Speed limit adjustment method and system based on beidou positioning for different road sections

Abstract

This invention relates to the field of intelligent transportation and assisted driving technology, and discloses a method and system for adjusting speed limits on different road segments based on BeiDou positioning. The method includes: acquiring real-time positioning and multi-dimensional motion vector data of each operating node using the BeiDou system; generating a road segment group behavior benchmark by combining a biomimetic swarm intelligence model; monitoring the motion vector deviation of each node relative to the benchmark, and identifying dynamic fluctuations in road segment capacity based on its clustering characteristics; and calculating and broadcasting adaptive speed limit suggestions in real time to achieve dynamic closed-loop adjustment of speed limits. The system consists of BeiDou positioning perception, swarm intelligence collaborative computing, capacity assessment, and speed limit decision distribution modules. By integrating BeiDou positioning and a biomimetic swarm intelligence model, this invention can keenly capture fluctuations in group behavior caused by sudden local anomalies, realizing a shift from centralized passive control to proactive self-organized control, and improving road network identification accuracy, response speed, and traffic efficiency.

Description

Technical Field

[0001] This invention belongs to the field of intelligent transportation and assisted driving technology, specifically relating to a method and system for adjusting speed limits on different road sections based on BeiDou positioning. Background Technology

[0002] With the rapid development of intelligent transportation systems and high-precision satellite positioning technology, road safety management and traffic efficiency optimization have become research hotspots in the transportation and logistics field. In modern transportation networks, vehicle speed control is not only a core means of preventing traffic accidents, but also a key link in ensuring the dynamic allocation of road resources. Its technological level affects the operational efficiency of the national economy and the safety of public life and property.

[0003] Combining the BeiDou Navigation Satellite System with geographic information technology for speed limit adjustment is a crucial technological approach for achieving precise traffic management. This technology acquires real-time vehicle spatial positioning information and matches it with corresponding road grades and traffic regulations, aiming to establish a standardized vehicle speed intervention mechanism to maintain road traffic order at a macro level.

[0004] Existing technologies still have shortcomings when dealing with complex and ever-changing road scenarios: speed limit adjustments are mainly based on preset static map data, lacking sensitivity to dynamic risks such as sudden local anomalies, slippery roads, or debris spills, and cannot be flexibly adjusted according to real-time conditions; traditional solutions rely excessively on centralized control logic, lacking collaborative perception capabilities among operational nodes, resulting in an excessively long response chain when facing sudden obstacles; data processing dimensions are relatively singular, failing to integrate features such as vehicle motion vector deviation and yaw rate, making it difficult to identify real-time fluctuations in road segment capacity through group behavior logic; the lack of biomimetic-level group obstacle avoidance and collaboration mechanisms makes individual vehicles make isolated decisions when responding to environmental changes, making it difficult to form an adaptive cluster control scheme. To address the shortcomings of traditional speed limit adjustment technologies, such as lack of sensitivity to dynamic risks, lack of collaborative perception capabilities among operational nodes, failure to integrate vehicle motion vector characteristics, and lack of group obstacle avoidance and collaboration mechanisms, this invention proposes a speed limit adjustment method and system based on BeiDou positioning for different road segments, which is particularly important. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for adjusting speed limits on different road sections based on BeiDou positioning, which can solve the problems mentioned in the background. Addressing the shortcomings of existing technologies where speed limit adjustments are mainly based on preset static map data, lacking sensitivity to dynamic risks such as sudden local anomalies, slippery roads, or debris spills, and the over-reliance on centralized control logic leading to excessively long system response chains and a lack of collaborative perception capabilities between operational nodes, this invention provides a dynamic speed limit coordination scheme that integrates the high-precision positioning of the BeiDou satellite navigation system with biomimetic swarm intelligence fishing logic.

[0006] To achieve the above objectives, this invention proposes a method for adjusting speed limits on different road sections based on BeiDou positioning, comprising the following steps: S1. Real-time three-dimensional positioning information and multi-dimensional motion vector data of multiple operation nodes within the target road section are obtained through the Beidou satellite navigation system; S2. Based on the real-time three-dimensional positioning information, construct a dynamic spatiotemporal topology mapping, and combine it with a biomimetic swarm intelligence model to collaboratively represent the multi-dimensional motion vector data, generating a road segment group behavior benchmark. S3. Monitor the motion vector deviation of each operation node relative to the road segment group behavior benchmark in real time, and identify the dynamic fluctuation state of the road segment's traffic capacity based on the clustering characteristics of the motion vector deviation. S4. Based on the dynamic fluctuation state, calculate the adaptive speed limit suggestion in real time, and distribute the adaptive speed limit suggestion to each of the operation nodes through the broadcast link to realize the dynamic closed-loop adjustment of the road section speed limit.

[0007] Preferably, step S2 specifically includes the following steps: S21. Perform spatiotemporal alignment processing on the acquired three-dimensional positioning information of multiple operation nodes to ensure that different nodes perform interactive calculations under a unified time scale and spatial coordinate system. S22. Introduce a group motion model from bionics, treat each of the work nodes as an individual in the group, and establish a dynamic neighborhood relationship based on the relative distance and relative speed between the work nodes. S23. Based on the dynamic neighborhood relationship, calculate the expected motion vector of the group in the local area during steady-state driving. The expected motion vector includes the group's average velocity direction, average speed, and preset driving trajectory envelope.

[0008] Preferably, step S3 specifically includes the following steps: S31. Calculate the instantaneous speed, instantaneous yaw rate, and Euclidean distance between the travel trajectory and the desired motion vector for each of the operation nodes to obtain the single-node motion deviation value; S32. Extract the single-node motion deviation values ​​of all the work nodes within the preset time period, and identify the set of nodes with abnormal avoidance actions or abnormal deceleration behavior through statistical distribution function. S33. Calculate the spatial distribution density and behavior duration of the node set in the target road segment. When the spatial distribution density exceeds a preset congestion threshold and the behavior duration exceeds a preset response period, determine that the traffic capacity of the target road segment has decreased.

[0009] Preferably, the multidimensional motion vector data includes at least longitude coordinates, latitude coordinates, altitude coordinates, instantaneous speed, driving azimuth angle, and yaw rate. The biomimetic swarm intelligence model adopts an improved Reynolds swarm model, which simulates the self-organized obstacle avoidance behavior of vehicles in complex road environments by setting separation force, alignment force, and cohesion force.

