A Vehicle-Road Cooperative Information Interaction Method and System Based on Intelligent Transportation

Abstract

This invention discloses a vehicle-road cooperative information interaction method and system based on intelligent transportation, belonging to the field of vehicle-road cooperative technology. The method includes S1: multi-source data acquisition, S2: 3D environment modeling, S3: structured analysis of traffic conditions, S4: multi-source collaborative fusion, S5: vehicle-end collaboration, and S6: adaptive optimization configuration. This invention establishes a multi-source clock synchronization benchmark, accurately measures and compensates for network transmission delay, eliminates time deviations caused by link asymmetry, and achieves motion distortion correction of laser point clouds and spatiotemporal alignment of multi-sensor data. This effectively eliminates motion distortion of lidar in dynamic traffic scenarios, enabling high-density point clouds to accurately reflect the instantaneous state of the road, balancing perception accuracy and real-time processing efficiency, improving perception reliability under adverse weather conditions, and achieving lane-level precise cooperative control and high-precision vehicle positioning capabilities.

Description

Technical Field

[0001] This invention relates to the field of vehicle-road cooperative technology, and in particular to a vehicle-road cooperative information interaction method and system based on intelligent transportation. Background Technology

[0002] Intelligent transportation is a key development direction for improving road traffic efficiency, ensuring traffic safety, and realizing intelligent urban management. Vehicle-road cooperation, as its core supporting technology, aims to achieve global perception and collaborative control of traffic elements through real-time information interaction and intelligent decision-making between vehicles, roads, and the cloud. Efficient and accurate environmental perception is the foundation for the operation of vehicle-road cooperative systems, currently relying primarily on perception arrays composed of roadside cameras and radar equipment.

[0003] Currently, vehicle-road cooperative systems mainly rely on roadside perception solutions using cameras and millimeter-wave radar. While this solution can achieve basic target detection, it lacks reliability in adverse weather conditions and has limited perception accuracy, making it difficult to support advanced applications such as lane-level precise cooperative control and high-precision vehicle positioning.

[0004] To improve perception accuracy, lidar has been introduced to roadside environments. It can generate high-density 3D point clouds, which are the foundation for building accurate digital environment models. However, lidar faces a core challenge of motion distortion in dynamic traffic scenarios. The relative motion between the sensor and the target during scanning causes point cloud distortion, and the instantaneous environment model directly established cannot reflect the real road conditions. This makes it difficult to balance the accuracy of motion compensation with processing efficiency, which limits the effective application of lidar in vehicle-to-infrastructure (V2I) scenarios with stringent real-time requirements. Summary of the Invention

[0005] The purpose of this invention is to provide a vehicle-road cooperative information interaction method and system based on intelligent transportation, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a vehicle-road cooperative information interaction method based on intelligent transportation, comprising the following steps:

[0007] S1: Multi-source data acquisition, deploying a roadside perception array consisting of lidar, millimeter-wave radar and cameras to simultaneously collect traffic environment data streams;

[0008] S2: 3D environment modeling. Motion compensation and time axis correction of laser point cloud are performed through multi-source clock synchronization technology. High-precision clock protocol and spatiotemporal integration processing of multi-source data are used to obtain a 3D model of road environment.

[0009] S3: Structural analysis of traffic conditions. The fused 3D model is subjected to structured feature analysis, and parametric extraction and encoding transformation are used to obtain a standardized set of traffic condition information.

[0010] S4: Multi-source collaborative fusion, the vehicle-mounted communication terminal receives the standardized information set from the roadside, adopts cross-verification and complementary fusion of multi-source perception data, and constructs a dynamic environmental cognitive picture of all elements to support collaborative decision-making;

[0011] S5: Vehicle-to-vehicle collaboration. The vehicle system receives roadside information and fuses and verifies it with vehicle perception. Based on the fusion results, it performs motion planning and control, and at the same time feeds back the vehicle's status and intentions to the roadside system in real time, forming a closed-loop interaction.

[0012] S6: Adaptive optimization configuration, which continuously monitors the operating status indicators of each part of the system and uses a dynamic feedback adjustment mechanism for real-time optimization.

[0013] Preferably, step S2 uses multi-source clock synchronization technology to perform motion compensation and time axis correction on the laser point cloud, and employs high-precision clock protocol and multi-source data spatiotemporal integration processing to obtain a three-dimensional model of the road environment, including:

[0014] S21: Perform multi-source clock synchronization calibration to establish a nanosecond-level synchronized clock reference among roadside sensing nodes to ensure the consistency of data acquisition timestamps from each sensor.

[0015] S22: Perform point cloud motion compensation. Utilize motion parameters collected by the inertial measurement unit to perform six-degree-of-freedom motion distortion correction on the original point cloud of the lidar, eliminating point cloud deformation caused by carrier motion.

[0016] S23: Implement spatiotemporal alignment of multi-source data, align the corrected point cloud data with millimeter-wave radar points and video frames based on a unified clock reference, and achieve spatial unification of multi-sensor coordinate systems through a coordinate transformation matrix;

[0017] S24: Digitally sign the fused environment model through encryption encoding to form three-dimensional spatial data with anti-counterfeiting and integrity verification functions.

[0018] Preferably, step S21 performs multi-source clock synchronization calibration to establish a nanosecond-level synchronized clock reference among roadside sensing nodes, ensuring the consistency of data acquisition timestamps from each sensor.

