Lane capacity prediction method and system based on big data analysis

Abstract

The application relates to the technical field of traffic roads, and provides a lane traffic capacity prediction method and system based on big data analysis. The method comprises the following steps: acquiring multi-angle lane information by performing multi-angle acquisition on a target lane based on an image acquisition device; acquiring target lane basic data; monitoring the state in the target lane based on an infrared detection technology; outputting a theoretical vehicle flow with a time identifier based on a real-time theoretical traffic model, and predicting a theoretical driving track of a target vehicle according to the theoretical vehicle flow with the time identifier; acquiring an actual track of the target vehicle based on a path tracking algorithm; performing deviation analysis, and predicting the traffic capacity of the target lane based on a deviation analysis result. The application solves the technical problems that the workload of investigation is huge, the data precision is low, and the reliability is low in the prior art, and achieves the technical effect of improving the prediction precision of vehicle road traffic capacity.

Description

Technical Field

[0001] This application relates to the field of traffic and road technology, and in particular to a method and system for predicting lane capacity based on big data analysis. Background Technology

[0002] Traditional lane capacity surveys rely on manual methods to obtain raw traffic flow parameters. This approach is extremely labor-intensive, results in low data accuracy, and leads to a small sample size and low reliability. However, with advancements in technology, a wide range of high-speed traffic information can be collected through various methods, accumulating a vast amount of raw traffic flow data.

[0003] In summary, this application solves the technical problems of existing technologies, such as the huge workload of investigations, low accuracy of data acquisition, and low reliability. Summary of the Invention

[0004] Therefore, it is necessary to address the aforementioned technical problems by providing a lane capacity prediction method and system based on big data analysis that can improve the prediction accuracy of vehicle traffic routes. This achieves the technical effect of improving the prediction accuracy of lane capacity and ensuring traffic safety.

[0005] In a first aspect, embodiments of this application provide a lane capacity prediction method based on big data analysis. The method includes: acquiring multi-angle lane information by using an image acquisition device; comparing and analyzing the multi-angle lane information with N lane information in big data to obtain basic data of the target lane; monitoring the state within the target lane using infrared detection technology to obtain the real-time state of the target lane; inputting the basic data of the target lane and the real-time state of the target lane into a real-time theoretical traffic model to output a theoretical traffic flow with time stamps; predicting the theoretical driving trajectory of the target vehicle based on the theoretical traffic flow with time stamps to obtain the theoretical trajectory of the target vehicle; obtaining the actual trajectory of the target vehicle based on a path tracking algorithm; performing deviation analysis by traversing the theoretical trajectory and the actual trajectory of the target vehicle; and predicting the capacity of the target lane based on the deviation analysis results.

[0006] Secondly, embodiments of this application provide a lane capacity prediction system based on big data analysis. The system includes: a multi-angle lane information acquisition module, used to acquire multi-angle lane information from a target lane using an image acquisition device; a target lane basic data acquisition module, used to compare and analyze the multi-angle lane information with N lane information from big data to acquire target lane basic data; a target lane real-time status acquisition module, used to monitor the status within the target lane using infrared detection technology to acquire the real-time status of the target lane; and a target vehicle theoretical trajectory acquisition module. The system comprises the following modules: a target vehicle theoretical trajectory acquisition module, which inputs the target lane's basic data and real-time status into a real-time theoretical traffic model, outputs a time-stamped theoretical traffic flow, and predicts the target vehicle's theoretical trajectory based on the time-stamped theoretical traffic flow; a target vehicle actual trajectory acquisition module, which acquires the target vehicle's actual trajectory based on a path tracking algorithm; and a capacity prediction module, which performs deviation analysis by traversing the target vehicle's theoretical trajectory and actual trajectory, and predicts the target lane's capacity based on the deviation analysis results.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] First, multi-angle image acquisition is used to capture lane information from the target lane. Second, this multi-angle lane information is compared and analyzed with N lane information from a large dataset to obtain basic data for the target lane. Next, infrared detection technology is used to monitor the state within the target lane to obtain its real-time status. Then, the basic data and real-time status of the target lane are input into a real-time theoretical traffic model, which outputs a time-stamped theoretical traffic flow. Based on this time-stamped theoretical traffic flow, the theoretical trajectory of the target vehicle is predicted to obtain its theoretical trajectory. Then, a path tracking algorithm is used to obtain the actual trajectory of the target vehicle. Finally, by traversing the theoretical trajectory and the actual trajectory of the target vehicle, a deviation analysis is performed, and the traffic capacity of the target lane is predicted based on the deviation analysis results. This application solves the technical problems of the existing technology, such as the huge workload of investigation, low data accuracy, and low reliability, achieving the technical effect of improving the prediction accuracy of lane traffic and ensuring traffic safety.