[0010] Preferably, the real-time three-dimensional positioning information is centimeter-level coordinate data obtained by processing carrier phase observations and pseudorange observations from the BeiDou satellite navigation system using a differential positioning algorithm.

[0011] Preferably, step S4 specifically includes the following steps: S41. Based on the severity of the decrease in traffic capacity as determined, retrieve the corresponding initial speed limit value from the preset speed limit gradient library; S42. Combining the slope, curvature of the curve, and historical meteorological parameters of the target road section, the initial speed limit value is weighted and corrected to generate a temporary adaptive speed limit value for the current abnormal operating conditions. S43. The temporary adaptive speed limit value is encapsulated into a standard command format and sent to the vehicle terminals of all work nodes within the influence range of the road section via a wireless communication network.

[0012] Preferably, the clustering feature identification of the motion vector deviation includes: using a density spatial clustering algorithm to spatially cluster the operation nodes that produce yaw or deceleration; if the cluster center is located within the driving lane and the size of the cluster shows an expanding trend, it is identified as the presence of a sudden local obstacle or abnormal road surface friction.

[0013] Preferably, the construction of the road segment group behavior benchmark further includes: establishing normalized group behavior fingerprints for different time periods based on historical traffic data, and comparing the real-time acquired motion vector data with the corresponding normalized group behavior fingerprints to eliminate the interference of periodic congestion on dynamic anomaly identification.

[0014] Preferably, the process of generating the adaptive speed limit suggestion also includes a feedback adjustment mechanism: after the system issues a temporary speed limit suggestion, it continuously monitors the convergence of the motion vectors of subsequent work nodes. If the motion deviation of subsequent nodes decreases and the group speed returns to a stable state, the speed limit value is gradually increased until it returns to the original static speed limit state.

[0015] A speed limit adjustment system based on BeiDou positioning for different road segments, used to implement the above-mentioned method, includes: a BeiDou positioning sensing module, used to collect high-precision spatial positions and kinematic parameters of all operation nodes within the target road segment in real time, and aggregate the data to a data preprocessing unit; a swarm intelligence collaborative computing module, electrically connected to the BeiDou positioning sensing module, configured to analyze the motion vector collaborative relationship between operation nodes using a biomimetic swarm intelligence model, construct a group behavior benchmark, and identify motion deviation clustering features; a traffic capacity assessment module, connected to the swarm intelligence collaborative computing module, used to assess the real-time traffic status of the road segment based on the statistical characteristics of motion deviation, and determine whether there are any sudden anomalies; and a speed limit decision distribution module, which generates adaptive speed limit instructions based on the traffic capacity assessment results, and uses a vehicle network communication link to realize the real-time distribution and execution feedback of the instructions.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention introduces a dynamic speed-limiting collaborative algorithm for biomimetic swarm fishing, which deeply integrates the high-precision three-dimensional positioning data of the BeiDou satellite navigation system with the swarm motion model in bionics, thereby changing the excessive reliance of traditional speed-limiting management on static maps and centralized control.

[0017] 2. By treating each work vehicle equipped with the BeiDou positioning system as an intelligent node with sensing and feedback capabilities, and analyzing the motion vector deviations of the group of nodes, the system can, like a flock of birds or a school of fish avoiding obstacles, keenly detect subtle abnormalities in group behavior caused by sudden fog, slippery road surfaces, or spillage on different surfaces within a road segment. This group-based collaborative perception model improves the system's accuracy and response speed in identifying sudden environmental risks, solving the pain point of existing technologies' slow response to unpredictable obstacles.

[0018] 3. This invention utilizes the high-precision spatiotemporal reference provided by the BeiDou Navigation Satellite System and combines it with biomimetic swarm collaboration logic to construct a speed limit adjustment mechanism possessing self-perception, dynamic decision-making, and swarm collaboration characteristics. This mechanism enhances the robustness of the traffic system to complex environmental disturbances and has practical application value and technological advancement in reducing traffic accident rates and improving the overall traffic efficiency of the road network.

[0019] 4. This invention integrates existing BeiDou navigation infrastructure, achieving a qualitative leap in traffic management quality at a relatively low marginal cost, resulting in both economic and social benefits. This solution not only addresses the persistent problem of current static speed limits being unable to cope with dynamic risks, but also lays a solid technical foundation for precise traffic flow control in the future era of fully autonomous driving. Attached Figure Description

[0020] Figure 1This is a schematic diagram of the overall technical solution architecture of the present invention; Figure 2 This is a schematic diagram of the core principle framework for constructing road segment group behavior benchmarks based on a biomimetic swarm intelligence model in this invention; Figure 3 This is a flowchart of the logical process for assessing road segment capacity based on motion vector deviation clustering in this invention. Figure 4 This is a flowchart of the adaptive speed limit suggestion generation logic based on multi-dimensional parameter weighted correction in this invention. Figure 5 This is a schematic diagram of the multi-level interaction relationship and data flow based on BeiDou high-precision positioning and vehicle network distribution in this invention. Detailed Implementation

[0021] To further illustrate the technical means and effects adopted by the present invention to achieve the intended purpose, the following description is provided in conjunction with the appendix. Figure 1 To be continued Figure 5 The following is a detailed description of the specific implementation methods, structures, features, and effects of the present invention, as well as preferred embodiments.

[0022] Example 1: This example provides a method for adjusting speed limits on different road sections based on BeiDou positioning. This method integrates the high-precision spatiotemporal reference of the BeiDou satellite navigation system with bionic swarm collaboration logic to achieve dynamic and precise control of vehicle speed in complex road environments.

[0023] In the initial stage of this embodiment, namely step S1, the system activates the BeiDou positioning and sensing module. This module integrates a multi-frequency receiver supporting the BeiDou-3 global satellite navigation system, capable of receiving carrier signals from satellites at multiple orbital altitudes. The received weak satellite signals are amplified with low noise, down-converted, and processed via analog-to-digital conversion through a radio frequency front-end, converting the original electromagnetic signals into baseband digital signals. During processing, the system acquires real-time three-dimensional positioning information for all work nodes within the target road segment. This real-time three-dimensional positioning information includes not only basic longitude and latitude coordinates but also elevation information, forming a complete spatial three-dimensional vector. The system collects multi-dimensional motion vector data. This data is output in real-time by the BeiDou receiver's dynamic calculation engine, specifically including the instantaneous velocity of the work nodes, driving azimuth angle, and yaw rate reflecting the severity of vehicle steering. To ensure extremely high data accuracy, the system uses real-time differential processing of carrier phase observations and pseudorange observations. By introducing correction information from the reference station, it compensates for ionospheric delay, tropospheric delay, and satellite ephemeris errors, obtaining centimeter-level coordinate positioning data. These data are updated at a preset sampling frequency of 20 Hz or higher, providing a high-frequency, high-precision raw sample data stream for subsequent group behavior analysis.

