A method and system for tracking hit-and-run accidents based on unmanned vehicles and air-ground integration.

Abstract

This invention relates to the field of intelligent transportation technology, specifically to a method and system for tracking hit-and-run accidents based on unmanned vehicles and air-ground integration. The method includes: installing road traffic monitoring equipment at various road nodes to monitor road traffic conditions and the characteristics of fleeing vehicles in real time, and transmitting this information to an unmanned enforcement vehicle and a regional traffic data processing center; upon receiving the fleeing vehicle's characteristics, the unmanned enforcement vehicle triggers an automatic takeoff procedure for a drone; after takeoff, the drone locks onto the fleeing vehicle based on the provided characteristics and transmits real-time information about the fleeing vehicle to the regional traffic data processing center; the regional traffic data processing center receives all information and generates a real-time dynamic trajectory map of the fleeing vehicle, and formulates a deployment plan. This invention integrates information from monitoring equipment, unmanned enforcement vehicles, and drones to achieve the discovery, tracking, and deployment of vehicles fleeing road traffic accidents, improving the detection rate and processing efficiency of hit-and-run cases.

Description

Technical Field

[0001] This invention relates to the field of intelligent transportation technology, specifically to a method and system for tracking hit-and-run accidents based on unmanned vehicles and air-ground integration. Background Technology

[0002] There are many problems that urgently need to be solved in the current handling of hit-and-run cases. At present, the main reliance is on fixed road traffic monitoring equipment for preliminary accident monitoring and evidence collection. However, these fixed devices have significant monitoring blind spots and cannot fully cover all road sections, especially in some remote areas, suburban areas, and sections with complex road conditions, where the monitoring coverage is seriously insufficient.

[0003] Traditional traffic enforcement relies mainly on law enforcement officers patrolling and handling incidents in law enforcement vehicles. After receiving a report of a hit-and-run accident, the response speed of law enforcement vehicles to the scene and to launch a pursuit is slow. Moreover, in situations such as traffic congestion and complex road networks, the pursuit is extremely difficult, often resulting in the escape of the fleeing vehicle, which increases the difficulty and time cost of solving the case.

[0004] Although drone technology has been applied to some extent in the field of traffic monitoring, enabling aerial monitoring and tracking, the use of drones alone has problems such as limited endurance, susceptibility to weather conditions, and untimely information exchange with ground law enforcement forces. This makes it difficult to form a continuous and efficient tracking situation, which seriously affects the tracking effect and final capture efficiency of vehicles fleeing traffic accidents.

[0005] Therefore, there is an urgent need for a technical solution that can integrate multiple monitoring and law enforcement resources to achieve air-ground coordination, rapid response, and efficient tracking, in order to solve the aforementioned defects in existing road traffic accident hit-and-run tracking technologies. Summary of the Invention

[0006] To address the aforementioned technical shortcomings, this invention provides a method and system for tracking hit-and-run accidents based on unmanned vehicles and air-ground integration, thereby solving the problems of numerous blind spots, slow response speed, poor tracking effect, untimely information exchange, and low capture efficiency in existing road traffic accident hit-and-run tracking technologies.

[0007] This invention is achieved through the following technical solution:

[0008] A method for tracking hit-and-run accidents based on unmanned vehicles and air-to-ground integration is provided. The method includes the following steps:

[0009] Step S10: Install road traffic monitoring equipment at each road node to monitor road traffic conditions in real time, and send the road traffic conditions information and the escape vehicle feature information obtained through image recognition and intelligent analysis to the unmanned law enforcement vehicle and the regional traffic data processing center.

[0010] Step S20: Distribute unmanned law enforcement vehicles to patrol dynamically along preset patrol routes on various roads. Each unmanned law enforcement vehicle is equipped with a drone. When the unmanned law enforcement vehicle receives the characteristic information of the escaped vehicle, it enters the emergency response state and triggers the drone's automatic take-off procedure.

[0011] Step S30: After the drone takes off, it locks the target vehicle based on the characteristic information of the fleeing vehicle provided by the unmanned law enforcement vehicle and monitoring equipment. During the tracking process, it captures images and dynamic videos of the fleeing vehicle in real time and transmits accurate information to the unmanned law enforcement vehicle and the regional traffic data processing center in real time.

[0012] Step S40: The regional traffic data processing center receives and integrates all information from monitoring equipment, unmanned law enforcement vehicles, and drones, and generates a real-time dynamic trajectory map of the escaped vehicle. The command center formulates a deployment plan based on the map, combined with the traffic network layout and police force distribution in the jurisdiction, and implements precise deployment.

[0013] Preferably, the road traffic monitoring equipment installed at each road node in step S10 includes high-definition cameras, millimeter-wave radar, and infrared sensors, etc., to monitor road traffic conditions in real time, including road traffic images, vehicle speed, vehicle trajectory, and traffic flow.

[0014] Preferably, the step S10, which involves obtaining the escape vehicle feature information through image recognition and intelligent analysis, includes:

[0015] Image data preprocessing: The acquired road traffic images are denoised, enhanced, and have regions of interest (ROIs) extracted. Spatiotemporal calibration is performed on vehicle speeds and trajectories acquired by millimeter-wave radar with the road traffic images to correct data misalignment caused by equipment installation angle deviations or signal delays. This ensures accurate matching of image features and motion parameters for the same vehicle, avoiding misjudgments. Denoising involves using Gaussian filtering or median filtering algorithms to remove salt-and-pepper noise (Gaussian noise) from the road traffic images. Image enhancement uses histogram equalization to improve the contrast of low-light images and the Retinex algorithm to correct uneven image brightness caused by backlighting and shadows, ensuring the clarity of key targets such as vehicles and road markings. ROI extraction is based on road geographic coordinates and lane line recognition results, cropping meaningless areas from the image and retaining only lane driving areas and road edge areas to reduce the amount of data for subsequent analysis and improve processing efficiency.