[0009] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0010] Figure 1 This is a flowchart illustrating a lane capacity prediction method based on big data analysis in one embodiment.

[0011] Figure 2 This is a flowchart illustrating the process of determining the theoretical driving trajectory of a target vehicle using a lane capacity prediction method based on big data analysis in one embodiment.

[0012] Figure 3 This is an internal structure diagram of a lane capacity prediction method based on big data analysis in one embodiment.

[0013] Explanation of reference numerals in the attached diagram: 11 Multi-angle lane information acquisition module, 12 Target lane basic data acquisition module, 13 Target lane real-time status acquisition module, 14 Target vehicle theoretical trajectory acquisition module, 15 Target vehicle actual trajectory acquisition module, 16 Traffic capacity prediction module. Detailed Implementation

[0014] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0015] After introducing the basic principles of this application, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments of this application. It should be understood that this application is not limited to the exemplary embodiments described herein. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application. It should also be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all of them.

[0016] like Figure 1 As shown, this application provides a lane capacity prediction method based on big data analysis, the method comprising:

[0017] S100: Based on image acquisition equipment, multi-angle acquisition of target lane information is obtained;

[0018] Specifically, lane capacity refers to the ability of road facilities to manage traffic flow. It is the capacity of road facilities to handle traffic flow points within a certain time period (usually 15 minutes or 1 hour) and under normal road, traffic, control, and operational quality requirements.

[0019] The target lane refers to any lane for which lane capacity prediction is required; the image acquisition device refers to the device that uses the photoelectric function of photoelectric devices to convert the light image on the photosensitive surface into an electrical signal that is proportional to the light image, such as charge-coupled devices, contact image sensors, cameras, scanners, etc., specifically referring to the camera that captures the target lane; multi-angle lane information refers to the target lane captured by cameras set at different angles, such as capturing the target lane directly above it, capturing the target lane at a 45° angle directly in front of it, etc.

[0020] The image acquisition device captures images of the target lane from different angles, and collects these images to obtain multi-angle lane information. Images of the target lane are obtained based on different deployment locations, ensuring the completeness of the acquired multi-angle lane information.

[0021] S200: By comparing and analyzing the multi-angle lane information with N lane information in the big data, the basic data of the target lane is obtained;

[0022] Specifically, N lane information refers to the basic characteristic data of multiple lanes, such as the length and width of the lanes. In big data, it refers to the data that has already been recorded in big data. The basic data of the target lane refers to the basic data such as the length and width of the target lane.

[0023] The basic data of the target lane is obtained by comparing and analyzing the multi-angle lane information with N lane information in the big data.

[0024] Furthermore, the steps in this application also include:

[0025] S210: Based on the multi-angle lane information, extract the lane top view image information using an image acquisition device deployed perpendicular to the center point of the target lane;

[0026] S220: Using the lane top-view image information, compare the lane top-view image information with N lane top-view information with lane data in the big data to obtain the lane width and lane curvature of the target lane.

[0027] S230: Add the lane width and lane curvature of the target lane to the target lane base data.

[0028] Specifically, lane top-view image information refers to the image information captured by a camera directly above the center point of the target lane; in big data, N lane top-view information with lane data refers to the data that already exists and is recorded in big data, including lane top-view information for multiple lanes, such as lane length, lane width, etc.

[0029] Based on the images obtained from the multi-angle lane information, a camera positioned perpendicularly above the center point of the target lane is located. The top-down view of the lane is extracted. This top-down view is then compared with N lane top-down views containing lane data from a large dataset. The width and curvature of the target lane are identified based on consistent data, and these parameters are added to the target lane's basic data. By finding the top-down view, the specific basic data of the target lane can be obtained, providing a foundation for subsequent calculations and analysis.