[0024] In step S2, the system performs in-depth spatial and logical reconstruction of the massive amount of collected data. Step S21, i.e., spatiotemporal alignment processing, is then executed. Because the BeiDou terminals at different work nodes may have slight internal clock deviations, or varying link delays in signal transmission to the cloud processing center, the system utilizes the nanosecond-level high-precision time synchronization function built into the BeiDou system to attach all collected positioning coordinates to a unified timestamp. By resampling and linear interpolating the time series, the system ensures that the spatial positional relationships of all nodes are comparable at the same instant. The system constructs a dynamic spatiotemporal topology mapping. This mapping process transforms the originally isolated vehicle coordinates into a topological network with interacting relationships.

[0025] In step S22, the system introduces a swarm motion model from bionics. Under this model, each task node is no longer an independent individual but is considered an intelligent particle within a bionic swarm. The system uses a dynamic search algorithm to define the neighborhood of each node in real time based on the relative distance and relative speed between task nodes. When the three-dimensional straight-line distance between two task nodes is lower than a preset interaction radius threshold, a dynamic neighborhood relationship is established between them. This relationship simulates the perception range of flocks of birds or schools of fish in nature.

[0026] In step S23, based on the established dynamic neighborhood relationships, the system combines a biomimetic swarm intelligence model to collaboratively represent multidimensional motion vector data and generate a road segment group behavior benchmark. The system calculates the expected motion vector of the group during steady-state driving within a local area. This expected motion vector is not a simple arithmetic average, but rather based on the Reynolds model logic in bionics. The system calculates three core driving forces: first, a separation force, which ensures a necessary safe distance between adjacent vehicles to prevent physical collisions; this manifests as a repulsive force that increases exponentially with decreasing distance. Second, an alignment force, which guides all vehicles within the neighborhood to tend towards similar speeds and directions, smoothing the velocity field. Third, a cohesion force, which drives vehicles deviating from the main flow towards the group center, maintaining the integrity of the traffic flow. Through the textual logical calculation of these three forces, the system derives the group's average speed direction, average speed, and preset trajectory envelope in the current road segment and time period. This trajectory envelope defines the allowable fluctuation range of vehicles in three-dimensional space under normal traffic conditions.

[0027] In step S3, the system monitors the motion vector deviation of each work node relative to the road segment group behavior benchmark in real time. During step S31, the system compares the current instantaneous speed, instantaneous yaw rate, and actual travel trajectory of each work node with the previously generated expected motion vector in real time. By calculating the square root of the sum of the squares of the three-dimensional coordinate differences, the Euclidean distance deviation of the vehicle in space is obtained. The absolute difference between the instantaneous speed and the average speed of the group, as well as the azimuth angle deviation from the average direction of the group, are calculated. These parameters together constitute the single-node motion deviation value.

[0028] In step S32, the system extracts the single-node motion deviation values ​​of all work nodes within a preset time period (e.g., the past 30 seconds). Using the normal distribution function or Gaussian mixture model from statistics, probability density analysis is performed on these deviation values ​​at the purely textual logic level. When the deviation value of a node falls within a region of extremely low probability distribution, the system identifies it as an abnormal avoidance action or abnormal deceleration behavior and classifies it into the abnormal node set.

[0029] In step S33, the system performs spatiotemporal clustering analysis on the node set. The system calculates the spatial distribution density of the node set in the target road segment. When multiple anomalous nodes cluster within a specific kilometer marker range, and the duration of this behavior exceeds the system's preset response period, such as lasting more than 10 seconds, the system determines that the traffic capacity of the road segment has decreased. This determination logic can identify collective deceleration and avoidance caused by sudden road debris, localized fog, or road icing.

[0030] In step S4, the system calculates adaptive speed limit suggestions in real time based on the determined dynamic fluctuation state. Step S41 involves the system retrieving the corresponding initial speed limit value from a pre-set speed limit gradient library based on the severity of the capacity reduction, such as the proportion of abnormal nodes or the magnitude of speed reduction. For example, when the speed reduction magnitude is in the first interval, a first-level speed limit suggestion is retrieved. Step S42 involves the system adjusting the initial speed limit value based on the physical parameters of the target road segment. These physical parameters include the average slope of the road segment calculated using BeiDou elevation information and the curvature of curves obtained through trajectory fitting. The system employs a weighted correction logic; when the slope is downward and the curvature is large, the initial speed limit value is further reduced. Combined with accessed historical meteorological parameters, such as rainfall and visibility records, a temporary adaptive speed limit value is generated for the current abnormal operating conditions.

[0031] In step S43, the system encapsulates the generated temporary adaptive speed limit value into a standard instruction format and sends it to the onboard terminals of all work nodes within the affected area of ​​the road segment via a wireless communication network, such as a 5G vehicle-to-everything (V2X) dedicated frequency band. Upon receiving the instruction, the onboard terminal reminds the driver to slow down via voice or interface display, or intervenes in the speed control loop of the autonomous vehicle to achieve dynamic closed-loop adjustment of the road segment's speed limit.

[0032] This embodiment introduces a biomimetic swarm intelligence fishing logic, so that speed limit adjustment no longer relies on delayed traffic police reports or static map markings, but instead uses the stress response of the group of vehicles to infer abnormal road conditions, thus improving the real-time performance and accuracy of the system.

[0033] Example 2: Based on Example 1, this example deeply optimizes the parameter correction mechanism of the speed limit adjustment method to meet the special needs of high-altitude mountainous areas and road sections prone to extreme weather.

[0034] In mountainous environments, the signal of the BeiDou Navigation Satellite System may be affected by multipath effects due to mountain obstruction, or by severe fluctuations in tropospheric thickness. In step S1 of this embodiment, the BeiDou positioning sensing module introduces signal quality monitoring logic. The system calculates the carrier-to-noise ratio in real time and evaluates the positioning accuracy in real time based on the geometric accuracy factor of the satellite distribution. For nodes where the positioning error may exceed the standard, the system uses angular velocity and linear acceleration data provided by the inertial navigation unit to perform position reinforcement through extended Kalman filter logic, ensuring that the operating node can still output a continuous and stable motion vector in tunnel entrances or deep canyon sections.