[0016] Vehicle and road feature extraction: Based on a deep learning model, vehicle target detection is performed on the preprocessed road traffic images, and key features of each vehicle are extracted, including appearance features and motion features. Appearance features include vehicle type, body markings, and license plate number. Motion features include real-time driving speed, acceleration, steering angle, relative motion direction and distance with surrounding vehicles, combined with millimeter-wave radar data, to form the dynamic motion trajectory vector of each vehicle. Road features are extracted through image segmentation algorithms.

[0017] Traffic accident determination: Based on extracted vehicle and road features, traffic accidents are identified through abnormal motion state determination combined with road surface abnormality determination. Abnormal motion state determination includes vehicle speed changes and vehicle collision features. Vehicle speed changes are specifically defined as sudden deceleration of the vehicle, with an absolute value of vehicle acceleration ≥ 2.5 m / s², i.e., emergency braking, lasting for more than 1 second, abnormal stopping of the vehicle, and the vehicle remaining stationary in a non-parking area for more than 3 seconds without normal parking behaviors such as turning on hazard lights or engaging the handbrake. Vehicle collision features are specifically defined as the movement trajectories of two or more vehicles intersecting and overlapping, with at least one vehicle braking suddenly at the moment of overlap, with an absolute value of acceleration ≥ 2.5 m / s², and vehicle deformation appearing in the image. Road surface abnormality determination is achieved by setting road surface conditions. When the set road surface conditions occur, they are correlated with the abnormal motion state determination to confirm the occurrence of a traffic accident. The set road surface conditions specifically include abnormal traffic behaviors such as debris on the road surface after a vehicle collision, pedestrians gathering in the accident area, and non-motorized vehicles detouring, and abnormal temperature increases on the hood and body of the accident vehicle detected in the infrared image.

[0018] Confirmation of vehicle escape behavior: After identifying a traffic accident, further dynamic tracking and analysis are used to confirm whether there is any vehicle escape behavior.

[0019] The steps in confirming vehicle escape behavior by further verifying whether vehicle escape behavior exists through dynamic tracking and analysis after identifying a traffic accident include:

[0020] Accident-related vehicle locking: Based on the images and vehicle movement trajectories at the moment of the accident, vehicles directly related to the accident are locked, marked as accident-related vehicle groups, and the position coordinates and movement status of each vehicle are recorded 3 seconds before the accident, at the time of the accident, and 3 seconds after the accident.

[0021] Vehicle dwell time determination: Monitor the dwell time of each vehicle in the accident-related vehicle group at the accident scene. If a vehicle remains within 50 meters of the accident center point after the accident and the driver gets out of the vehicle to check or call the police, it is determined to be a non-hit-and-run vehicle. If a vehicle does not stop and accelerates away from the accident scene within 10 seconds after the accident, the speed limit can be dynamically adjusted according to the road section. For example, it can be adjusted to 5 seconds on highways. Or if the vehicle suddenly turns or accelerates away after a short dwell time of ≤3 seconds without turning on the hazard lights or placing warning signs, it is initially determined to be a hit-and-run vehicle.

[0022] Escape trajectory verification: Real-time vehicle trajectory tracking is performed on the initially identified escape vehicle. Vehicle driving conditions are set. When the set vehicle driving conditions are met, the escape behavior and escape vehicle are confirmed. The set vehicle driving conditions include the vehicle's departure direction being inconsistent with the normal driving direction before the accident, the vehicle's departure speed being significantly higher than the driving speed before the accident, and millimeter-wave radar detecting abnormal driving behaviors such as continuous lane changes and weaving during the vehicle's departure to avoid monitoring or hinder tracking.

[0023] Escape information solidification: After confirming the escape behavior and the escape vehicle, the complete feature information of the escape vehicle is extracted, including vehicle type, color, complete license plate number or partially clear license plate number, vehicle markings, initial escape direction, and precise location coordinates of the departure time. The feature information is then integrated with the collected road traffic condition information to obtain the escape vehicle feature information, including images, video clips and vehicle motion data of the traffic accident, which is transmitted in real time to the unmanned law enforcement vehicle and the regional traffic data processing center through an encrypted communication link.

[0024] Preferably, the unmanned law enforcement vehicle in step S20 has autonomous navigation, environmental perception, and wireless communication functions. It conducts dynamic patrols on roads within its jurisdiction according to a preset patrol route. The unmanned law enforcement vehicle is equipped with a high-definition panoramic camera, lidar, GPS positioning unit, and emergency response processing unit. It can perceive the surrounding traffic environment in real time and receive information from road traffic monitoring equipment and regional traffic data processing center. When the unmanned law enforcement vehicle receives escape vehicle characteristic information transmitted by road traffic monitoring equipment or discovers escape behavior of traffic accidents through environmental perception, it immediately enters emergency response state. Each unmanned law enforcement vehicle is equipped with at least one drone. The drone is placed on a dedicated take-off and landing platform on the top of the unmanned law enforcement vehicle and has vertical take-off and landing, long endurance, real-time transmission of high-definition images, autonomous obstacle avoidance, and remote control functions. The drone is equipped with a high-definition infrared camera, a high-definition visible light camera, a GPS / BeiDou dual-mode positioning unit, a data transmission radio, and a battery endurance unit.

[0025] Preferably, in step S30, when the unmanned law enforcement vehicle enters the emergency response state, the drone's automatic take-off procedure is triggered. After take-off, the drone quickly locks onto the target vehicle based on the escape vehicle information provided by the road traffic monitoring equipment or the unmanned law enforcement vehicle itself. During the tracking process, the drone captures dynamic videos and images of the escape vehicle in real time, obtains escape vehicle tracking information, including the escape vehicle's precise location, driving trajectory, and surrounding road conditions, and transmits this information in real time to the unmanned law enforcement vehicle and the regional traffic data processing center via a wireless communication link. At the same time, the drone can autonomously adjust its flight trajectory and altitude according to the escape vehicle's driving status to maintain continuous tracking of the escape vehicle. When encountering obstacles or complex terrain, it can autonomously plan a detour route.