[0019] S211: Master clock node election and configuration: elect a master clock node in the roadside sensing network and configure it as the time reference source for the entire network.

[0020] S212: Clock synchronization message exchange. The master clock node periodically broadcasts synchronization messages carrying precise timestamps to its subordinate slave nodes. The slave nodes record the arrival time of the messages and calculate the initial time deviation from the master clock.

[0021] S213: Transmission delay measurement and compensation, measuring the network transmission delay between master and slave nodes, and dynamically correcting the time deviation based on the link symmetry assumption;

[0022] S214: Clock frequency locking and phase adjustment, the slave node adjusts the frequency and phase of the local clock in real time based on the calculated time deviation and transmission delay to keep it synchronized with the master clock at the nanosecond level;

[0023] S215: Synchronization status monitoring and resynchronization triggering. It continuously monitors the synchronization error of each slave node. When the error exceeds the preset threshold or synchronization fails due to network jitter, it triggers the resynchronization process to restore the consistency of the clock reference.

[0024] Preferably, step 213 includes:

[0025] S2131: Hardware timestamp mark, which performs hardware stamping on the sending and receiving times of synchronization messages at the physical layer of master and slave nodes to eliminate time jitter caused by protocol stack software processing;

[0026] S2132: Bidirectional delay request-response interaction. The slave node sends a delay request message to the master node and records the sending time T3. After receiving the message, the master node marks the arrival time T4 and replies with a delay response message carrying T4. The slave node calculates the total round-trip delay based on this.

[0027] S2133: Link symmetry determination and compensation. The one-way transmission delay is calculated by taking half of the round-trip delay. At the same time, a link quality monitoring mechanism is introduced. If the uplink and downlink paths are detected to be asymmetrical, the asymmetric delay compensation algorithm is used to correct the time deviation.

[0028] S2134: Dynamic delay filtering process, which smooths the transmission delay values ​​measured multiple times in succession, eliminates outliers caused by instantaneous network jitter, and outputs a stable delay estimate for clock synchronization.

[0029] Preferably, the link symmetry determination and compensation calculates the one-way transmission delay by taking half of the round-trip delay, and simultaneously introduces a link quality monitoring mechanism. If uplink and downlink asymmetry is detected, an asymmetric delay compensation algorithm is used to correct the time deviation, including:

[0030] Real-time link quality monitoring involves continuously sending probe messages between master and slave nodes to monitor the signal strength, signal-to-noise ratio, bit error rate, and routing path changes of uplink and downlink, and to establish a dynamic database of link quality.

[0031] Asymmetric coefficient calibration: In the initial stage of deployment, the actual one-way delay between master and slave nodes is calibrated using a high-precision reference source, the asymmetric coefficient of uplink and downlink delay is calculated, and an asymmetric compensation lookup table is established.

[0032] Dynamic asymmetric detection compares the changing trend of round-trip delay in real time during normal operation. If the rate of change of uplink and downlink delay is inconsistent, dynamic asymmetry is determined to have occurred, and a compensation mechanism is triggered.

[0033] Adaptive compensation calculation: Based on the asymmetric coefficient lookup table and dynamic detection results, a weighted compensation algorithm is used to correct the round-trip delay by taking half the value, and the corrected one-way delay estimate is output.

[0034] The compensation effect is verified and feedback is provided by comparing the compensated clock synchronization error with a preset threshold. If the threshold is exceeded, the asymmetry coefficient is adjusted or the calibration process is retried to form a closed-loop optimization.

[0035] Preferably, during the real-time link quality monitoring process, the signal strength, signal-to-noise ratio, bit error rate, and routing path changes of the uplink and downlink are continuously collected, and each parameter is compared with a preset anomaly threshold.

[0036] If the signal strength is lower than the preset minimum signal strength threshold, it is determined that the current link has an excessive attenuation problem;

[0037] If the signal-to-noise ratio is lower than the preset minimum signal-to-noise ratio threshold, the current link is determined to be subject to strong noise interference.

[0038] If the bit error rate is higher than the preset maximum bit error rate threshold, the data transmission quality of the current link is determined to be degraded.

[0039] If the routing path changes or the number of hops changes significantly, it is determined that the network topology has changed.

[0040] Preferably, in the asymmetric coefficient calibration process, the actual one-way delay truth value of the uplink and downlink between master and slave nodes is obtained through a high-precision reference source, and the asymmetric coefficient is calculated accordingly. The calculated asymmetric coefficient is then compared with a preset calibration effective range.

[0041] If the asymmetric coefficient falls within the preset valid calibration range, the calibration result is deemed valid, and the coefficient is stored in the asymmetric compensation lookup table for subsequent use.

[0042] If the asymmetry coefficient exceeds the preset calibration range, it is determined that the asymmetry of the current link exceeds the system's compensability. In this case, the calibration result is not adopted, and a link topology check or hardware troubleshooting process is triggered.

[0043] Preferably, during normal system operation, the ratio of the uplink delay change rate to the downlink delay change rate in multiple consecutive round-trip delay measurements is calculated in real time, and this ratio is compared with a preset dynamic asymmetric detection threshold.

[0044] If the ratio of the uplink latency change rate to the downlink latency change rate and its reciprocal are both within the preset threshold range, it is determined that the uplink and downlink latency change trends are basically consistent, and the current link still maintains a symmetrical state.

[0045] If the ratio or its reciprocal exceeds the preset threshold range, it is determined that there is a significant asymmetric change in the uplink and downlink delays, and the dynamic asymmetric compensation mechanism is immediately triggered.