[0035] In step S2, when constructing the group behavior baseline, this embodiment additionally considers environmental constraints. In addition to separation, alignment, and cohesion forces, the system introduces road boundary constraint logic. By online correlation between BeiDou positioning data and a high-precision vector map, the system determines the vehicle's drivable space in the lateral dimension. The biomimetic swarm intelligence model treats the road boundary as an infinitely large repulsive potential field, ensuring that the generated desired motion vector is always locked near the compliant lane centerline.

[0036] In the deviation monitoring step S3, this embodiment enhances the sensitivity to yaw rate. On winding mountain roads, normal steering can result in a large yaw rate. To distinguish between normal steering and emergency steering to avoid obstacles, the system compares the current yaw rate data with the curve curvature parameters on the map. Only when the deviation between the actual yaw rate and the theoretical yaw rate calculated from the current vehicle speed and curve radius exceeds a preset threshold is it identified as an abnormal motion vector.

[0037] In step S4, the speed limit generation process, this embodiment introduces a meteorological auxiliary discrimination factor. The system monitors the rate of change of tropospheric delay generated when BeiDou satellite signals penetrate the atmosphere to invert atmospheric humidity and pressure fluctuations above the road section. When an abnormal step change in the rate of change of delay is detected, the system, combined with data from the vehicle's windshield wiper operating frequency sensor, determines that the current road section is in a state of heavy rainfall. When calculating the adaptive speed limit suggestion, the system uses the attenuation of the road surface friction coefficient as the core correction weight. By consulting a preset road surface adhesion attenuation table, the calculated speed limit value is adjusted downwards based on the value under standard atmospheric pressure.

[0038] This embodiment adds a distribution layer for rate-limiting commands. Considering the discontinuous coverage of communication base stations in mountainous areas, the system adopts multi-hop ad hoc networking technology. When a working node receives an adaptive rate-limiting suggestion, if it detects nearby nodes that cannot connect to the base station, it forwards the rate-limiting command through a short-range communication link, ensuring that the rate-limiting information spreads rapidly within the group and eliminating the security risks caused by communication dead zones.

[0039] Example 3: This example provides a speed limit adjustment system based on BeiDou positioning for different road sections. This system is mainly used in closed automated operation scenarios such as large ports and open-pit mines, aiming to achieve intelligent collaborative speed limiting of high-density unmanned operation vehicles.

[0040] The system includes: a BeiDou positioning and sensing module, a crowdsourced collaborative computing module, a traffic capacity assessment module, and a speed limit decision distribution module.

[0041] The BeiDou positioning and sensing module is installed at the top center of each operating vehicle, such as unmanned mining trucks and AGV flatbed trucks. This module integrates a dual-antenna BeiDou receiver, which not only acquires high-precision three-dimensional coordinates but also calculates the vehicle's heading angle in real time through the phase difference between the two antennas, with a heading accuracy better than 0.1 degrees. The module collects high-precision spatial positions and kinematic parameters of the operating nodes in real time and aggregates the data to the central data preprocessing unit via an industrial-grade wireless link.

[0042] The swarm intelligence collaborative computing module runs on edge computing nodes in the mining area. This module is configured to analyze the collaborative relationships of motion vectors between work nodes using a biomimetic swarm intelligence model. Due to the diverse types of vehicles in the mining area, including heavy-duty unloading vehicles and light-duty auxiliary vehicles, this module assigns different mass weights to vehicles of different tonnages when applying the Reynolds model. Heavy vehicles have greater kinetic energy in the group, and the impact weight of their motion vector changes on the group baseline is set to three times that of light vehicles. The group behavior baseline constructed by the system includes standard speed profiles on each preset work route.

[0043] The traffic capacity assessment module is connected to the collaborative computing module. In mining environments, road dust and falling debris are common risks. This module assesses the real-time traffic status of a road segment by analyzing the statistical characteristics of motion deviations, particularly identifying the spatial queuing features of abnormal deceleration behavior. If multiple unmanned mining trucks deviate from their central trajectories near the same coordinate point, even if the deviation is small, the module will determine the presence of an obstacle based on the stability of clustering features and conclude that the road segment's traffic capacity has decreased.

[0044] The speed limit decision distribution module generates speed limit instructions for specific areas based on the assessment results. In closed scenarios, this module employs a more proactive feedback adjustment mechanism. After issuing a temporary speed limit suggestion, the system continuously monitors the convergence of motion vectors at subsequent operation nodes. If the motion vector deviation of subsequent vehicles decreases rapidly after executing the speed limit instruction, and the group speed stabilizes at the new low-speed level, it indicates that the risk has been controlled. Over time, if abnormal nodes disappear, the system gradually increases the speed limit value in increments of 5 kilometers per second until it returns to the original static speed limit state.

[0045] The system in this embodiment also has a historical data learning function. The database unit continuously records the triggering parameters and final effects of each speed limit adjustment. By mining long-term data, the system can identify common risk areas caused by road subsidence or poor drainage, and automatically adjust the speed limit benchmarks for these areas under different seasons and loads.

[0046] Example 4: This example further elaborates on the deep logic of motion vector deviation clustering feature recognition in the speed limit adjustment method for different road sections based on Beidou positioning, and how to use these features to accurately identify the dynamic fluctuations in road section traffic capacity.

[0047] In step S3, the clustering feature identification of motion vector deviation is not merely a simple data statistics process, but a complex pattern recognition process. The system utilizes density spatial clustering algorithms, such as density clustering logic with noise processing capabilities, to spatially cluster operational nodes that produce significant yaw or deceleration. In the algorithm logic described in plain text, the system searches for other anomalous nodes in the neighborhood of each anomalous node. If, within a preset spatial step size, the number of anomalous nodes exceeds a defined density threshold, these nodes are grouped into a cluster.

[0048] The system analyzes the geometric center of the cluster in real time. If the cluster center is located in the core area of ​​the driving lane, rather than the shoulder or emergency stopping lane, the system determines that the anomaly is not caused by a single vehicle malfunction leading to a pullover, but rather by a group behavior triggered by a common road obstacle. The system monitors the size evolution trend of the cluster. If the number of nodes in the cluster increases over time, or if the cluster extends upstream in the opposite direction of traffic, it is identified as indicating a sudden local obstacle or extreme anomalies in road friction, such as a large-scale oil spill or icing.

