions when faced with multiple alternative routes. Ultimately, without increasing hardware costs or communication frequency, the effective frequency of location updates is increased from 30 seconds to 5 seconds, significantly enhancing the real-time performance, trajectory smoothness, and prediction accuracy of the monitoring system.

[0067] According to another aspect of the embodiments of this application, an apparatus for predicting the location of a trajectory point is also provided for implementing the above-described method for predicting the location of a trajectory point. As shown in FIG6, the apparatus includes: a road network matching module 601, used for performing road network matching based on a preset number of consecutive trajectory points before the current trajectory point to obtain the location of the road segment where the current trajectory point is located; a travel distance calculation module 602, used for calculating the vehicle travel distance based on vehicle speed information and a prediction time interval; a judgment module 603, used for judging whether the predicted location is within the current road segment based on the location of the current trajectory point on the road segment, the direction of the road segment, and the vehicle travel distance; and a trajectory prediction module 604, used for predicting the predicted location as the next predicted trajectory point if the predicted location is within the current road segment, and predicting the vehicle's exit from the road segment based on the vehicle's historical travel characteristics if the predicted location is not within the current road segment, and determining the predicted location of the vehicle in the exit road segment.

[0068] It should be noted that the above embodiments of the device for predicting trajectory point positions are only illustrated by the division of the above functional modules when executing the method for predicting trajectory point positions. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the device for predicting trajectory point positions and the method embodiments for predicting trajectory point positions provided in the above embodiments belong to the same concept, and the implementation process is detailed in the method embodiments, which will not be repeated here.

[0069] According to another aspect of the present application, an electronic device corresponding to the method for predicting trajectory point positions provided in the foregoing embodiments is also provided, for performing the above-described method for predicting trajectory point positions.

[0070] Please refer to Figure 7, which shows a schematic diagram of an electronic device provided by some embodiments of this application. As shown in Figure 7, the electronic device includes: a processor 700, a memory 701, a bus 702, and a communication interface 703. The processor 700, the communication interface 703, and the memory 701 are connected via the bus 702. The memory 701 stores a computer program that can run on the processor 700. When the processor 700 runs the computer program, it executes the method for predicting the position of trajectory points provided by any of the foregoing embodiments of this application.

[0071] The memory 701 may include high-speed random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Communication between this system network element and at least one other network element is achieved through at least one communication interface 703 (which can be wired or wireless), such as the Internet, wide area network, local area network, or metropolitan area network.

[0072] Bus 702 can be an ISA bus, PCI bus, or EISA bus, etc. Buses can be divided into address buses, data buses, control buses, etc. Memory 701 is used to store programs. After receiving execution instructions, processor 700 executes the program. The method for predicting trajectory point positions disclosed in any of the aforementioned embodiments of this application can be applied to processor 700, or implemented by processor 700.

[0073] The processor 700 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of the processor 700 or by instructions in software form. The processor 700 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules may reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in memory 701. Processor 700 reads the information in memory 701 and, in conjunction with its hardware, completes the steps of the above method.

[0074] The electronic device provided in this application embodiment and the method for predicting trajectory point positions provided in this application embodiment are based on the same inventive concept and have the same beneficial effects as the methods they employ, operate, or implement.

[0075] According to another aspect of the present application, a computer-readable storage medium corresponding to the method for predicting the location of trajectory points provided in the foregoing embodiments is also provided, wherein a computer program (i.e., a program product) is stored thereon, and when the computer program is run by a processor, it executes the method for predicting the location of trajectory points provided in any of the foregoing embodiments.

[0076] It should be noted that examples of computer-readable storage media may also include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other optical and magnetic storage media, which will not be elaborated here.

[0077] The computer-readable storage medium provided in the above embodiments of this application and the method for predicting trajectory point positions provided in the embodiments of this application are based on the same inventive concept and have the same beneficial effects as the methods used, run or implemented by the applications stored therein.

[0078] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0079] The above embodiments merely illustrate several implementation methods of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this patent should be determined by the appended claims.