[0030] S300: Based on infrared detection technology, monitor the state within the target lane and obtain the real-time state of the target lane;

[0031] Specifically, infrared detection technology is a non-destructive testing technology that uses the characteristic of objects radiating infrared light to perform non-contact infrared temperature recording. In this application, this technology is used to detect the distance between vehicles in the target lane and obstacles in the lane, and the object carrying it is an infrared probe; the real-time status of the target lane refers to the status of the target lane at the same time during the monitoring process, such as whether there is a car accident or traffic jam during the monitoring.

[0032] The status of the target lane is monitored using infrared detection technology to obtain the real-time status of the target lane.

[0033] Furthermore, the steps in this application also include:

[0034] S310: Using infrared detectors deployed in the target lane, infrared detection is performed on vehicles in the target lane to obtain the vehicle spacing.

[0035] S320: Detect the length of adjacent vehicles based on the vehicle spacing, monitor obstacles in the target lane according to the detection results, and obtain obstacle information;

[0036] S330: Determine the real-time status of the target lane based on the impact of the obstacle information on the vehicle's driving within the target lane.

[0037] Specifically, infrared detection refers to a detection method that converts incident infrared radiation signals into electrical signals; vehicle spacing refers to the distance between the front of one vehicle and the rear of the adjacent vehicle in the target lane; obstacles refer to things that prevent vehicles from moving normally, such as accident scenes, barriers in the lane, or litter in the lane; driving impact refers to how obstacles affect the movement of vehicles in the target lane, such as an accident scene making it impossible for vehicles to pass, or litter in the lane generally not affecting normal driving; and the real-time status of the target lane refers to the state in which the target vehicle is, such as being able to pass, needing to detour, or being prohibited from passing.

[0038] Infrared detectors deployed within the target lane are used to detect vehicles in the target lane and obtain vehicle spacing. Based on the vehicle spacing, the length of adjacent vehicles is detected. Obstacles within the target lane are monitored according to the detection results to obtain obstacle information. The real-time status of the target lane is determined based on the impact of the obstacle information on vehicle travel within the target lane.

[0039] S400: By inputting the target lane basic data and the target lane real-time status into the real-time theoretical traffic model, outputting the theoretical traffic flow with time stamp, and predicting the theoretical driving trajectory of the target vehicle based on the theoretical traffic flow with time stamp, the theoretical trajectory of the target vehicle is obtained.

[0040] Specifically, the target vehicle refers to the vehicle being monitored in the lane; the real-time theoretical traffic model refers to the construction of a model that, based on the target lane's basic data and real-time status, yields a conceptual model of the target vehicle's real-time passage in the target lane; the time-stamped theoretical traffic flow refers to the traffic flow within a specific time period of the target lane, where the time stamp is an arbitrary selection of a time period within the lane and a corresponding marker is assigned to that time period. For example, if the real-time status of the target lane is determined by infrared monitoring between 10:00 and 10:10, the theoretical traffic flow is the traffic flow that should have passed between 10:00 and 10:10 without external influences. Traffic flow refers to the number of vehicles passing through the target lane per unit time, representing the number of vehicles passing through a certain point on a highway within a given time period. Infrared detection can determine the traffic flow in different time periods; the theoretical driving trajectory refers to the trajectory of the vehicle under the influence of no external factors, specifically the aforementioned obstacles.

[0041] By inputting the target lane's basic data and real-time status into a real-time theoretical traffic model, the model outputs a time-stamped theoretical traffic flow. Based on this time-stamped theoretical traffic flow, the theoretical driving trajectory of the target vehicle is predicted, such as the possible trajectory of the target vehicle when traffic flow is high and the possible trajectory of the target vehicle when traffic flow is low, thereby obtaining the target vehicle's theoretical trajectory. This can be compared with the vehicle's actual trajectory later.

[0042] like Figure 2 As shown, the steps of this application further include:

[0043] S410: Collect the target lane length of the target lane;

[0044] S420: Based on the theoretical traffic flow with time identifier, extract the first training trajectory from the training trajectory database based on each time period, and the first training trajectory has a first lane length identifier.

[0045] S430: Determine whether the length difference between the target lane length and the first training trajectory meets the predetermined lane length difference threshold;

[0046] S440: If the conditions are met, the first training trajectory is taken as the theoretical trajectory of the target vehicle in the target lane.