[0049] In this scenario, the speed limit adjustment logic will trigger emergency mode. The calculation of adaptive speed limit suggestions will no longer solely rely on the speed limit gradient library but will incorporate a collision warning time constant. The system dynamically adjusts the lead time for issuing speed limits by calculating the estimated time when upstream normal-moving groups enter the clustered area. Through the wireless communication network, the system begins distributing the first-level deceleration command one kilometer upstream of the cluster center point and the second-level mandatory speed limit command 500 meters upstream, forming a stepped-decreasing speed guidance zone.

[0050] This embodiment also introduces the concept of crowd fingerprinting in the construction of road segment crowd behavior benchmarks. Based on historical traffic data from the past few months, the system establishes normalized crowd behavior fingerprints for different date types (e.g., weekdays, holidays) and time periods (e.g., morning and evening rush hours, midnight). These fingerprints record the natural traffic flow fluctuation patterns of the road segment when there are no abnormal disturbances. During real-time processing, the system first compares the currently acquired motion vector data with the corresponding normalized crowd behavior fingerprints. This subtraction logic eliminates the interference of the decrease in average vehicle speed caused by periodic congestion on dynamic anomaly identification. Only when the deviation value differs from the historical fingerprints for that time period does the system initiate deep calculations using the biomimetic crowd intelligence model, reducing the system's false alarm rate and ensuring the authority and scientific validity of speed limit adjustments.

[0051] Example 5: This example focuses on describing the specific implementation details of a speed limit adjustment system based on BeiDou positioning in terms of multi-source data fusion and hardware interaction.

[0052] At the hardware level, the BeiDou positioning and sensing module employs a high-precision positioning SoC chip with multiple systems and frequencies. This chip integrates a highly sensitive radio frequency acquisition unit capable of processing signals from multiple systems such as BeiDou, GPS, and GLONASS. However, in its computational logic, it uses the novel signals from BeiDou-3 as the primary reference, leveraging its unique inter-satellite link technology to improve orbit determination accuracy. During data preprocessing, the system incorporates a nonlinear filtering scheme. Because vehicles experience vibrations and acceleration jumps during operation, the system uses a built-in microelectromechanical system (MEMS) inertial sensor to compensate for carrier phase cycle slips in real time, ensuring the continuity of three-dimensional positioning information even at high speeds.

[0053] The swarm intelligence collaborative computing module is deployed on a cloud architecture based on high-performance processors. To support concurrent computing with tens of thousands of nodes, the module employs distributed streaming processing logic. Each road segment is divided into multiple logical computing partitions, each responsible for maintaining the dynamic spatiotemporal topology mapping within its region. When a working node crosses a partition boundary, the system ensures the continuous relay of motion vector data through a smooth migration protocol. Parallel acceleration technology is used when executing the biomimetic swarm intelligence model computation. Due to the prohibition of formula description, its logic can be expressed as follows: the system performs batch interpolation calculations on the coordinate differences of all nodes within the region, then performs a weighted sum of these differences according to preset weight coefficients, deriving the expected trend of swarm behavior within a very short period.

[0054] The traffic capability assessment module incorporates a multi-dimensional risk evaluation matrix. The matrix's input includes not only clustering features of motion vector deviations but also underlying bus data such as anti-lock braking system (ABS) trigger signals and traction control system intervention frequencies. When BeiDou positioning indicates unexpected vehicle yaw, and simultaneously the bus feedback triggers an emergency braking signal, the traffic capability assessment module increases the risk coefficient. Through this cross-dimensional verification, the system can distinguish between the driver's active lane-changing behavior and the vehicle's passive skidding behavior after losing traction.

[0055] The speed limit decision distribution module employs a dedicated short-range communication protocol with priority in its distribution protocol. Speed ​​limit commands are given the highest level of service priority, allowing them to utilize the bandwidth of other non-safety-related services during network congestion. Upon receiving the command, the onboard terminal sends a speed limit request to the vehicle's powertrain control unit via the controller area network bus. For vehicles with driver assistance features, the system automatically adjusts the target speed setpoint for adaptive cruise control and displays a prominent red icon on the instrument panel indicating the reason for the speed limit, such as: slippery road surface 500 meters ahead.

[0056] The system in this embodiment also includes a closed-loop feedback evaluation unit. This unit is responsible for calculating the compliance rate of vehicles within the target road segment after the speed limit instruction is issued. If it is detected that most vehicles are not slowing down as advised, the system will increase the intensity of the reminder through the broadcast link or coordinate with the roadside electronic information boards to issue warning information. This closed-loop mechanism ensures that the speed limit adjustment is not merely a command, but is truly transformed into a speed reshaping of road traffic flow.

[0057] Example 6: This example explores the application of the present invention in the scenario of smart city logistics management, especially the dynamic speed limit adjustment logic in the area where urban expressways and ground roads switch.

[0058] In urban environments, speed limit characteristics on road sections exhibit spatial heterogeneity. When a logistics operation node transitions from a closed expressway ramp to an open surface road, the expected value of its motion vector undergoes a step change. In this embodiment, the method introduces semantic map overlay logic during step S2 when constructing the dynamic spatiotemporal topology mapping. The system projects the three-dimensional coordinates obtained from BeiDou positioning onto an electronic map with semantic labels, automatically identifying the road level where the vehicle is currently located.

[0059] In step S3, when identifying dynamic fluctuations, the system employs lane-specific deviation monitoring to address the multi-directional traffic characteristics of urban intersections. Utilizing BeiDou's centimeter-level positioning capabilities, the system precisely locks vehicles within specific lane lines. If a group of vehicles in a lane becomes stagnant while adjacent lanes are operating normally, the system identifies this as a disabled vehicle or localized construction in that lane. Speed ​​limit adjustments are no longer applied to the entire road segment but instead generate directional lane-specific speed limit suggestions, guiding upstream vehicles to change lanes in advance and avoiding drastic deceleration and collisions at problem points.

[0060] In step S4, when generating speed limit suggestions, this embodiment additionally considers pedestrian and non-motorized vehicle traffic flow factors. Although BeiDou positioning mainly covers operational vehicles, the system, by accessing the perception results of roadside smart cameras, transforms the distribution density of vulnerable traffic participants into a dynamic potential field in a biomimetic model. When pedestrian crossing traffic in the intersection area exceeds a safe threshold, this potential field generates a downward velocity traction force on the expected motion vector of the operational node, automatically lowering the speed limit suggestion value.