[0046] Preferably, the adaptive compensation calculation formula is:

[0047]

[0048] Where D1 represents the corrected one-way delay estimate, D2 represents the measured total round-trip delay, and α represents the uplink and downlink asymmetry coefficient, which is the ratio of the uplink one-way delay to the downlink one-way delay.

[0049] This invention also provides a vehicle-road cooperative information interaction system based on intelligent transportation, comprising:

[0050] The multi-source data acquisition module deploys a roadside sensing array consisting of lidar, millimeter-wave radar, and cameras to simultaneously collect traffic environment data streams.

[0051] The 3D environment modeling module uses multi-source clock synchronization technology to perform motion compensation and time axis correction on laser point clouds. It adopts high-precision clock protocol and spatiotemporal integration processing of multi-source data to obtain a 3D model of the road environment.

[0052] The traffic situation structured analysis module performs structured feature analysis on the fused 3D model, and uses parametric extraction and encoding transformation to obtain a standardized traffic state information set;

[0053] The multi-source collaborative fusion module receives a standardized roadside information set from the vehicle-mounted communication terminal, and uses cross-validation and complementary fusion of multi-source perception data to construct a dynamic environmental cognitive picture that supports collaborative decision-making.

[0054] The vehicle-side collaboration module receives roadside information from the vehicle system and fuses and verifies it with the vehicle's perception. Based on the fusion results, it performs motion planning and control, and at the same time feeds back the vehicle's status and intentions to the roadside system in real time, forming a closed-loop interaction.

[0055] The adaptive optimization configuration module continuously monitors the operational status indicators of each part of the system and uses a dynamic feedback adjustment mechanism for real-time optimization.

[0056] The technical effects and advantages of this invention are as follows:

[0057] This invention deploys a multi-source sensing array on the roadside, consisting of lidar, millimeter-wave radar, and cameras. By establishing a multi-source clock synchronization reference, it accurately measures and compensates for network transmission delays, eliminating time deviations caused by link asymmetry. This achieves motion distortion correction of the lidar point cloud and spatiotemporal alignment of multi-sensor data. Based on this, the fused 3D environment model undergoes structured analysis and standardized encoding to form a dynamic cognitive picture supporting collaborative decision-making. Furthermore, it constructs a closed-loop interaction and adaptive optimization mechanism between the vehicle and the roadside, effectively eliminating motion distortion of lidar in dynamic traffic scenarios. This allows the high-density point cloud to accurately reflect the instantaneous state of the road, balancing perception accuracy and real-time processing efficiency. It also improves perception reliability under adverse weather conditions, achieving lane-level precise collaborative control and high-precision vehicle positioning capabilities. This solves the pain points of existing systems in complex traffic environments, such as insufficient perception accuracy and limited collaborative control capabilities. Attached Figure Description

[0058] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention, but do not constitute a limitation thereof. In the drawings:

[0059] Figure 1 This is a flowchart of the vehicle-road cooperative information interaction method based on intelligent transportation according to the present invention. Detailed Implementation

[0060] 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.

[0061] This invention provides, for example Figure 1 The vehicle-road cooperative information interaction method based on intelligent transportation, as shown, includes the following steps:

[0062] S1: Multi-source data acquisition, deploying a roadside perception array consisting of lidar, millimeter-wave radar and cameras to simultaneously collect traffic environment data streams;

[0063] S2: 3D environment modeling. Motion compensation and time axis correction of laser point cloud are performed through multi-source clock synchronization technology. High-precision clock protocol and spatiotemporal integration processing of multi-source data are used to obtain a 3D model of road environment.

[0064] S3: Structural analysis of traffic conditions. The fused 3D model is subjected to structured feature analysis, and parametric extraction and encoding transformation are used to obtain a standardized set of traffic condition information.

[0065] S4: Multi-source collaborative fusion, the vehicle-mounted communication terminal receives the standardized information set from the roadside, adopts cross-verification and complementary fusion of multi-source perception data, and constructs a dynamic environmental cognitive picture of all elements to support collaborative decision-making;

[0066] S5: Vehicle-to-vehicle collaboration. The vehicle system receives roadside information and fuses and verifies it with vehicle perception. Based on the fusion results, it performs motion planning and control, and at the same time feeds back the vehicle's status and intentions to the roadside system in real time, forming a closed-loop interaction.

[0067] S6: Adaptive optimization configuration, which continuously monitors the operating status indicators of each part of the system and uses a dynamic feedback adjustment mechanism for real-time optimization.

[0068] Step S2 uses multi-source clock synchronization technology to perform motion compensation and time axis correction on the laser point cloud. It employs a high-precision clock protocol and multi-source data spatiotemporal integration processing to obtain a three-dimensional model of the road environment, including:

[0069] S21: Perform multi-source clock synchronization calibration to establish a nanosecond-level synchronized clock reference among roadside sensing nodes to ensure the consistency of data acquisition timestamps from each sensor.

[0070] S22: Perform point cloud motion compensation. Utilize motion parameters collected by the inertial measurement unit to perform six-degree-of-freedom motion distortion correction on the original point cloud of the lidar, eliminating point cloud deformation caused by carrier motion.

[0071] S23: Implement spatiotemporal alignment of multi-source data, align the corrected point cloud data with millimeter-wave radar points and video frames based on a unified clock reference, and achieve spatial unification of multi-sensor coordinate systems through a coordinate transformation matrix;

[0072] S24: Digitally sign the fused environment model through encryption encoding to form three-dimensional spatial data with anti-counterfeiting and integrity verification functions.