[0047] Specifically, the training trajectory database refers to the collection of data on the possible trajectories of vehicles traveling in the current target lane. This database contains the driving trajectories of all required vehicles within the lane, allowing for comparison of data such as the future trajectories of vehicles and the presence of curves in the lane. For example, if a lane allows two vehicles to turn simultaneously, but one vehicle is blocking the way, only one vehicle can turn. To ensure the predicted passage of the lane, the trajectory of the target vehicle must be tracked, and the data collected and compared. The first training trajectory refers to a randomly selected training trajectory from the training trajectory database. The predetermined lane length difference threshold is a manually set threshold, specifically the maximum difference between the target lane length and the first training trajectory.

[0048] The target lane length of the target lane is collected; based on the theoretical traffic flow with time identifier, the first training trajectory in the training trajectory database is extracted based on each time period, and the first training trajectory has a first lane length identifier; it is determined whether the length difference between the target lane length and the first training trajectory meets the predetermined lane length difference threshold. If it does, the first training trajectory is used as the theoretical trajectory of the target vehicle in the target lane.

[0049] S500: Obtains the actual trajectory of the target vehicle based on a path tracking algorithm;

[0050] Specifically, the path tracking algorithm refers to a trajectory tracking control method that achieves lateral control of a vehicle based on the kinematic relationship between the vehicle's position and a reference trajectory. It features high computation speed, strong real-time performance, and good adaptability. Its principle is based on the geometric relationship between the vehicle and the reference path. An algorithm is established between fixed parameters such as the aiming distance, the aiming point, and the vehicle's position to calculate the turning radius and driving curvature. Finally, based on the known vehicle wheelbase and aiming distance, the front wheel steering angle control value is calculated. Under the action of this control value, the vehicle can continuously approach the desired path, achieving trajectory tracking. The actual trajectory of the target vehicle is obtained according to the algorithm.

[0051] Furthermore, the steps in this application include:

[0052] S510: Extract the first target image from the target lane image time sequence;

[0053] S520: Locate the first pixel region of the target lane in the first target image, and collect the first longest diameter pixel number of the first pixel region;

[0054] S530: Read the number of pre-aimed pixels, and draw a target circle with the center of the first longest diameter pixel count as the center and the number of pre-aimed pixels as the radius;

[0055] S540: The intersection point of the target circle and the theoretical trajectory of the target vehicle is taken as the pre-aiming pixel point;

[0056] S550: Obtain the first actual movement direction of the first pixel region, and take the angle between the actual movement direction of the target vehicle and the pre-aimed pixel as the first deviation angle;

[0057] S560: Calculate the first actual turning angle of the target vehicle based on the first longest diameter pixel count, the pre-aiming pixel count, and the first deviation angle;

[0058] S570: Extract the second target image from the target lane image time sequence, and analyze the second target image to obtain the second actual turning angle, wherein the first target image and the second target image are adjacent images;

[0059] S580: Based on the first actual turning angle and the second actual turning angle, generate the actual trajectory of the target vehicle.

[0060] Specifically, the target lane image sequence refers to images with time stamps acquired by an image acquisition device; the first target image refers to the first image after arranging the target lane images in chronological order; the first pixel region refers to the position of the target lane in the first image; the first longest diameter pixel count corresponds to the length of the target lane; the target circle refers to the intersection point used to calculate the first deviation angle; the intersection point refers to the point where lines intersect or lines intersect with surfaces; the pre-aiming pixel count refers to a preset number of plotting pixels, which can be set to the center of the rear wheel of the vehicle; the first actual direction of movement refers to the actual direction of movement of the target vehicle on the target lane. The included angle refers to the smallest positive angle formed by the intersection of two straight lines (or vectors); the first deviation angle is determined by the fact that a theoretical trajectory is preset to predict the traffic capacity of the target vehicle channel. Since the actual vehicle trajectory differs from the predicted theoretical trajectory, the deviation angle between the actual movement direction of the target vehicle and the theoretical trajectory is used for judgment. Therefore, Yuyao uses the angle difference between the actual movement direction of the target vehicle and the pre-aimed pixel as the first deviation angle; the first actual turning angle is obtained based on the number of pixels with the first longest diameter, the number of pre-aimed pixels, and the first deviation angle according to the formula:

[0061]

[0062] Where L is the number of pixels with the first longest diameter, Ld is the number of pixels to be aimed, and θ is the first deviation angle; from the above, the second actual turning angle can be obtained. The angle between the actual movement direction of the target vehicle and the number of pixels to be aimed is the second deviation angle. The second actual turning angle is obtained according to the formula based on the number of pixels with the first longest diameter, the number of pixels to be aimed, and the second deviation angle.