[0061] This embodiment also proposes a virtual coupling speed-limiting logic for logistics fleets. When multiple operational nodes belonging to the same logistics company form a convoy, the system utilizes the high-precision time synchronization function of BeiDou to achieve microsecond-level synchronization within the convoy. The alignment force in the biomimetic swarm intelligence model is enhanced, enabling the entire convoy to behave like a single organism, adjusting its overall speed rhythm according to changes in road capacity. This method not only ensures safety but also reduces the overall energy consumption of the logistics fleet by minimizing unnecessary braking and acceleration.

[0062] Example 7: This example further illustrates the safety redundancy design of the present invention in response to sudden extreme environmental risks, such as strong fog or sudden crosswinds.

[0063] In highway operations, dense fog is characterized by its sudden onset and extremely low visibility, making it a major cause of chain-reaction rear-end collisions. In step S1 of this embodiment, the method analyzes the signal-to-noise ratio fluctuations of the BeiDou satellite signal and combines this with data from roadside meteorological stations to establish an environmental risk perception and early warning mechanism. When the signal penetrates the dense fog area and causes scattering, resulting in a specific pattern of carrier-to-noise ratio attenuation, the system immediately enters a high-alert state.

[0064] In step S2, the road segment group behavior benchmark generated by the system is dynamically shifted according to the diffusion rate of the fog. The cohesion in the biomimetic swarm intelligence model is adjusted to a safe following logic, that is, guiding vehicles to maintain a constant longitudinal distance based on BeiDou positioning coordinates at extremely low line of sight, rather than relying solely on visual recognition of the vehicle in front.

[0065] In step S3, the system's tolerance for motion vector deviation is tightened. Under normal conditions, a speed deviation of five kilometers per hour might be considered noise, but in extreme conditions, a deviation of two kilometers can trigger a warning. The system utilizes a density spatial clustering algorithm to accurately detect the physical phenomenon of deceleration waves propagating upstream due to emergency braking by vehicles in the center of the fog zone.

[0066] The speed limit command generated in step S4 includes a dynamic safe distance suggestion. The onboard terminal not only displays the speed limit value but also provides the driver with the precise physical distance between themselves and the vehicles in front and behind, calculated based on BeiDou positioning. If the distance is lower than the safe coasting distance calculated based on the current speed limit, the system will issue a high-frequency alarm.

[0067] The system in this embodiment has the capability to operate in shadow mode. Before the speed control command is officially issued, the system continuously simulates the expected effectiveness of various speed limiting strategies in the background. By comparing the time it takes for the group motion vector to converge to a steady state under different strategies, the optimal speed limiting gradient is selected. This decision-making logic based on real-time simulation ensures that every issued speed limiting suggestion has sufficient scientific basis and can balance safety and efficiency.

[0068] Example 8: This example describes the collaborative application of this system in a large open-pit mine. In the complex, unstructured terrain of a mine, traditional speed limit signs often struggle to cover the constantly changing work surfaces.

[0069] The BeiDou positioning and sensing module in this embodiment adopts an earthquake-resistant and reinforced design, which can withstand the severe vibrations generated by heavy machinery during blasting and excavation. The system achieves continuous and precise point positioning in complex terrain of the mining area through the BeiDou ground-based augmentation system. Since the roads in the mining area are mostly temporary structures with extremely steep slopes and very small curve radii, the system reads the data from the load-bearing weight sensor of the mining truck in real time when constructing the reference in step S2.

[0070] In this scenario, the biomimetic swarm intelligence model evolves into a dynamic force model. The system calculates the braking distance redundancy of heavy-duty mining trucks at a specific slope. When a mining truck is fully loaded and descending a slope, its speed weight in the group behavior benchmark is set to a negative value, logically manifesting as forcibly decelerating the entire group through its own speed fluctuations.

[0071] In step S3, the system focuses on monitoring the lateral deviation of the vehicle. Since the edges of the mining road are often cliffs or spoil heaps, once the deviation of the vehicle's trajectory from the preset work path by the Euclidean distance exceeds 50 centimeters, the traffic capacity assessment module immediately determines that the traffic conditions of the section have deteriorated, and there may be roadbed collapse or boulders blocking the way. The speed limit recommendation will be instantly reduced to a creeping speed.

[0072] In the distribution logic of step S4, the system utilizes low-Earth orbit satellite links as backup for ground base stations. In blind spots deep within the mining area where signals cannot reach, speed limit commands are relayed via satellite, ensuring absolute delivery of speed control instructions. This all-weather, all-space speed control capability reduces the accident rate in mining operations and improves the overall efficiency of automated mining.

[0073] Example 9: This example explores the compatibility and synergy of this method in mixed traffic flow of manual and autonomous driving, in light of the future development of smart highways.

[0074] In mixed traffic flows, different types of work nodes exhibit varying response times and execution accuracy to speed limit commands. Manually driven vehicles suffer from significant perception delays and operational randomness, while autonomous vehicles possess millisecond-level response capabilities. In step S2 of this embodiment, the work nodes are labeled as either rigid autonomous driving nodes or flexible manually driven nodes.

[0075] In calculating collaborative forces, the biomimetic swarm intelligence model utilizes rigid nodes as stabilizing anchors for the group. The system prioritizes guiding autonomous vehicles to adjust to the suggested speed limit, using their physical barriers to force surrounding human-driven vehicles into a uniform following state. This logic leverages the harmonizing effect of a small number of intelligent vehicles on the overall traffic flow.

[0076] In step S3, the traffic capacity assessment module analyzes the chaos of group behavior. If the motion vector deviation of the manually driven vehicle exhibits divergent characteristics, i.e., the driver displays panicked or uncertain steering actions, the system will identify it as the road segment's traffic capacity entering a highly unstable state.

[0077] In step S4, the adaptive speed limit suggestions generated by the system are distributed differentiatedly based on node attributes. For autonomous driving nodes, a control vector containing a precise acceleration curve is sent; for manually driven nodes, a speed limit value with psychological compensation is sent, for example, slightly lower than the theoretically calculated value by 3 to 5 kilometers per hour, to allow for fluctuations in human operation. Through this hierarchical control, the system achieves a smooth transition of heterogeneous traffic flows under complex risks, preventing mass traffic disturbances caused by single-vehicle malfunctions.

[0078] Example 10: This example details the database management and self-evolution mechanism of the system of the present invention.

[0079] The system's database unit is not merely a storage medium, but a logical module with deep learning capabilities. It records in real time the BeiDou positioning characteristics, motion vector deviation distribution, meteorological parameters, road geometric features, and the final traffic flow recovery time at the time each speed limit adjustment command is triggered.