[0026] Preferably, in step S40, the regional traffic data processing center receives and integrates all data from road traffic monitoring equipment, unmanned enforcement vehicles, and drones, performs fusion analysis and processing on the data, and generates a real-time dynamic trajectory map of the escaped vehicle. The public security traffic command center is connected to the regional traffic data processing center in real time, and can synchronously obtain the escaped vehicle's characteristic information and tracking information, including the escaped vehicle's precise location, driving trajectory, and surrounding road conditions. Based on the escaped vehicle's characteristic information and tracking information, the command center formulates a deployment plan in conjunction with the traffic network layout and police force distribution in the jurisdiction. It sends instructions to surrounding traffic police officers, other unmanned enforcement vehicles, and traffic signal control systems along the road via wireless communication links to implement precise deployment. The road traffic monitoring equipment, unmanned enforcement vehicles, drones, regional traffic data processing center, and public security traffic command center establish a real-time, high-speed, and stable information interaction channel through 5G / 6G wireless communication networks and dedicated wireless data transmission links. Information transmission is encrypted to ensure data security and confidentiality, and each device has collaborative control logic.

[0027] Furthermore, to achieve the above objectives, this invention also proposes a traffic accident hit-and-run tracking system based on unmanned vehicles and air-ground integration, wherein the traffic accident hit-and-run tracking system based on unmanned vehicles and air-ground integration includes:

[0028] Road traffic monitoring module: Used to install road traffic monitoring equipment at various road nodes to monitor road traffic conditions in real time, and send road traffic conditions information and escape vehicle feature information obtained through image recognition and intelligent analysis to unmanned law enforcement vehicles and regional traffic data processing centers;

[0029] Unmanned law enforcement vehicle dynamic patrol module: used to distribute unmanned law enforcement vehicles on various roads to patrol dynamically according to preset patrol routes. Each unmanned law enforcement vehicle is equipped with a drone. When the unmanned law enforcement vehicle receives the characteristic information of the escaped vehicle, it enters the emergency response state and triggers the drone's automatic take-off procedure.

[0030] Drone aerial tracking module for escaping vehicles: After the drone takes off, it locks onto the target vehicle based on the characteristic information of the escaping vehicle provided by the unmanned law enforcement vehicle and monitoring equipment. During the tracking process, it captures images and dynamic videos of the escaping vehicle in real time and transmits accurate information to the unmanned law enforcement vehicle and the regional traffic data processing center in real time.

[0031] The regional traffic data processing center and the public security traffic command center linkage deployment module is used by the regional traffic data processing center to receive and integrate all information from monitoring equipment, unmanned law enforcement vehicles and drones, generate real-time dynamic trajectory maps of escaped vehicles, and formulate deployment plans based on the maps in combination with the traffic network layout and police force distribution in the jurisdiction to implement precise deployment.

[0032] Furthermore, to achieve the above objectives, the present invention also proposes a traffic accident escape tracking device based on unmanned vehicles and air-ground integration. The device includes: a memory, a processor, and a program for unmanned vehicles and air-ground integration traffic accident escape tracking stored in the memory and executable on the processor. The program for unmanned vehicles and air-ground integration traffic accident escape tracking comprises the steps for implementing the traffic accident escape tracking method based on unmanned vehicles and air-ground integration as described above.

[0033] In addition, to achieve the above objectives, the present invention also provides a computer program product, which includes programs such as unmanned vehicle and air-ground integrated traffic accident escape tracking. When the unmanned vehicle and air-ground integrated traffic accident escape tracking programs are executed by a processor, they implement the traffic accident escape tracking method based on unmanned vehicle and air-ground integrated technology as described above.

[0034] The advantages and effects of this invention are:

[0035] This invention proposes a method and system for tracking hit-and-run accidents based on unmanned vehicles and air-ground integration. By combining real-time monitoring by road traffic monitoring equipment with dynamic patrols by unmanned enforcement vehicles, it can detect and trigger tracking procedures within a very short time after a hit-and-run accident occurs, significantly shortening the response time and avoiding the tracking delays caused by alarm delays and slow police deployment in traditional enforcement models. Simultaneously, the air-ground integrated tracking method allows unmanned enforcement vehicles to cover ground roads, while drones provide all-around aerial monitoring. The collaborative work of both effectively compensates for the blind spots of road traffic monitoring equipment and the limitations of single tracking methods, achieving comprehensive, blind-spot-free tracking of fleeing vehicles. Furthermore, the real-time data collection by drones and unmanned enforcement vehicles... The system collects dynamic information on fleeing vehicles and transmits it to the command center via a high-speed communication network. This ensures that the command center can promptly and accurately grasp the whereabouts and status of fleeing vehicles, providing reliable data support for deployment decisions. Based on the real-time information on fleeing vehicles, the public security traffic command center can quickly formulate and implement precise deployment plans, mobilizing multiple law enforcement forces for coordinated interception. This significantly improves the capture rate of fleeing vehicles, effectively combats hit-and-run traffic accidents, and maintains road traffic order and public safety. Furthermore, the entire system, from the discovery of fleeing behavior and the collection and transmission of tracking information to the issuance of deployment commands, has achieved a high degree of automation and intelligence, reducing reliance on manual operation, lowering the workload of law enforcement personnel, and avoiding errors caused by human factors. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is a flowchart of the traffic accident escape tracking method based on unmanned vehicles and air-ground integration according to the present invention.

[0038] Figure 2 This is a schematic diagram of the structure of the traffic accident escape tracking system based on unmanned vehicles and air-ground integration of the present invention.

[0039] Figure 3 This is a schematic block diagram of the electronic device for tracking hit-and-run accidents based on unmanned vehicles and air-ground integration according to the present invention. Detailed Implementation

[0040] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0041] like Figure 1 As shown, in one embodiment of the present invention, a traffic accident hit-and-run tracking method based on unmanned vehicles and air-ground integration includes the following steps:

[0042] Step S10: Install road traffic monitoring equipment at each road node to monitor road traffic conditions in real time, and send the road traffic conditions information and the escape vehicle feature information obtained through image recognition and intelligent analysis to the unmanned law enforcement vehicle and the regional traffic data processing center.