[0073] Step S2 constructs a high-precision 3D model of the road environment through multi-source clock synchronization and spatiotemporal fusion. First, a nanosecond-level synchronized clock reference is established to unify the timestamps of each sensor. Then, an inertial measurement unit is used to correct motion distortion of the lidar point cloud, eliminating scanning deformation. Subsequently, the point cloud, radar traces, and video frames are spatiotemporally aligned based on the unified clock. Finally, digital signatures ensure model integrity and anti-counterfeiting. This process fundamentally solves the problems of time misalignment and motion distortion in multi-source data, ensuring the accurate correspondence and reliability of the fused data, and providing a reliable environmental perception foundation for vehicle-road cooperative decision-making.

[0074] Step S21 performs multi-source clock synchronization calibration to establish a nanosecond-level synchronized clock reference among roadside sensing nodes, ensuring the consistency of data acquisition timestamps from each sensor:

[0075] S211: Master clock node election and configuration: elect a master clock node in the roadside sensing network and configure it as the time reference source for the entire network.

[0076] S212: Clock synchronization message exchange. The master clock node periodically broadcasts synchronization messages carrying precise timestamps to its subordinate slave nodes. The slave nodes record the arrival time of the messages and calculate the initial time deviation from the master clock.

[0077] S213: Transmission delay measurement and compensation, measuring the network transmission delay between master and slave nodes, and dynamically correcting the time deviation based on the link symmetry assumption;

[0078] S214: Clock frequency locking and phase adjustment, the slave node adjusts the frequency and phase of the local clock in real time based on the calculated time deviation and transmission delay to keep it synchronized with the master clock at the nanosecond level;

[0079] S215: Synchronization status monitoring and resynchronization triggering. It continuously monitors the synchronization error of each slave node. When the error exceeds the preset threshold or synchronization fails due to network jitter, it triggers the resynchronization process to restore the consistency of the clock reference.

[0080] Step S21 establishes a nanosecond-level high-precision clock reference among roadside sensing nodes through a complete clock synchronization process. This process first determines a unified time reference source for the entire network through master clock node election and configuration; then, through clock synchronization message exchange, slave nodes acquire the initial time deviation from the master clock; based on this, transmission delay measurement and compensation are performed to accurately calculate network transmission delay and dynamically correct time deviations; next, clock frequency locking and phase adjustment ensure that the local clocks of slave nodes maintain nanosecond-level synchronization with the master clock; finally, synchronization status monitoring and resynchronization triggering continuously monitor synchronization errors and restore consistency in case of anomalies. Thus, through the master-slave architecture and dynamic correction mechanism, the impact of network transmission delay and clock drift on synchronization accuracy is effectively eliminated, ensuring strict uniformity of the timestamps of data acquisition from various sensors. This provides a reliable time foundation for subsequent spatiotemporal fusion of multi-source data, fundamentally guaranteeing the accuracy and real-time performance of multi-sensor data alignment in the vehicle-road cooperative system.

[0081] Step 213 includes:

[0082] S2131: Hardware timestamp mark, which performs hardware stamping on the sending and receiving times of synchronization messages at the physical layer of master and slave nodes to eliminate time jitter caused by protocol stack software processing;

[0083] S2132: Bidirectional delay request-response interaction. The slave node sends a delay request message to the master node and records the sending time T3. After receiving the message, the master node marks the arrival time T4 and replies with a delay response message carrying T4. The slave node calculates the total round-trip delay based on this.

[0084] S2133: Link symmetry determination and compensation. The one-way transmission delay is calculated by taking half of the round-trip delay. At the same time, a link quality monitoring mechanism is introduced. If the uplink and downlink paths are detected to be asymmetrical, the asymmetric delay compensation algorithm is used to correct the time deviation.

[0085] S2134: Dynamic delay filtering process, which smooths the transmission delay values ​​measured multiple times in succession, eliminates outliers caused by instantaneous network jitter, and outputs a stable delay estimate for clock synchronization.

[0086] Step S213 uses hardware timestamps to accurately record message transmission and reception times at the physical layer, eliminating software jitter. Then, through bidirectional delay request-response interaction, it accurately calculates the total round-trip delay. A link symmetry determination and compensation mechanism is introduced to correct time deviations for asymmetric paths. Finally, dynamic delay filtering smoothing removes instantaneous network jitter anomalies and outputs a stable delay estimate. Hardware timestamps ensure measurement accuracy, bidirectional interaction enables accurate delay acquisition, asymmetric compensation solves path asymmetry problems, and dynamic filtering suppresses network interference. These four elements work together to provide a high-precision, highly stable delay data foundation for clock synchronization, supporting nanosecond-level synchronization accuracy.

[0087] Link symmetry determination and compensation involves calculating the one-way transmission delay by halving the round-trip delay. A link quality monitoring mechanism is also introduced; if uplink and downlink asymmetry is detected, an asymmetric delay compensation algorithm is used to correct the time deviation, including:

[0088] Real-time link quality monitoring involves continuously sending probe messages between master and slave nodes to monitor the signal strength, signal-to-noise ratio, bit error rate, and routing path changes of uplink and downlink, and to establish a dynamic database of link quality.

[0089] Asymmetric coefficient calibration: In the initial stage of deployment, the actual one-way delay between master and slave nodes is calibrated using a high-precision reference source, the asymmetric coefficient of uplink and downlink delay is calculated, and an asymmetric compensation lookup table is established.