[0080] Through offline analysis of tens of thousands of speed limit events, the system uses machine learning algorithms to continuously correct the force coefficients in the biomimetic swarm intelligence model. For example, if data shows that the existing speed limit value fails to reduce motion vector deviation when rainfall is 10 millimeters per hour, the system will automatically increase the weight of alignment force and lower the trigger threshold of the speed limit gradient in the next similar operating condition.

[0081] This self-evolutionary capability allows the system to adapt to different driving habits and road material characteristics in different regions. On asphalt roads with low slip coefficients and on concrete roads with high friction, the system will automatically evolve two completely different sets of speed limit correction weights.

[0082] The database unit also has road segment health status diagnosis capabilities. By monitoring the frequency of motion vector deviations in specific road segments over a long period, the system can identify habitual deviation areas caused by unreasonable design, such as the apex of convex vertical curves with limited visibility. The system will feed these findings back to the traffic planning department and, in daily operation, set more preventative speed limit benchmarks for these areas.

[0083] In summary, this invention utilizes the high-precision spatiotemporal reference provided by the BeiDou Navigation Satellite System and combines it with biomimetic swarm collaboration logic to construct a speed limit adjustment mechanism characterized by self-perception, dynamic decision-making, and swarm collaboration. This mechanism enhances the robustness of the traffic system to complex environmental disturbances and has practical application value and technological advancement in reducing traffic accident rates and improving the overall traffic efficiency of the road network. This solution not only solves the persistent problem that current static speed limits cannot cope with dynamic risks but also lays a solid technical foundation for precise traffic flow control in the future era of fully autonomous driving. In practical deployment, this invention can integrate existing BeiDou navigation infrastructure to achieve a qualitative leap in traffic management quality at a relatively low marginal cost, resulting in both economic and social benefits.

[0084] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for adjusting speed limits on different road sections based on BeiDou positioning, characterized in that, Includes the following steps: S1. Differential positioning calculation is performed using carrier phase observations and pseudorange observations from the BeiDou satellite navigation system to obtain centimeter-level real-time three-dimensional positioning information and multi-dimensional motion vector data of multiple operation nodes within the target road segment in real time. S2. Based on the timing function of the Beidou system, the three-dimensional positioning information of the multiple operation nodes is spatiotemporally aligned and a dynamic spatiotemporal topology mapping is constructed. A biomimetic swarm intelligence model based on the Reynolds swarm model is introduced, and each operation node is regarded as an individual in the swarm. A dynamic neighborhood relationship is established based on the relative distance and relative speed between the operation nodes. By logically calculating the separation force, alignment force and cohesion force between the operation nodes in the dynamic neighborhood relationship, the expected motion vector of the swarm in the local area during steady-state driving is obtained, and a road segment swarm behavior benchmark containing the swarm's average speed direction, average speed and driving trajectory envelope is generated. S3. Real-time monitoring of the motion vector deviation of each operation node relative to the road segment group behavior benchmark, extraction of the single node motion deviation value of all operation nodes within a preset time period, identification of the node set of abnormal avoidance actions or abnormal deceleration behavior through statistical distribution function, spatial clustering of the node set using density spatial clustering algorithm, and identification of the dynamic fluctuation state of road segment traffic capacity based on the geometric position of the cluster center and the size evolution trend of the cluster. S4. Based on the determined dynamic fluctuation state, retrieve the corresponding initial speed limit value from the preset speed limit gradient library, and combine it with the average slope of the road section calculated by Beidou elevation information, the curvature of the curve obtained by trajectory fitting, and historical meteorological parameters to perform weighted correction on the initial speed limit value, generate a temporary adaptive speed limit value for the current abnormal working condition, encapsulate the temporary adaptive speed limit value into a standard command format and send it to all the operation nodes within the influence range of the road section through the wireless communication network.

2. The method for adjusting speed limits on different road sections based on BeiDou positioning according to claim 1, characterized in that, The spatiotemporal alignment process described in S2 includes: using the timing signal provided by the BeiDou system to attach the positioning coordinates of all the collected operation nodes to a unified timestamp; By performing resampling and linear interpolation on the time series, the data asynchrony caused by internal clock deviation and link transmission delay between different work nodes is eliminated, ensuring that the spatial positional relationship of different work nodes at the same instant is logically comparable. The construction process of the dynamic spatiotemporal topology mapping includes: transforming the coordinates of isolated job nodes into a topology network with interactive relationships; using a dynamic search algorithm to define the neighborhood range of each job node in real time; and establishing a dynamic neighborhood relationship between two job nodes when the three-dimensional straight-line distance between them is lower than a preset interaction radius threshold.

3. The method for adjusting speed limits on different road sections based on BeiDou positioning according to claim 2, characterized in that, The logical calculation process of the desired motion vector in S2 includes: the separation force logic is configured to ensure a safe distance between adjacent work nodes, which is manifested as a repulsive force that increases as the distance decreases; The alignment logic is configured to guide all work nodes within the dynamic neighborhood relationship to move at a consistent speed and direction, thereby smoothing the velocity field. The cohesion logic is configured to drive work nodes that deviate from the main flow toward the group center, thereby maintaining the integrity of the traffic flow. By performing textual logical calculations on the separation force, alignment force, and cohesion force, the average velocity vector of the group in the current road segment and current time period is determined, and the trajectory envelope range of the operation node in the three-dimensional space is defined based on the average velocity vector.

4. The method for adjusting speed limits on different road sections based on BeiDou positioning according to claim 3, characterized in that, S3 specifically includes: calculating the absolute difference between the instantaneous speed of each of the work nodes and the average speed in the road segment group behavior benchmark, calculating the deviation of the instantaneous azimuth angle from the average direction of the group, and calculating the square and square root of the spatial coordinate difference between the actual driving trajectory and the expected motion vector to obtain the single node motion deviation value. Using a Gaussian mixture model, probability density analysis is performed on all the single-node motion deviation values ​​obtained within a preset time period. Work nodes whose deviation values ​​fall into the region of extremely low probability distribution are identified as abnormal nodes and included in the node set. The density spatial clustering algorithm searches for other abnormal nodes in the neighborhood of each abnormal node. When the number of abnormal nodes within a preset spatial step size exceeds a defined density threshold, the abnormal nodes are grouped into the same cluster.