[0043] Specifically, the road traffic monitoring equipment installed at each road node in step S10 includes high-definition cameras, millimeter-wave radar, and infrared sensors, which monitor road traffic conditions in real time, including road traffic images, vehicle speed, vehicle trajectory, and traffic flow.

[0044] Specifically, step S10, which involves obtaining the escape vehicle's characteristic information through image recognition and intelligent analysis, includes:

[0045] Image data preprocessing: The acquired road traffic images are denoised, enhanced, and regions of interest (ROIs) extracted. Spatiotemporal calibration is performed on vehicle speeds and trajectories acquired by millimeter-wave radar with the road traffic images to correct data misalignment caused by equipment installation angle deviations or signal delays. This ensures accurate matching of image features and motion parameters for the same vehicle, avoiding misjudgments. Denoising involves using Gaussian filtering or median filtering algorithms to remove salt-and-pepper noise (Gaussian noise) from road traffic images, such as raindrop reflections on rainy days and headlight interference at night. Image enhancement uses histogram equalization to improve the contrast of low-light images and the Retinex algorithm to correct uneven image brightness caused by backlighting and shadows, ensuring the clarity of key targets such as vehicles and road markings. ROI extraction is based on road geographic coordinates and lane line recognition results, cropping meaningless areas such as roadside trees and buildings, retaining only lane driving areas and road edge areas to reduce the amount of data for subsequent analysis and improve processing efficiency.

[0046] Vehicle and road feature extraction: Based on deep learning models such as YOLOv8 and Faster R-CNN, vehicle target detection is performed on preprocessed road traffic images to extract key features of each vehicle, including appearance features and motion features. Appearance features include vehicle type, such as sedan, SUV, truck, etc.; body color, such as accurate classification through HSV color space analysis; body markings, such as stickers, headlight shape, wheel style; and license plate number, such as extracting the license plate area and recognizing characters based on the character recognition algorithm CRNN with an accuracy of no less than 99%. Motion features include combining millimeter-wave radar data to extract the vehicle's real-time driving speed, acceleration, steering angle, and relative motion direction and distance with surrounding vehicles to form the dynamic motion trajectory vector of each vehicle. Road feature extraction is performed by using image segmentation algorithms, such as U-Net, to identify road markings, such as lane lines, stop lines, and zebra crossings, as well as road obstacles, such as debris and depressions. At the same time, the system monitors for abnormal stationary targets on the road, such as fallen pedestrians and overturned objects, to provide road environment information for subsequent traffic accident determination.

[0047] Traffic accident identification: Based on extracted vehicle and road features, traffic accidents are identified through abnormal motion state determination combined with abnormal road surface conditions. Abnormal motion state determination includes vehicle speed changes and vehicle collision characteristics. Vehicle speed changes specifically refer to sudden deceleration of the vehicle, with an absolute value of vehicle acceleration ≥2.5 m / s² (i.e., emergency braking), lasting for more than 1 second; abnormal stopping of the vehicle, where the vehicle is in a non-parking area, such as within a lane or a no-parking zone, and remains stationary for more than 3 seconds without normal parking behaviors such as activating hazard lights or engaging the handbrake. Vehicle collision characteristics specifically refer to the overlapping of the movement trajectories of two or more vehicles, with the overlap accompanied by at least one vehicle suddenly... Braking, with an absolute acceleration value ≥2.5m / s², and the presence of vehicle deformation in the image, such as dented front end, damaged rear end, broken windows, and detached body parts, etc., are considered collision marks. The road surface abnormality auxiliary judgment is based on setting road surface conditions. When the set road surface conditions occur, they are correlated with the abnormal motion state judgment to confirm that a traffic accident has occurred. The set road surface conditions specifically include: after a vehicle collision, there are scattered objects on the road surface, such as parts, glass fragments, and oil stains, etc.; abnormal traffic behavior such as pedestrians gathering and non-motorized vehicles detouring around the accident area; and abnormal temperature rise of the engine hood and body parts of the accident vehicle detected in the infrared image, such as mechanical failure caused by the collision.

[0048] Confirmation of vehicle escape behavior: After identifying a traffic accident, further dynamic tracking and analysis are used to confirm whether there is any vehicle escape behavior.

[0049] Specifically, the steps in confirming vehicle escape behavior by further verifying whether vehicle escape behavior has occurred after identifying a traffic accident through dynamic tracking and analysis include:

[0050] Accident-related vehicle locking: Based on the images and vehicle movement trajectories at the moment of the accident, lock the vehicles directly related to the accident, such as the two parties involved in the collision and the vehicles involved in the scrape, mark them as accident-related vehicle groups, and record the position coordinates and movement status of each vehicle 3 seconds before the accident, at the time of the accident, and 3 seconds after the accident.

[0051] Vehicle dwell time determination: Monitor the dwell time of each vehicle in the accident-related vehicle group at the accident scene. If a vehicle remains within 50 meters of the accident center point after the accident and the driver gets out of the vehicle to check or call the police, it is determined to be a non-hit-and-run vehicle. If a vehicle does not stop and accelerates away from the accident scene within 10 seconds after the accident, the speed limit can be dynamically adjusted according to the road section. For example, it can be adjusted to 5 seconds on highways. Or if the vehicle suddenly turns or accelerates away after a short dwell time of ≤3 seconds without turning on the hazard lights or placing warning signs, it is initially determined to be a hit-and-run vehicle.

[0052] Escape trajectory verification: Real-time vehicle trajectory tracking is performed on the initially identified escape vehicle. Vehicle driving conditions are set. When the set vehicle driving conditions are met, the escape behavior and escape vehicle are confirmed. The set vehicle driving conditions include the vehicle's departure direction being inconsistent with the normal driving direction before the accident, such as sudden lane change or U-turn to escape; the vehicle's departure speed being significantly higher than the driving speed before the accident, such as the speed before the accident being 60 km / h and the speed increasing to more than 90 km / h when leaving; and millimeter-wave radar detecting abnormal driving behaviors such as continuous lane changes and weaving during the vehicle's departure to avoid monitoring or hinder tracking.