[0063] Extract a first target image from the time sequence of the target lane image; locate the first pixel region of the target lane in the first target image and collect the first longest diameter pixel count of the first pixel region; read the pre-aimed pixel count and draw a target circle with the center of the first longest diameter pixel count as the center and the pre-aimed pixel count as the radius; take the intersection of the target circle and the theoretical trajectory of the target vehicle as the pre-aimed pixel point; obtain the first actual movement direction of the first pixel region and take the angle between the actual movement direction of the target vehicle and the pre-aimed pixel point as the first deviation angle; calculate the first actual turning angle of the target vehicle based on the first longest diameter pixel count, the pre-aimed pixel count and the first deviation angle; extract a second target image from the time sequence of the target lane image and analyze the second target image to obtain a second actual turning angle, wherein the first target image and the second target image are adjacent images; generate the actual trajectory of the target vehicle based on the first actual turning angle and the second actual turning angle.

[0064] S600: By traversing the theoretical trajectory of the target vehicle and the actual trajectory of the target vehicle, deviation analysis is performed, and the traffic capacity of the target lane is predicted based on the deviation analysis results.

[0065] Specifically, traversal literally means going through all nodes. In program code, it means visiting each node in the tree once and only once along a certain search route. The operations performed on each node depend on the specific application problem. In this embodiment, traversal means visiting all the theoretical trajectory and actual trajectory data of the target vehicle. Deviation analysis means that since there may be deviations between the theoretical trajectory and the actual trajectory, there will be differences between the lane capacity predicted when driving according to the theoretical trajectory and the lane capacity when driving according to the actual trajectory. Predicting the lane capacity of the target lane based on the deviation analysis results means predicting the lane capacity based on the magnitude of this difference.

[0066] By performing deviation analysis between the theoretical trajectory and the actual trajectory of the target vehicle, the traffic capacity of the target lane is predicted based on the deviation analysis results.

[0067] Furthermore, the steps in this application include:

[0068] S610: Obtain the target theoretical trajectory by traversing the training trajectory database;

[0069] S620: Based on the theoretical trajectory of the target lane, analyze the time series of the target lane image to obtain the actual trajectory of the target lane;

[0070] S630: Compare the theoretical trajectory of the target with the actual trajectory of the target to obtain the deviation analysis results.

[0071] Specifically, the target theoretical trajectory refers to the theoretical driving trajectory of the target vehicle; the deviation refers to the non-overlapping part between the target theoretical trajectory and the target actual trajectory.

[0072] The target theoretical trajectory is obtained by accessing all data in the training trajectory database; based on the target theoretical trajectory, the target lane image time series is analyzed to obtain the target actual trajectory of the target lane; the target theoretical trajectory is compared with the target actual trajectory to obtain the deviation analysis result. This application solves the technical problems of huge workload, low accuracy and reliability of data acquisition in the prior art, and achieves the technical effect of improving the prediction accuracy of vehicle road traffic and ensuring traffic safety.

[0073] like Figure 3 As shown, this application provides a lane capacity prediction method and system based on big data analysis, the system comprising:

[0074] Multi-angle lane information acquisition module 11, the multi-angle lane information acquisition module 11 is used to acquire multi-angle lane information by acquiring the target lane based on the image acquisition device;

[0075] The target lane basic data acquisition module 12 is used to compare and analyze the multi-angle lane information with N lane information in the big data to obtain the target lane basic data.

[0076] The target lane real-time status acquisition module 13 is used to monitor the status within the target lane based on infrared detection technology and acquire the real-time status of the target lane.

[0077] The target vehicle theoretical trajectory acquisition module 14 is used to input the target lane basic data and the target lane real-time status into a real-time theoretical traffic model, output a theoretical traffic flow with time stamp, and predict the theoretical driving trajectory of the target vehicle based on the theoretical traffic flow with time stamp to obtain the theoretical trajectory of the target vehicle.

[0078] The target vehicle actual trajectory acquisition module 15 acquires the actual trajectory of the target vehicle based on a path tracking algorithm.

[0079] The capacity prediction module 16 is used to perform deviation analysis by traversing the theoretical trajectory of the target vehicle and the actual trajectory of the target vehicle, and predict the capacity of the target lane based on the deviation analysis results.

[0080] Furthermore, embodiments of this application include:

[0081] The lane top-view image information extraction module is used to extract lane top-view image information based on the multi-angle lane information and according to the image acquisition device deployed perpendicular to the center point of the target lane.