Claims

1. A method for predicting the position of trajectory points, characterized in that, include: Based on a preset number of consecutive trajectory points preceding the current trajectory point, road network matching is performed to obtain the location of the road segment where the current trajectory point is located. Calculate the vehicle's travel distance based on vehicle speed information and the predicted time interval; Based on the location of the current trajectory point on the road segment, the direction of the road segment, and the distance traveled by the vehicle, it is determined whether the predicted location is within the current road segment range. If the predicted location is within the current road segment range, the predicted location is used as the next predicted trajectory point. If the predicted location is not within the current road segment range, the vehicle is predicted to exit the road segment based on the vehicle's historical driving characteristics, and the predicted location of the vehicle in the exiting road segment is determined.

2. The method according to claim 1, characterized in that, If the predicted location is not within the current road segment, the vehicle exits the road segment based on its historical driving characteristics, and the predicted location of the vehicle in the exit road segment is determined. This includes: if the predicted location is not within the current road segment, taking the current road segment as the entry road segment, querying the statistical results of each exit road segment at the intersection in the current time period from the preset historical traffic characteristic database according to the intersection marker at the road segment boundary to see if there is a statistical result for the vehicle; if there is an exit road segment for the vehicle in the statistical results, then the exit road segment is taken as the vehicle's exit road segment, and the predicted location of the vehicle in the exit road segment is determined based on the vehicle's travel distance.

3. The method according to claim 2, characterized in that, Also includes: If the statistical results show that the vehicle has multiple exit segments, calculate the number of times the vehicle passed through each exit segment. The exit section with the most frequent traffic is designated as the vehicle exit section, and the predicted position of the vehicle in the exit section is determined based on the vehicle's travel distance.

4. The method according to claim 3, characterized in that, Also includes: If the statistical results show that there are multiple exit segments for the vehicle, and there are also multiple exit segments with the most passages of the vehicle, the current segment is used as the entry segment, and the multiple segments with the most passages of the current vehicle are used as exit segments as the query conditions. The statistical values ​​of each vehicle type at the current preset time period are queried in the preset historical traffic feature database. The road segment with the most frequent exits of the same type of vehicle as the current vehicle is taken as the vehicle exit segment, and the predicted position of the vehicle in the exit segment is determined based on the vehicle's travel distance.

5. The method according to claim 4, characterized in that, Also includes: If there are multiple exit segments with the most frequent trips for the current vehicle type, randomly select one exit segment as the vehicle's exit segment, and determine the vehicle's predicted position on the exit segment based on the vehicle's travel distance.

6. The method according to claim 2, characterized in that, Also includes: If the statistical results for each exit segment at the intersection during the current time period do not include the statistical results for this vehicle identifier, then the current segment will be used as the entry segment, and the statistical values ​​for each vehicle type's exit segments at the intersection during the current preset time period will be queried. The road segment from which the vehicle of the same type as the current vehicle leaves the road the most times will be designated as the vehicle's exit segment. If there are multiple exit segments with the most frequent trips for a particular vehicle type, then one exit segment will be randomly selected as the vehicle's exit segment.

7. The method according to any one of claims 1 to 6, characterized in that, Also includes: Each intersection on the road network is numbered, and the entry and exit sections of each intersection are recorded. Road network matching is performed based on historical vehicle trajectory data. The number of vehicles entering and exiting each intersection is counted to obtain the number of times each vehicle enters and exits the intersection within a preset time period, as well as the number of times each type of vehicle passes through. The statistical data within the preset time period is stored in the historical traffic feature database.

8. A device for predicting the position of trajectory points, characterized in that, include: The road network matching module is used to perform road network matching based on a preset number of consecutive trajectory points before the current trajectory point to obtain the location of the road segment where the current trajectory point is located; The driving distance calculation module is used to calculate the vehicle's driving distance based on vehicle speed information and prediction time interval; the judgment module is used to determine whether the predicted position is within the current road segment based on the current trajectory point's location on the road segment, road segment direction, and vehicle driving distance; the trajectory prediction module is used to use the predicted position as the next predicted trajectory point if the predicted position is within the current road segment, and to predict the vehicle's exit from the road segment based on the vehicle's historical driving characteristics if the predicted position is not within the current road segment, thus determining the vehicle's predicted position in the exiting road segment.

9. An electronic device, characterized in that, It includes a processor and a memory storing program instructions, the processor being configured to, when executing the program instructions, perform the method for predicting the location of trajectory points as described in any one of claims 1 to 7.

10. A computer-readable medium, characterized in that, It stores computer-readable instructions that are executed by a processor to implement a method for predicting the location of trajectory points as described in any one of claims 1 to 7.

Images