[0053] Escape information solidification: After confirming the escape behavior and the escape vehicle, the complete feature information of the escape vehicle is extracted, including vehicle type, color, complete license plate number or partially clear license plate number, vehicle markings, initial escape direction, and precise location coordinates of the departure time. The feature information is then integrated with the collected road traffic condition information to obtain the escape vehicle feature information, including images, video clips and vehicle motion data of the traffic accident, which is transmitted in real time to the unmanned law enforcement vehicle and the regional traffic data processing center through an encrypted communication link.

[0054] Step S20: Distribute unmanned law enforcement vehicles to patrol dynamically along preset patrol routes on each road. Each unmanned law enforcement vehicle is equipped with a drone. When the unmanned law enforcement vehicle receives the characteristic information of the escaped vehicle, it enters emergency response mode and triggers the drone's automatic take-off procedure.

[0055] Specifically, the unmanned law enforcement vehicle in step S20 has autonomous navigation, environmental perception, and wireless communication functions. It conducts dynamic patrols on roads within its jurisdiction according to a preset patrol route. The unmanned law enforcement vehicle is equipped with a high-definition panoramic camera, lidar, GPS positioning unit, and emergency response processing unit. It can perceive the surrounding traffic environment in real time and receive information from road traffic monitoring equipment and regional traffic data processing center. When the unmanned law enforcement vehicle receives the escape vehicle characteristic information transmitted by the road traffic monitoring equipment or discovers the escape behavior of a traffic accident through environmental perception, it immediately enters the emergency response state. Each unmanned law enforcement vehicle is equipped with at least one drone. The drone is placed on a dedicated take-off and landing platform on the top of the unmanned law enforcement vehicle. It has vertical take-off and landing, long endurance, real-time transmission of high-definition images, autonomous obstacle avoidance, and remote control functions. The drone is equipped with a high-definition infrared camera, a high-definition visible light camera, a GPS / BeiDou dual-mode positioning unit, a data transmission radio, and a battery endurance unit.

[0056] Step S30: After the drone takes off, it locks onto the target vehicle based on the characteristic information of the fleeing vehicle provided by the unmanned law enforcement vehicle and monitoring equipment. During the tracking process, it captures images and dynamic videos of the fleeing vehicle in real time and transmits accurate information to the unmanned law enforcement vehicle and the regional traffic data processing center in real time.

[0057] Specifically, in step S30, when the unmanned law enforcement vehicle enters the emergency response state, it triggers the automatic take-off procedure of the drone. After take-off, the drone quickly locks onto the target vehicle based on the information of the fleeing vehicle provided by the road traffic monitoring equipment or the unmanned law enforcement vehicle itself. During the tracking process, the drone captures dynamic videos and images of the fleeing vehicle in real time, and obtains tracking information of the fleeing vehicle, including the precise location of the fleeing vehicle, its driving trajectory and surrounding road conditions, etc., and transmits it to the unmanned law enforcement vehicle and the regional traffic data processing center in real time through the wireless communication link. At the same time, the drone can autonomously adjust its flight trajectory and altitude according to the driving status of the fleeing vehicle to maintain continuous tracking of the fleeing vehicle. When encountering obstacles or complex terrain, it can autonomously plan a detour route.

[0058] Step S40: The regional traffic data processing center receives and integrates all information from monitoring equipment, unmanned law enforcement vehicles, and drones, and generates a real-time dynamic trajectory map of the escaped vehicle. The command center formulates a deployment plan based on the map, combined with the traffic network layout and police force distribution in the jurisdiction, and implements precise deployment.

[0059] Specifically, in step S40, the regional traffic data processing center receives and integrates all data from road traffic monitoring equipment, unmanned enforcement vehicles, and drones. It performs fusion analysis and processing on the data to generate a real-time dynamic trajectory map of the fleeing vehicle. The public security traffic command center is connected to the regional traffic data processing center in real time, enabling it to simultaneously acquire the fleeing vehicle's characteristic information and tracking information, including the vehicle's precise location, driving trajectory, and surrounding road conditions. Based on the fleeing vehicle's characteristic information and tracking information, and combined with the traffic network layout and police force distribution within its jurisdiction, the command center formulates a deployment plan. It sends instructions via wireless communication links to surrounding traffic police officers, other unmanned enforcement vehicles, and traffic signal control systems along the road to implement precise deployment. For example, the command center can instruct the traffic lights at the intersection ahead to adjust their timing to guide the fleeing vehicle into a pre-set interception area. The system commands nearby unmanned enforcement vehicles to encircle the fleeing vehicle along its predicted route; it also notifies nearby traffic police officers to intercept the vehicle at a designated location. A real-time, high-speed, and stable information exchange channel is established between road traffic monitoring equipment, unmanned enforcement vehicles, drones, the regional traffic data processing center, and the public security traffic command center via a 5G / 6G wireless communication network and a dedicated wireless data transmission link. Information transmission is encrypted to ensure data security and confidentiality. Each device possesses collaborative control logic; for example, when a drone's battery level falls below a preset threshold, the unmanned enforcement vehicle can adjust its route based on the drone's location information, approaching the drone to recharge or replace it with a backup drone. When fixed monitoring equipment detects a fleeing vehicle entering a new monitoring area, it automatically transfers monitoring authority and related information for that area to the corresponding unmanned enforcement vehicle and drone, achieving seamless integration of tracking tasks.

[0060] In addition, such as Figure 2 As shown, in one embodiment of the present invention, a traffic accident hit-and-run tracking system based on unmanned vehicles and air-ground integration is proposed. The system includes:

[0061] Road traffic monitoring module: Used to install road traffic monitoring equipment at various road nodes to monitor road traffic conditions in real time, and send road traffic conditions information and escape vehicle feature information obtained through image recognition and intelligent analysis to unmanned law enforcement vehicles and regional traffic data processing centers;

[0062] Unmanned law enforcement vehicle dynamic patrol module: used to distribute unmanned law enforcement vehicles on various roads to patrol dynamically according to preset patrol routes. Each unmanned law enforcement vehicle is equipped with a drone. When the unmanned law enforcement vehicle receives the characteristic information of the escaped vehicle, it enters the emergency response state and triggers the drone's automatic take-off procedure.