[0082] The target lane curvature acquisition module is used to compare the lane top view image information with N lane top view information with lane data in the big data to obtain the lane width and lane curvature of the target lane.

[0083] The target lane basic data module is used to add the lane width and lane curvature of the target lane to the target lane basic data.

[0084] Furthermore, embodiments of this application include:

[0085] The vehicle spacing acquisition module is used to perform infrared detection on vehicles in the target lane by using infrared probes deployed in the target lane to acquire the vehicle spacing.

[0086] An obstacle information acquisition module is used to detect the length of adjacent vehicles based on the vehicle spacing, monitor obstacles in the target lane according to the detection results, and acquire obstacle information.

[0087] The target lane real-time status determination module is used to determine the real-time status of the target lane based on the impact of the obstacle information on the vehicle's driving within the target lane.

[0088] Furthermore, embodiments of this application also include:

[0089] A target lane length acquisition module, wherein the target lane length acquisition module is used to acquire the target lane length of the target lane;

[0090] The training trajectory database data extraction module is used to extract the first training trajectory from the training trajectory database based on the theoretical traffic flow with time identifier and each time period, and the first training trajectory has a first lane length identifier.

[0091] The lane length difference threshold determination module is used to determine whether the length difference between the target lane length and the first lane length meets the predetermined lane length difference threshold.

[0092] The theoretical trajectory acquisition module is used to, if a match is found, use the first training trajectory as the theoretical trajectory of the target vehicle within the target lane.

[0093] Furthermore, embodiments of this application also include:

[0094] A first target image extraction module is used to extract a first target image from the target lane image time sequence;

[0095] The first longest diameter pixel acquisition module is used to locate the target lane in the first pixel region of the first target image and acquire the first longest diameter pixel number of the first pixel region.

[0096] A target circle drawing module is used to read the number of pre-aimed pixels and draw a target circle with the center of the first longest diameter pixel count as the center and the number of pre-aimed pixels as the radius.

[0097] A pre-aiming pixel acquisition module is used to take the intersection of the target circle and the theoretical trajectory of the target vehicle as the pre-aiming pixel.

[0098] The first deviation angle acquisition module is used to acquire the first actual movement direction of the first pixel region and take the angle between the actual movement direction of the target vehicle and the pre-aimed pixel as the first deviation angle.

[0099] The first actual turning angle acquisition module is used to calculate the first actual turning angle of the target vehicle based on the first longest diameter pixel count, the pre-aiming pixel count and the first deviation angle.

[0100] The second target image extraction module is used to extract the second target image in the time sequence of the target lane image and analyze the second target image to obtain the second actual turning angle, wherein the first target image and the second target image are adjacent images;

[0101] The actual trajectory generation module is used to generate the actual trajectory of the target vehicle based on the first actual turning angle and the second actual turning angle.

[0102] Furthermore, embodiments of this application also include:

[0103] A target theoretical trajectory acquisition module is used to traverse the training trajectory database to obtain the target theoretical trajectory.

[0104] The target actual trajectory acquisition module is used to obtain the target actual trajectory of the target lane by analyzing the time series of the target lane image based on the target theoretical trajectory.

[0105] The deviation analysis result acquisition module is used to compare the theoretical trajectory of the target with the actual trajectory of the target to obtain the deviation analysis result.

[0106] For specific embodiments of the lane capacity prediction system based on big data analysis, please refer to the embodiments of the lane capacity prediction method based on big data analysis described above, which will not be repeated here. The above modules can be embedded in hardware or independent of the processor in a computer device, or stored in software in the memory of a computer device, so that the processor can call and execute the operations corresponding to each module.

[0107] 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.