[0090] Dynamic asymmetric detection compares the changing trend of round-trip delay in real time during normal operation. If the rate of change of uplink and downlink delay is inconsistent, dynamic asymmetry is determined to have occurred, and a compensation mechanism is triggered.

[0091] Adaptive compensation calculation: Based on the asymmetric coefficient lookup table and dynamic detection results, a weighted compensation algorithm is used to correct the round-trip delay by taking half the value, and the corrected one-way delay estimate is output.

[0092] The compensation effect is verified and feedback is provided by comparing the compensated clock synchronization error with a preset threshold. If the threshold is exceeded, the asymmetry coefficient is adjusted or the calibration process is retried to form a closed-loop optimization.

[0093] This process effectively solves the path asymmetry problem that is common in real networks through end-to-end closed-loop optimization, eliminates the synchronization error caused by the inconsistency between uplink and downlink delays, and ensures that high-precision single-way delay estimates can still be output in complex network environments, providing stable and reliable data support for clock synchronization.

[0094] During real-time link quality monitoring, signal strength, signal-to-noise ratio, bit error rate, and routing path changes of uplink and downlink are continuously collected, and each parameter is compared with preset anomaly thresholds:

[0095] If the signal strength is lower than the preset minimum signal strength threshold, it is determined that the current link has an excessive attenuation problem;

[0096] If the signal-to-noise ratio is lower than the preset minimum signal-to-noise ratio threshold, the current link is determined to be subject to strong noise interference.

[0097] If the bit error rate is higher than the preset maximum bit error rate threshold, the data transmission quality of the current link is determined to be degraded.

[0098] If the routing path changes or the number of hops changes significantly, it is determined that the network topology has changed.

[0099] By conducting multi-dimensional real-time monitoring and threshold comparison, link anomalies can be detected in a timely manner and unreliable data can be marked, thus avoiding the impact of link quality degradation on clock synchronization accuracy and providing a data foundation with controllable quality for subsequent delay measurements.

[0100] During the asymmetric coefficient calibration process, the actual one-way delay of the uplink and downlink between master and slave nodes is obtained through a high-precision reference source, and the asymmetric coefficient is calculated accordingly. The calculated asymmetric coefficient is then compared with the preset calibration effective range.

[0101] If the asymmetric coefficient falls within the preset valid calibration range, the calibration result is deemed valid, and the coefficient is stored in the asymmetric compensation lookup table for subsequent use.

[0102] If the asymmetry coefficient exceeds the preset calibration range, it is determined that the asymmetry of the current link exceeds the system's compensability. In this case, the calibration result is not adopted, and a link topology check or hardware troubleshooting process is triggered.

[0103] By setting a validity threshold, we ensure that only asymmetric coefficients within the compensable range are used for subsequent compensation, thus preventing the compensation algorithm from failing due to extreme link asymmetry. At the same time, we can promptly trigger hardware checks to ensure the reliability of the synchronization system from the root.

[0104] During normal system operation, the ratio of the uplink delay change rate to the downlink delay change rate in multiple consecutive round-trip delay measurements is calculated in real time, and this ratio is compared with a preset dynamic asymmetric detection threshold.

[0105] If the ratio of the uplink latency change rate to the downlink latency change rate and its reciprocal are both within the preset threshold range, it is determined that the uplink and downlink latency change trends are basically consistent, and the current link still maintains a symmetrical state.

[0106] If the ratio or its reciprocal exceeds the preset threshold range, it is determined that there is a significant asymmetric change in the uplink and downlink delays, and the dynamic asymmetric compensation mechanism is immediately triggered.

[0107] By monitoring the ratio of latency change rate in real time, we can keenly detect the occurrence of dynamic asymmetry in the link and trigger compensation as soon as asymmetry occurs. This avoids the accumulation of synchronization errors caused by network congestion or routing changes and ensures that clock synchronization accuracy is always maintained at the nanosecond level in dynamic network environments.

[0108] The adaptive compensation calculation formula is as follows:

[0109]

[0110] Where D1 represents the corrected one-way delay estimate, D2 represents the measured total round-trip delay, and α represents the uplink and downlink asymmetry coefficient, which is the ratio of the uplink one-way delay to the downlink one-way delay.

[0111] Step S3, the structured analysis of traffic conditions, includes the following steps:

[0112] S31: Deconstruction and extraction of key target elements. Multi-scale scanning and pattern recognition are performed on the 3D model to separate and extract core parameters such as 3D bounding boxes, position coordinates, velocity vectors and category labels of key traffic participants and static facilities, including vehicles, pedestrians, traffic lights, traffic signs, etc. This transforms the continuous physical world into discrete, parameterized digital objects, laying a standardized data foundation for subsequent semantic understanding and information interaction, and greatly improving the system's efficiency and accuracy in analyzing complex traffic scenarios.

[0113] S32: Scene Relationship and Topology Construction. Based on the extracted target elements, graph theory models or spatiotemporal relationship networks are used to establish the relative positions, kinematic relationships (such as following and changing lanes), and right-of-way logical relationships (such as lane ownership and stop line priority) between targets. This elevates the scattered target information into a scene knowledge graph with contextual logic, providing crucial structured support for predicting the intentions of traffic participants and detecting conflicts.