[0063] Drone aerial tracking module for escaping vehicles: After the drone takes off, it locks onto the target vehicle based on the characteristic information of the escaping vehicle provided by the unmanned law enforcement vehicle and monitoring equipment. During the tracking process, it captures images and dynamic videos of the escaping vehicle in real time and transmits accurate information to the unmanned law enforcement vehicle and the regional traffic data processing center in real time.

[0064] The regional traffic data processing center and the public security traffic command center linkage deployment module is used by the regional traffic data processing center to receive and integrate all information from monitoring equipment, unmanned law enforcement vehicles and drones, generate real-time dynamic trajectory maps of escaped vehicles, and formulate deployment plans based on the maps in combination with the traffic network layout and police force distribution in the jurisdiction to implement precise deployment.

[0065] The traffic accident escape tracking system based on unmanned vehicles and air-ground integration provided in this application adopts the traffic accident escape tracking method based on unmanned vehicles and air-ground integration in the above embodiments. It can solve the technical problems of existing road traffic accident escape vehicle tracking technologies, such as many monitoring blind spots, slow response speed, poor tracking effect, untimely information interaction, and low capture efficiency. Compared with the prior art, the beneficial effects of the traffic accident escape tracking system based on unmanned vehicles and air-ground integration provided in this application are the same as those of the traffic accident escape tracking method based on unmanned vehicles and air-ground integration provided in the above embodiments. Moreover, other technical features of the traffic accident escape tracking system based on unmanned vehicles and air-ground integration are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0066] This application provides a traffic accident escape tracking device based on unmanned vehicles and air-ground integration. The traffic accident escape tracking device based on unmanned vehicles and air-ground integration includes: at least one processor; and a memory communicatively connected to at least one processor; wherein the memory stores instructions that can be executed by at least one processor, and the instructions are executed by at least one processor to enable at least one processor to execute the traffic accident escape tracking method based on unmanned vehicles and air-ground integration in the above embodiment 1.

[0067] like Figure 3As shown in the illustration, in one embodiment of the present invention, a structural schematic diagram of a traffic accident escape tracking device based on unmanned vehicles and air-to-ground integration, suitable for implementing the embodiments of this application, is presented. The traffic accident escape tracking device based on unmanned vehicles and air-to-ground integration in the embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), etc., as well as fixed terminals such as digital TVs, desktop computers, etc. Figure 3 The traffic accident escape tracking device based on unmanned vehicles and air-ground integration shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0068] Figure 3 The illustrated unmanned vehicle-integrated air-to-ground traffic accident hit-and-run tracking device may include a processor 1001 (e.g., a central processing unit, graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 into a machine-readable storage medium (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the unmanned vehicle-integrated air-to-ground traffic accident hit-and-run tracking device. The processor 1001, the read-only memory 1002, and the machine-readable storage medium 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and a communication unit 1009. Communication unit 1009 allows the unmanned vehicle-based, air-to-ground integrated traffic accident escape tracking device to exchange data with other devices wirelessly or via wired communication. Although the figure shows an unmanned vehicle-based, air-to-ground integrated traffic accident escape tracking device with various systems, it should be understood that it is not required to implement or possess all the systems shown. More or fewer systems can be implemented alternatively.

[0069] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication unit, or installed from storage device 1003, or installed from read-only memory 1002. When the computer program is executed by processor 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0070] The traffic accident escape tracking device based on unmanned vehicles and air-ground integration provided in this application adopts the traffic accident escape tracking method based on unmanned vehicles and air-ground integration in the above embodiments. It can solve the technical problems of existing road traffic accident escape vehicle tracking technologies, such as many monitoring blind spots, slow response speed, poor tracking effect, untimely information interaction, and low capture efficiency. Compared with the prior art, the beneficial effects of the traffic accident escape tracking device based on unmanned vehicles and air-ground integration provided in this application are the same as the beneficial effects of the traffic accident escape tracking method based on unmanned vehicles and air-ground integration provided in the above embodiments. Moreover, other technical features of the traffic accident escape tracking device based on unmanned vehicles and air-ground integration are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0071] The various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0072] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the traffic accident escape tracking method based on unmanned vehicles and air-ground integration as described above.

[0073] The computer program product provided in this application can solve the technical problems of existing road traffic accident hit-and-run vehicle tracking technologies, such as numerous monitoring blind spots, slow response speed, poor tracking effect, untimely information interaction, and low capture efficiency. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the traffic accident hit-and-run tracking method based on unmanned vehicles and air-ground integration provided in the above embodiments, and will not be repeated here.