5. The method for adjusting speed limits on different road sections based on BeiDou positioning according to claim 4, characterized in that, The dynamic fluctuation state of identifying road segment traffic capacity described in S3 includes: analyzing the geometric center position of the cluster; if the cluster center is located in the core area of ​​the driving lane rather than the shoulder or emergency stopping lane, then the dynamic fluctuation state is determined to be an abnormal group behavior caused by common road obstacles. Monitor the increase or decrease of the number of nodes in the cluster over time. If the number of nodes shows an increasing trend and the cluster extends upstream in the opposite direction of driving, it is identified as a sudden local obstacle or an extreme abnormality in road surface friction. The process of constructing the road segment group behavior benchmark also includes: establishing normalized group behavior fingerprints for different date types and different time periods based on historical traffic data; and performing subtraction logic operations on the currently acquired motion vector data and the corresponding normalized group behavior fingerprints during real-time processing to eliminate the interference of periodic congestion on dynamic anomaly identification.

6. The method for adjusting speed limits on different road sections based on BeiDou positioning according to claim 5, characterized in that, The specific logic of the weighted correction described in S4 is as follows: extract the geometric feature parameters of the target road segment, and when it is detected that the average slope of the road segment is downward and the curvature of the curve exceeds the preset curve threshold, perform a deduction operation on the initial speed limit value retrieved from the speed limit gradient library. By introducing meteorological auxiliary discrimination factors, the environmental conditions of the road section are determined by monitoring the rate of change of tropospheric delay generated when Beidou satellite signals penetrate the atmosphere, combined with the wiper operation frequency fed back by vehicle sensors. When it is determined that the working conditions are rain or heavy fog, the calculated speed limit value is adjusted downward based on the value under a standard atmosphere of pressure, according to the preset road adhesion attenuation logic. The distribution process also includes: distributing a first-level deceleration command upstream at a preset distance based on the geometric center position of the cluster, and distributing a second-level forced speed limit command near the cluster center position, forming a stepped speed guide belt.

7. The method for adjusting speed limits on different road sections based on BeiDou positioning according to claim 6, characterized in that, S4 is followed by feedback adjustment and self-evolution steps: After the system issues the temporary adaptive speed limit value, it continuously monitors the convergence of the motion vectors of subsequent work nodes in the target road segment. If the motion vector deviation of the subsequent work nodes decreases over time and the group speed returns to a stable level at the new low speed, the risk is determined to be under control. Once the abnormal nodes in the node set disappear, the system gradually increases the speed limit value by a preset step size until it returns to the original static speed limit state. The database unit records the location characteristics, deviation distribution, meteorological parameters, and traffic flow recovery time at each speed limit adjustment trigger in real time. It uses machine learning algorithms to iteratively correct the stress coefficients in the biomimetic swarm intelligence model and evolves differentiated speed limit correction weights based on the road surface material characteristics of different road sections.

8. A speed limit adjustment system for different road sections based on BeiDou positioning, characterized in that, The system for implementing the method as described in any one of claims 1 to 7 includes: a BeiDou positioning and sensing module, which uses the multi-frequency signals of the BeiDou-3 global satellite navigation system to perform low-noise amplification and digital conversion processing on the satellite signals through the receiver radio frequency front end, and combines the differential calculation of carrier phase observation values ​​and pseudorange observation values ​​to collect the centimeter-level spatial position and multi-dimensional kinematic parameters of all operation nodes in the target road segment in real time, and aggregates the data to the data preprocessing unit; The crowd intelligence collaborative computing module is connected to the Beidou positioning and sensing module. It is configured to perform spatiotemporal alignment of the work node data based on Beidou timing information and construct a dynamic spatiotemporal topology mapping. It uses the separation force, alignment force and cohesion force logic in the biomimetic crowd intelligence model to analyze the motion vector collaborative relationship between work nodes, construct the road segment group behavior benchmark and extract the single node motion deviation value. The traffic capacity assessment module is connected to the collective intelligence collaborative computing module. It is used to perform cluster analysis on operation nodes that generate large motion vector deviations using density space clustering algorithm. By analyzing the geometric center position and scale evolution trend of the clusters, it assesses the real-time traffic status of the road segment and identifies the dynamic fluctuations in traffic capacity. The speed limit decision distribution module is connected to the traffic capacity assessment module and is configured to retrieve the initial speed limit value from the speed limit gradient library based on the traffic capacity assessment results, and perform weighted correction by combining road segment geometric parameters and meteorological parameters to generate a temporary adaptive speed limit value. The speed limit command is then sent to the vehicle terminal through the vehicle network communication link or satellite link, and closed-loop feedback adjustment is performed based on the command compliance rate fed back by the vehicle.

9. The speed limit adjustment system for different road sections based on BeiDou positioning according to claim 8, characterized in that, The BeiDou positioning and sensing module integrates a high-sensitivity radio frequency acquisition unit, which supports the processing of BeiDou-3 system signals and utilizes inter-satellite link technology to improve orbit determination accuracy. The data preprocessing unit incorporates a nonlinear filtering scheme, which uses a microelectromechanical system inertial sensor to compensate for the cycle slip of the carrier phase in real time, ensuring the continuity of three-dimensional positioning information under high-speed motion. The crowd intelligence collaborative computing module is deployed on a high-performance cloud architecture or edge computing node. It uses distributed streaming processing logic to divide the road segment into multiple logical computing slices. Each slice is responsible for maintaining the dynamic spatiotemporal topology mapping within its area and executing parallel accelerated computing of the biomimetic crowd intelligence model. The traffic capacity assessment module has a preset multi-dimensional risk assessment matrix. The input of the multi-dimensional risk assessment matrix includes at least the clustering features of motion vector deviation, the anti-lock braking system trigger signal fed back by the vehicle, and the intervention frequency of the traction control system.

10. The speed limit adjustment system for different road sections based on BeiDou positioning according to claim 8, characterized in that, The system has the ability to process heterogeneous nodes with weights. The collective intelligence collaborative computing module sets different motion feature weights for large heavy-duty trucks and small passenger cars, and sets the influence weight of heavy vehicles in the group behavior analysis to be greater than that of small passenger cars. The speed limit decision distribution module adopts a dedicated short-range communication protocol with priority to ensure that the speed limit command is transmitted first in the case of network congestion, and the speed limit command is connected to the vehicle powertrain control unit through the controller local area network bus of the vehicle terminal. For mixed traffic flows consisting of manually driven vehicles and autonomous vehicles, the swarm intelligence collaborative computing module marks the operation nodes as rigid nodes and flexible nodes. It uses the physical barrier effect of rigid nodes to guide flexible nodes into a uniform speed following state and issues differentiated speed limit commands according to the node attributes.

Images