[0114] S33: State information encoding and standardized encapsulation. The target parameters and the constructed topological relationships are digitally encoded according to predefined communication protocols (such as V2X message set standards) and encapsulated into lightweight, low-latency standard information sets (such as data packets containing fields such as position, speed, heading, acceleration, object type, confidence level, and timestamp). This achieves lossless conversion from complex internal representations to standardized external interfaces, ensuring efficient and unambiguous transmission and parsing of information between vehicles and roads. It is a key link in opening up roadside perception and vehicle-side applications.

[0115] Step S4, multi-source collaborative fusion, includes the following steps:

[0116] S41: Spatiotemporal alignment and confidence weighting. The vehicle terminal first strictly aligns the received roadside standardized information set with the vehicle's high-precision spatiotemporal reference. Then, based on the inherent characteristics of the roadside and vehicle sensors (such as cameras and millimeter-wave radar), the current signal-to-noise ratio, and historical reliability, it dynamically assigns confidence weights to the same target information from different sources. This solves the data misalignment problem caused by differences in observation angle, timestamp, and sensor accuracy. Furthermore, intelligent weighting is used to initially screen out more reliable information sources, laying the foundation for high-quality fusion.

[0117] S42: Target-level and feature-level complementary fusion. Based on spatiotemporal alignment, for the same associated target, Kalman filtering or Bayesian inference methods are used to fuse its state estimation. For targets missed or obstructed by vehicle-mounted sensors, reliable perception results provided by the roadside are directly introduced to extend the perception range. This not only improves the estimation accuracy of existing target states (such as position and speed), but more importantly, it uses roadside sensing to make up for the inherent blind spots and obstruction defects of vehicle-mounted sensors, and constructs a perception field beyond line of sight and without blind spots.

[0118] S43: Full-element dynamic environmental cognition scene generation. It integrates all target information after fusion with the static elements of high-precision map to generate a four-dimensional (three-dimensional space + time) environmental cognition scene under a unified coordinate system, containing all dynamic and static elements and with time attributes. It provides a real-time, accurate and complete model for vehicle decision control system. It is the ultimate information foundation for safe, efficient and collaborative driving decisions, and fundamentally improves the vehicle's environmental understanding and risk prediction capabilities.

[0119] Step S5 vehicle-side collaboration includes the following steps:

[0120] S51: Multi-source perception fusion verification and decision planning. Based on the generated full-element dynamic cognitive picture, the vehicle performs threat assessment and trajectory prediction, and combined with the vehicle's task objectives (such as navigation path), it plans a safe and smooth future motion trajectory while taking into account traffic rules and comfort. The planned trajectory has higher safety redundancy, traffic efficiency and predictability, which significantly improves the level of intelligence and reliability of driving.

[0121] S52: Vehicle Status and Intent Encoding Feedback. The vehicle encodes its precise real-time status (such as position, speed, and control mode) and short-term future movement intentions (such as planned lane change and deceleration) according to standard protocol and actively broadcasts them to roadside units and other traffic participants via V2X communication. This transforms the vehicle from a passive information receiver to an active information contributor, forming a closed loop from roadside perception to vehicle-side decision-making and then to vehicle-side feedback. This enables the roadside system to dynamically optimize its perception focus area and information dissemination strategy based on the vehicle's true intentions, improving the overall system's collaborative efficiency and adaptability.

[0122] S53: Dynamic optimization of roadside strategies based on feedback. After receiving vehicle intent feedback, the roadside system can dynamically adjust its computing resource allocation, such as scanning the future path area of ​​relevant vehicles at a higher frequency or tracking it with higher precision. The advantages are: it enables on-demand allocation and precise deployment of system resources, avoids waste of computing and communication resources, and gives the entire collaborative perception network an attention mechanism, prioritizing the perception service quality of key areas and key targets.

[0123] Step S6, adaptive optimization configuration, includes the following steps:

[0124] S61: Real-time monitoring of multi-dimensional performance indicators, continuously collecting and analyzing key performance indicators of each part of the system to provide comprehensive and real-time diagnosis, providing accurate data input for dynamic optimization;

[0125] S62: Feedback-based bottleneck identification and root cause analysis utilizes monitoring data to identify bottlenecks affecting the overall system performance (such as performance degradation of specific sensors or unstable communication links), and analyzes their potential causes, thereby quickly locating the root cause of the problem and providing a basis for decision-making to achieve the optimal global configuration at the system level.

[0126] S63: Dynamic configuration strategy generation and execution. Based on root cause analysis results, or with the assistance of generating dynamic optimization strategies (such as adjusting sensor sampling frequency, switching communication channels, reallocating computing tasks, triggering device calibration, etc.), and issuing configuration instructions to the corresponding nodes for execution. This enables the system to maintain optimal or suboptimal performance even when external environmental changes (such as sudden changes in weather or traffic flow) and internal state fluctuations occur, greatly improving the system's robustness, reliability, and long-term service performance, and reducing operation and maintenance costs.

[0127] This invention also provides a vehicle-road cooperative information interaction system based on intelligent transportation, comprising:

[0128] The multi-source data acquisition module deploys a roadside sensing array consisting of lidar, millimeter-wave radar, and cameras to simultaneously collect traffic environment data streams.

[0129] The 3D environment modeling module uses multi-source clock synchronization technology to perform motion compensation and time axis correction on laser point clouds. It adopts high-precision clock protocol and spatiotemporal integration processing of multi-source data to obtain a 3D model of the road environment.