[0074] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for tracking hit-and-run accidents based on unmanned vehicles and air-ground integration, characterized in that, The method includes the following steps: Step S10: Install road traffic monitoring equipment at each road node to monitor road traffic conditions in real time, and send the road traffic conditions information and the escape vehicle feature information obtained through image recognition and intelligent analysis to the unmanned law enforcement vehicle and the regional traffic data processing center. Step S20: Distribute unmanned law enforcement vehicles to patrol dynamically along preset patrol routes on various roads. Each unmanned law enforcement vehicle is equipped with a drone. When the unmanned law enforcement vehicle receives the characteristic information of the escaped vehicle, it enters the emergency response state and triggers the drone's automatic take-off procedure. Step S30: After the drone takes off, it locks the target vehicle based on the characteristic information of the fleeing vehicle provided by the unmanned law enforcement vehicle and monitoring equipment. During the tracking process, it captures images and dynamic videos of the fleeing vehicle in real time and transmits accurate information to the unmanned law enforcement vehicle and the regional traffic data processing center in real time. Step S40: The regional traffic data processing center receives and integrates all information from monitoring equipment, unmanned law enforcement vehicles and drones, generates a real-time dynamic trajectory map of the escaped vehicle and sends it to the public security traffic command center. The step S10, which involves obtaining the escape vehicle's characteristic information through image recognition and intelligent analysis, includes: Image data preprocessing: Denoising, image enhancement, and region of interest (ROI) extraction are performed on the acquired road traffic images. The vehicle speed and trajectory acquired by the millimeter-wave radar are spatiotemporally calibrated with the road traffic images to correct data misalignment caused by equipment installation angle deviation or signal delay. Vehicle and road feature extraction: Based on a deep learning model, vehicle target detection is performed on the preprocessed road traffic images, and key features of each vehicle are extracted, including appearance features and motion features; Traffic accident determination: Based on the extracted vehicle and road features, traffic accidents are identified by combining abnormal motion state determination with abnormal road surface condition determination. Abnormal motion state determination includes vehicle speed changes and vehicle collision features. The abnormal road surface condition determination is achieved by setting road conditions. When the set road conditions occur, they are correlated with the abnormal motion state determination to confirm the occurrence of a traffic accident. Confirmation of vehicle escape behavior: After identifying a traffic accident, further dynamic tracking and analysis are used to confirm whether there is any vehicle escape behavior; The steps for confirming the vehicle's escape behavior include: Accident-related vehicle locking: Based on the images and vehicle movement trajectories at the moment of the accident, vehicles directly related to the accident are locked, marked as accident-related vehicle groups, and the position coordinates and movement status of each vehicle are recorded 3 seconds before the accident, at the time of the accident, and 3 seconds after the accident. Vehicle dwell time determination: Monitor the dwell time of each vehicle in the accident-related vehicle group at the accident scene. If a vehicle remains within 50 meters of the accident center point after the accident and the driver gets out of the vehicle to check or call the police, it is determined to be a non-hit-and-run vehicle. If a vehicle does not stop and accelerates away from the accident scene within 10 seconds after the accident, or stops briefly for ≤3 seconds and then turns or accelerates away without turning on hazard lights or placing warning signs, it is preliminarily determined to be a hit-and-run vehicle. Escape trajectory verification: Real-time vehicle trajectory tracking is performed on the initially identified escape vehicles. Vehicle driving conditions are set, and when the set vehicle driving conditions are met, the escape behavior and escape vehicle are confirmed. Escape information solidification: After confirming the escape behavior and the escape vehicle, the characteristic information of the escape vehicle is extracted and integrated with the collected road traffic condition information to obtain the escape vehicle characteristic information, which is transmitted in real time to the unmanned law enforcement vehicle and the regional traffic data processing center through an encrypted communication link. The unmanned law enforcement vehicle in step S20 has autonomous navigation, environmental perception and wireless communication functions. It conducts dynamic patrols on the roads in the jurisdiction according to the preset patrol route. The unmanned law enforcement vehicle is equipped with a panoramic camera, lidar, GPS positioning unit and emergency response processing unit. When the unmanned law enforcement vehicle receives the escape vehicle characteristic information transmitted by the road traffic monitoring equipment or discovers the escape behavior of traffic accident through environmental perception, it immediately enters the emergency response state. Each unmanned law enforcement vehicle is equipped with at least one drone. The drone is parked on a dedicated take-off and landing platform on the top of the unmanned law enforcement vehicle. The drone is equipped with an infrared camera, a visible light camera, a GPS / BeiDou dual-mode positioning unit, a data transmission radio and a battery endurance unit. In step S30, when the unmanned law enforcement vehicle enters the emergency response state, it triggers the automatic take-off procedure of the drone. After take-off, the drone locks onto the target vehicle based on the information of the fleeing vehicle provided by the road traffic monitoring equipment or the unmanned law enforcement vehicle itself. During the tracking process, the drone captures dynamic videos and images of the fleeing vehicle in real time, obtains tracking information of the fleeing vehicle, including the precise location, driving trajectory and surrounding road conditions of the fleeing vehicle, and transmits it to the unmanned law enforcement vehicle and the regional traffic data processing center in real time through the wireless communication link. At the same time, the drone autonomously adjusts its flight trajectory and altitude according to the driving status of the fleeing vehicle to maintain continuous tracking of the fleeing vehicle. When encountering obstacles or complex terrain, it autonomously plans a detour route.

2. The method for tracking hit-and-run accidents based on unmanned vehicles and air-ground integration as described in claim 1, characterized in that, The road traffic monitoring equipment installed at each road node in step S10 includes cameras, millimeter-wave radar, and infrared sensors, which monitor road traffic conditions in real time, including road traffic images, vehicle speed, vehicle trajectory, and traffic flow information.

3. The method for tracking hit-and-run accidents based on unmanned vehicles and air-ground integration as described in claim 1, characterized in that, In step S40, the regional traffic data processing center receives and integrates all data from road traffic monitoring equipment, unmanned enforcement vehicles, and drones. It performs fusion analysis and processing on the data to generate a real-time dynamic trajectory map of the escaped vehicle. The public security traffic command center is connected to the regional traffic data processing center in real time to synchronously obtain the escaped vehicle's characteristic information and tracking information, including the escaped vehicle's precise location, driving trajectory, and surrounding road conditions. The road traffic monitoring equipment, unmanned enforcement vehicles, drones, regional traffic data processing center, and public security traffic command center establish a real-time information interaction channel through a 5G / 6G wireless communication network and a dedicated wireless data transmission link. Information transmission is encrypted, and each device has collaborative control logic.