[0108] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A lane capacity prediction method based on big data analysis, characterized in that, The method includes: The target lane is captured from multiple angles using image acquisition equipment to obtain multi-angle lane information. By comparing and analyzing the multi-angle lane information with N lane information in the big data, the basic data of the target lane is obtained. Based on infrared detection technology, the state within the target lane is monitored to obtain the real-time state of the target lane; By inputting the target lane's basic data and the target lane's real-time status into the real-time theoretical traffic model, the theoretical traffic flow with time stamps is output. Based on the theoretical traffic flow with time stamps, the theoretical driving trajectory of the target vehicle is predicted, and the theoretical trajectory of the target vehicle is obtained. The actual trajectory of the target vehicle is obtained based on the path tracing algorithm; By performing deviation analysis on the theoretical trajectory and the actual trajectory of the target vehicle, the traffic capacity of the target lane is predicted based on the deviation analysis results. The method for predicting the theoretical driving trajectory of the target vehicle based on the theoretical traffic flow with time stamps to obtain the theoretical trajectory of the target vehicle further includes: Collect the target lane length of the target lane; Based on the theoretical traffic flow with time identifier, the first training trajectory in the training trajectory database is extracted based on each time period, and the first training trajectory has a first lane length identifier. Determine whether the length difference between the target lane length and the first lane length meets the predetermined lane length difference threshold; If the conditions are met, the first training trajectory will be used as the theoretical trajectory of the target vehicle within the target lane. The method for obtaining the actual trajectory of the target vehicle based on the path tracking algorithm further includes: Extract the first target image from the time sequence of the target lane image; Locate the first pixel region of the target lane in the first target image, and collect the first longest diameter pixel number of the first pixel region; Read the number of pre-aimed pixels, and draw a target circle with the center of the first longest diameter pixel count as the center and the number of pre-aimed pixels as the radius; The intersection of the target circle and the theoretical trajectory of the target vehicle is taken as the pre-aiming pixel point; Obtain the first actual movement direction of the first pixel region, and take the angle between the actual movement direction of the target vehicle and the pre-aimed pixel as the first deviation angle; The first actual turning angle of the target vehicle is calculated based on the first longest diameter pixel count, the pre-aiming pixel count, and the first deviation angle. Extract the second target image from the time sequence of the target lane image, and analyze the second target image to obtain the second actual turning angle, wherein the first target image and the second target image are adjacent images; The actual trajectory of the target vehicle is generated based on the first actual turning angle and the second actual turning angle.

2. The method as described in claim 1, characterized in that, The method for obtaining the basic data of the target lane further includes: Based on the multi-angle lane information, the lane top view image information is extracted using an image acquisition device deployed perpendicular to the center point of the target lane. By comparing the lane top-view image information with N lane top-view information with lane data in the big data, the lane width and lane curvature of the target lane are obtained. Add the lane width and lane curvature of the target lane to the target lane base data.

3. The method as described in claim 1, characterized in that, The method for obtaining the real-time status of the target lane further includes: Infrared detectors deployed within the target lane are used to detect vehicles within the target lane and obtain vehicle distances. The length of adjacent vehicles is detected based on the vehicle spacing, and obstacles in the target lane are monitored based on the detection results to obtain obstacle information; The real-time status of the target lane is determined based on the impact of the obstacle information on the vehicle's driving within the target lane.

4. The method as described in claim 1, characterized in that, The method for analyzing the deviation between the theoretical trajectory of the target vehicle and the actual trajectory of the target vehicle further includes: The target theoretical trajectory is obtained by traversing the training trajectory database. Based on the theoretical trajectory of the target lane, the actual trajectory of the target lane is obtained by analyzing the time series of the target lane image; The theoretical trajectory of the target is compared with the actual trajectory of the target to obtain the deviation analysis results.

5. A lane capacity prediction system based on big data analysis, characterized in that, The system is used to implement the lane capacity prediction method based on big data analysis as described in any one of claims 1 to 4, including: A multi-angle lane information acquisition module is used to acquire multi-angle lane information by acquiring the target lane from multiple angles based on an image acquisition device. The target lane basic data acquisition module is used to compare and analyze the multi-angle lane information with N lane information in the big data to obtain the target lane basic data. The target lane real-time status acquisition module is used to monitor the status within the target lane based on infrared detection technology and acquire the real-time status of the target lane. The target vehicle theoretical trajectory acquisition module is used to input the target lane basic data and the target lane real-time status into a real-time theoretical traffic model, output a time-stamped theoretical traffic flow, and predict the theoretical driving trajectory of the target vehicle based on the time-stamped theoretical traffic flow to obtain the theoretical trajectory of the target vehicle. A target vehicle actual trajectory acquisition module, wherein the target vehicle actual trajectory acquisition module acquires the actual trajectory of the target vehicle based on a path tracking algorithm; The capacity prediction module is used to perform deviation analysis by traversing the theoretical trajectory and the actual trajectory of the target vehicle, and predict the capacity of the target lane based on the deviation analysis results.

Images