[0130] The traffic situation structured analysis module performs structured feature analysis on the fused 3D model, and uses parametric extraction and encoding transformation to obtain a standardized traffic state information set;

[0131] The multi-source collaborative fusion module receives a standardized roadside information set from the vehicle-mounted communication terminal, and uses cross-validation and complementary fusion of multi-source perception data to construct a dynamic environmental cognitive picture that supports collaborative decision-making.

[0132] The vehicle-side collaboration module receives roadside information from the vehicle system and fuses and verifies it with the vehicle's perception. Based on the fusion results, it performs motion planning and control, and at the same time feeds back the vehicle's status and intentions to the roadside system in real time, forming a closed-loop interaction.

[0133] The adaptive optimization configuration module continuously monitors the operational status indicators of each part of the system and uses a dynamic feedback adjustment mechanism for real-time optimization.

[0134] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A vehicle-road cooperative information interaction method based on intelligent transportation, characterized in that, include: S1: Multi-source data acquisition, deploying a roadside perception array consisting of lidar, millimeter-wave radar and cameras to simultaneously collect traffic environment data streams; S2: 3D environment modeling. Motion compensation and time axis correction of laser point cloud are performed through multi-source clock synchronization technology. High-precision clock protocol and spatiotemporal integration processing of multi-source data are used to obtain a 3D model of road environment. S3: Structural analysis of traffic conditions. The fused 3D model is subjected to structured feature analysis, and parametric extraction and encoding transformation are used to obtain a standardized set of traffic condition information. S4: Multi-source collaborative fusion, the vehicle-mounted communication terminal receives the standardized information set from the roadside, adopts cross-verification and complementary fusion of multi-source perception data, and constructs a dynamic environmental cognitive picture of all elements to support collaborative decision-making; S5: Vehicle-to-vehicle collaboration. The vehicle system receives roadside information and fuses and verifies it with vehicle perception. Based on the fusion results, it performs motion planning and control, and at the same time feeds back the vehicle's status and intentions to the roadside system in real time, forming a closed-loop interaction. S6: Adaptive optimization configuration, which continuously monitors the operating status indicators of each part of the system and uses a dynamic feedback adjustment mechanism for real-time optimization. 2.The car-road cooperation information interaction method based on intelligent transportation according to claim 1, wherein, Step S2 uses multi-source clock synchronization technology to perform motion compensation and time axis correction on the laser point cloud. It employs a high-precision clock protocol and multi-source data spatiotemporal integration processing to obtain a three-dimensional model of the road environment, including: S21: Perform multi-source clock synchronization calibration to establish a nanosecond-level synchronized clock reference among roadside sensing nodes to ensure the consistency of data acquisition timestamps from each sensor. S22: Perform point cloud motion compensation. Utilize motion parameters collected by the inertial measurement unit to perform six-degree-of-freedom motion distortion correction on the original point cloud of the lidar, eliminating point cloud deformation caused by carrier motion. S23: Implement spatiotemporal alignment of multi-source data, align the corrected point cloud data with millimeter-wave radar points and video frames based on a unified clock reference, and achieve spatial unification of multi-sensor coordinate systems through a coordinate transformation matrix; S24: Digitally sign the fused environment model through encryption encoding to form three-dimensional spatial data with anti-counterfeiting and integrity verification functions.

3. The vehicle-road cooperative information interaction method based on intelligent transportation according to claim 2, characterized in that, Step S21 performs multi-source clock synchronization calibration, establishing a nanosecond-level synchronized clock reference among roadside sensing nodes to ensure the consistency of data acquisition timestamps from each sensor: S211: Master clock node election and configuration: elect a master clock node in the roadside sensing network and configure it as the time reference source for the entire network. S212: Clock synchronization message exchange. The master clock node periodically broadcasts synchronization messages carrying precise timestamps to its subordinate slave nodes. The slave nodes record the arrival time of the messages and calculate the initial time deviation from the master clock. S213: Transmission delay measurement and compensation, measuring the network transmission delay between master and slave nodes, and dynamically correcting the time deviation based on the link symmetry assumption; S214: Clock frequency locking and phase adjustment, the slave node adjusts the frequency and phase of the local clock in real time based on the calculated time deviation and transmission delay to keep it synchronized with the master clock at the nanosecond level; S215: Synchronization status monitoring and resynchronization triggering. It continuously monitors the synchronization error of each slave node. When the error exceeds the preset threshold or synchronization fails due to network jitter, it triggers the resynchronization process to restore the consistency of the clock reference.

4. The vehicle-road cooperative information interaction method based on intelligent transportation according to claim 3, characterized in that, Step 213 includes: S2131: Hardware timestamp mark, which performs hardware stamping on the sending and receiving times of synchronization messages at the physical layer of master and slave nodes to eliminate time jitter caused by protocol stack software processing; S2132: Bidirectional delay request-response interaction. The slave node sends a delay request message to the master node and records the sending time T3. After receiving the message, the master node marks the arrival time T4 and replies with a delay response message carrying T4. The slave node calculates the total round-trip delay based on this. S2133: Link symmetry determination and compensation. The one-way transmission delay is calculated by taking half of the round-trip delay. At the same time, a link quality monitoring mechanism is introduced. If the uplink and downlink paths are detected to be asymmetrical, the asymmetric delay compensation algorithm is used to correct the time deviation. S2134: Dynamic delay filtering process, which smooths the transmission delay values ​​measured multiple times in succession, eliminates outliers caused by instantaneous network jitter, and outputs a stable delay estimate for clock synchronization.