4. A traffic accident hit-and-run tracking system based on unmanned vehicles and air-ground integration, characterized in that: The system executes the traffic accident escape tracking method based on unmanned vehicles and air-ground integration as described in claim 1, including: Road traffic monitoring module: Used to install road traffic monitoring equipment at various road nodes to monitor road traffic conditions in real time, and send road traffic conditions information and escape vehicle feature information obtained through image recognition and intelligent analysis to unmanned law enforcement vehicles and regional traffic data processing centers; Unmanned law enforcement vehicle dynamic patrol module: used to distribute unmanned law enforcement vehicles on various roads to patrol dynamically according to preset patrol routes. Each unmanned law enforcement vehicle is equipped with a drone. When the unmanned law enforcement vehicle receives the characteristic information of the escaped vehicle, it enters the emergency response state and triggers the drone's automatic take-off procedure. Drone aerial tracking module for escaping vehicles: After the drone takes off, it locks onto the target vehicle based on the characteristic information of the escaping vehicle provided by the unmanned law enforcement vehicle and monitoring equipment. During the tracking process, it captures images and dynamic videos of the escaping vehicle in real time and transmits accurate information to the unmanned law enforcement vehicle and the regional traffic data processing center in real time. The regional traffic data processing center and the public security traffic command center linkage deployment module is used by the regional traffic data processing center to receive and integrate all information from monitoring equipment, unmanned law enforcement vehicles and drones, and generate real-time dynamic trajectory maps of escaped vehicles. The steps in the road traffic monitoring module to obtain the characteristic information of escaped vehicles through image recognition and intelligent analysis include: Image data preprocessing: Denoising, image enhancement, and region of interest (ROI) extraction are performed on the acquired road traffic images. The vehicle speed and trajectory acquired by the millimeter-wave radar are spatiotemporally calibrated with the road traffic images to correct data misalignment caused by equipment installation angle deviation or signal delay. Vehicle and road feature extraction: Based on a deep learning model, vehicle target detection is performed on the preprocessed road traffic images, and key features of each vehicle are extracted, including appearance features and motion features; Traffic accident determination: Based on the extracted vehicle and road features, traffic accidents are identified by combining abnormal motion state determination with abnormal road surface condition determination. Abnormal motion state determination includes vehicle speed changes and vehicle collision features. The abnormal road surface condition determination is achieved by setting road conditions. When the set road conditions occur, they are correlated with the abnormal motion state determination to confirm the occurrence of a traffic accident. Confirmation of vehicle escape behavior: After identifying a traffic accident, further dynamic tracking and analysis are used to confirm whether there is any vehicle escape behavior; The steps for confirming the vehicle's escape behavior include: Accident-related vehicle locking: Based on the images and vehicle movement trajectories at the moment of the accident, vehicles directly related to the accident are locked, marked as accident-related vehicle groups, and the position coordinates and movement status of each vehicle are recorded 3 seconds before the accident, at the time of the accident, and 3 seconds after the accident. Vehicle dwell time determination: Monitor the dwell time of each vehicle in the accident-related vehicle group at the accident scene. If a vehicle remains within 50 meters of the accident center point after the accident and the driver gets out of the vehicle to check or call the police, it is determined to be a non-hit-and-run vehicle. If a vehicle does not stop and accelerates away from the accident scene within 10 seconds after the accident, or stops briefly for ≤3 seconds and then turns or accelerates away without turning on hazard lights or placing warning signs, it is preliminarily determined to be a hit-and-run vehicle. Escape trajectory verification: Real-time vehicle trajectory tracking is performed on the initially identified escape vehicles. Vehicle driving conditions are set, and when the set vehicle driving conditions are met, the escape behavior and escape vehicle are confirmed. Escape information solidification: After confirming the escape behavior and the escape vehicle, the characteristic information of the escape vehicle is extracted and integrated with the collected road traffic condition information to obtain the escape vehicle characteristic information, which is transmitted in real time to the unmanned law enforcement vehicle and the regional traffic data processing center through an encrypted communication link. The unmanned law enforcement vehicle dynamic patrol module has autonomous navigation, environmental perception and wireless communication functions. It conducts dynamic patrols on the roads in the jurisdiction according to the preset patrol route. The unmanned law enforcement vehicle is equipped with a panoramic camera, LiDAR, GPS positioning unit and emergency response processing unit. When the unmanned law enforcement vehicle receives the characteristic information of the fleeing vehicle transmitted by the road traffic monitoring equipment or discovers the fleeing behavior of the traffic accident through environmental perception, it immediately enters the emergency response state. Each unmanned law enforcement vehicle is equipped with at least one drone. The drone is parked on a dedicated take-off and landing platform on the top of the unmanned law enforcement vehicle. The drone is equipped with an infrared camera, a visible light camera, a GPS / BeiDou dual-mode positioning unit, a data transmission radio and a battery endurance unit. In the aforementioned UAV aerial tracking module for escaped vehicles, when the unmanned enforcement vehicle enters emergency response mode, the UAV automatically takes off. After takeoff, the UAV locks onto the target vehicle based on escaped vehicle information provided by road traffic monitoring equipment or the unmanned enforcement vehicle itself. During the tracking process, the UAV captures dynamic videos and images of the escaped vehicle in real time, acquiring tracking information, including the vehicle's precise location, driving trajectory, and surrounding road conditions. This information is then transmitted in real time to the unmanned enforcement vehicle and the regional traffic data processing center via a wireless communication link. Simultaneously, the UAV autonomously adjusts its flight trajectory and altitude based on the escaped vehicle's driving status to maintain continuous tracking. When encountering obstacles or complex terrain, it autonomously plans a detour route.

5. A traffic accident escape tracking device based on unmanned vehicles and air-ground integration, characterized in that, include: The system includes a memory, a processor, and a traffic accident escape tracking program based on unmanned vehicles and air-ground integration, which is stored in the memory and can run on the processor. When the unmanned vehicle and air-ground integration traffic accident escape tracking program is executed by the processor, it implements the traffic accident escape tracking method based on unmanned vehicles and air-ground integration as described in any one of claims 1 to 3.

6. A computer program product, characterized in that, The computer program product includes a traffic accident escape tracking program based on unmanned vehicles and air-ground integration. When the unmanned vehicle and air-ground integration traffic accident escape tracking program is executed by the processor, it implements the traffic accident escape tracking method based on unmanned vehicles and air-ground integration as described in any one of claims 1 to 3.

Images