5. The vehicle-road cooperative information interaction method based on intelligent transportation according to claim 4, characterized in that, The link symmetry determination and compensation calculates the one-way transmission delay by taking half of the round-trip delay. Simultaneously, a link quality monitoring mechanism is introduced. If uplink and downlink asymmetry is detected, an asymmetric delay compensation algorithm is used to correct the time deviation, including: Real-time link quality monitoring involves continuously sending probe messages between master and slave nodes to monitor the signal strength, signal-to-noise ratio, bit error rate, and routing path changes of uplink and downlink, and to establish a dynamic database of link quality. Asymmetric coefficient calibration: In the initial stage of deployment, the actual one-way delay between master and slave nodes is calibrated using a high-precision reference source, the asymmetric coefficient of uplink and downlink delay is calculated, and an asymmetric compensation lookup table is established. Dynamic asymmetric detection compares the changing trend of round-trip delay in real time during normal operation. If the rate of change of uplink and downlink delay is inconsistent, dynamic asymmetry is determined to have occurred, and a compensation mechanism is triggered. Adaptive compensation calculation: Based on the asymmetric coefficient lookup table and dynamic detection results, a weighted compensation algorithm is used to correct the round-trip delay by taking half the value, and the corrected one-way delay estimate is output. The compensation effect is verified and feedback is provided by comparing the compensated clock synchronization error with a preset threshold. If the threshold is exceeded, the asymmetry coefficient is adjusted or the calibration process is retried to form a closed-loop optimization. 6.The car-road cooperation information interaction method based on intelligent transportation according to claim 5, characterized in that, During the real-time monitoring of link quality, the signal strength, signal-to-noise ratio, bit error rate, and routing path changes of the uplink and downlink are continuously collected, and each parameter is compared with a preset anomaly threshold. If the signal strength is lower than the preset minimum signal strength threshold, it is determined that the current link has an excessive attenuation problem; If the signal-to-noise ratio is lower than the preset minimum signal-to-noise ratio threshold, the current link is determined to be subject to strong noise interference. If the bit error rate is higher than the preset maximum bit error rate threshold, the data transmission quality of the current link is determined to be degraded. If the routing path changes or the number of hops changes significantly, it is determined that the network topology has changed. 7.The vehicle-road cooperation information interaction method based on intelligent transportation according to claim 5, characterized in that, In the asymmetric coefficient calibration process, the actual one-way delay of the uplink and downlink between master and slave nodes is obtained through a high-precision reference source, and the asymmetric coefficient is calculated accordingly. The calculated asymmetric coefficient is then compared with a preset calibration effective range. If the asymmetric coefficient falls within the preset valid calibration range, the calibration result is deemed valid, and the coefficient is stored in the asymmetric compensation lookup table for subsequent use. If the asymmetry coefficient exceeds the preset calibration range, it is determined that the asymmetry of the current link exceeds the system's compensability. In this case, the calibration result is not adopted, and a link topology check or hardware troubleshooting process is triggered. 8.The vehicle-road cooperative information interaction method based on intelligent transportation according to claim 5, characterized in that, During normal system operation, the ratio of the uplink delay change rate to the downlink delay change rate in multiple consecutive round-trip delay measurements is calculated in real time, and this ratio is compared with a preset dynamic asymmetric detection threshold. If the ratio of the uplink latency change rate to the downlink latency change rate and its reciprocal are both within the preset threshold range, it is determined that the uplink and downlink latency change trends are basically consistent, and the current link still maintains a symmetrical state. If the ratio or its reciprocal exceeds the preset threshold range, it is determined that there is a significant asymmetric change in the uplink and downlink delays, and the dynamic asymmetric compensation mechanism is immediately triggered. 9.The vehicle-road cooperation information interaction method based on intelligent transportation according to claim 5, characterized in that, The adaptive compensation calculation formula is as follows: Where D1 represents the corrected one-way delay estimate, D2 represents the measured total round-trip delay, and α represents the uplink and downlink asymmetry coefficient, which is the ratio of the uplink one-way delay to the downlink one-way delay.

10. A vehicle-road cooperative information interaction system based on intelligent transportation, applied to the vehicle-road cooperative information interaction method based on intelligent transportation in any one of claims 1-9, characterized in that, include: The multi-source data acquisition module deploys a roadside sensing array consisting of lidar, millimeter-wave radar, and cameras to simultaneously collect traffic environment data streams. The 3D environment modeling module uses multi-source clock synchronization technology to perform motion compensation and time axis correction on laser point clouds. It adopts high-precision clock protocol and spatiotemporal integration processing of multi-source data to obtain a 3D model of the road environment. The traffic situation structured analysis module performs structured feature analysis on the fused 3D model, and uses parametric extraction and encoding transformation to obtain a standardized traffic state information set; The multi-source collaborative fusion module receives a standardized roadside information set from the vehicle-mounted communication terminal, and uses cross-validation and complementary fusion of multi-source perception data to construct a dynamic environmental cognitive picture that supports collaborative decision-making. The vehicle-side collaboration module receives roadside information from the vehicle system and fuses and verifies it with the vehicle's perception. Based on the fusion results, it performs motion planning and control, and at the same time feeds back the vehicle's status and intentions to the roadside system in real time, forming a closed-loop interaction. The adaptive optimization configuration module continuously monitors the operational status indicators of each part of the system and uses a dynamic feedback adjustment mechanism for real-time optimization.

Images