Intelligent road system for vehicle-road cooperative safety alarm

By building a vehicle-road collaborative safety alarm system and using visible light communication and road stud networks to generate communication quality heat maps, the problems of insufficient risk prediction accuracy and poor dynamic adaptability of alarm strategies have been solved. Accurate early warning and dynamic resource scheduling in high-risk areas have been achieved, and the real-time and reliability of vehicle-road collaborative safety alarms have been improved.

CN120748200AActive Publication Date: 2025-10-03ANHUI ZHONGYI NEW MATERIAL TECH CO LTD

Patent Information

Application Number
CN202511004243.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-07-21
Publication Date
2025-10-03
Estimated Expiration
2045-07-21

AI Technical Summary

Technical Problem

Existing vehicle-road collaborative safety alarm systems find it difficult to effectively integrate vehicle dynamic data, road environment parameters, and historical accident characteristics in risk level assessment, resulting in insufficient risk prediction accuracy. In addition, the alarm strategy lacks dynamic coupling analysis of risk propagation paths and vehicle motion trajectories, making it impossible to achieve collaborative processing of multi-source alarm information and intelligent generation of priority strategies.

Method used

Through visible light communication, it receives and analyzes danger signals from vehicle lights and manually triggered devices, generates initial alarm signals based on surrounding environmental data, uses the status of the road stud network for path planning and communication evaluation, generates a communication quality heat map, outputs priority transmission parameters based on the risk decision module, and realizes real-time response through sound and light alarm and information transmission modules.

Benefits of technology

It improves the accuracy of risk prediction and the dynamic adaptability of alarm strategies, realizes accurate early warning and dynamic resource scheduling in high-risk areas, and improves the real-time and reliability of vehicle-road collaborative safety alarms.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120748200A_ABST
    Figure CN120748200A_ABST
Patent Text Reader

Abstract

The invention discloses an intelligent road system for vehicle-road cooperation safety alarm, and relates to the technical field of vehicle-road cooperation alarm, and the system comprises a communication evaluation module which inputs an optimal communication path into a communication mapping model, a space matching layer carries out the fusion of road geometry and communication topology, a state mapping layer marks the signal intensity and spike node states, and a state analysis module which carries out the analysis of the optimal communication path; generating a communication quality thermodynamic diagram; the risk decision module is used for evaluating the risk level of each region based on the communication quality thermodynamic diagram and outputting priority transmission parameters; and the information transmission module starts sound-light alarm according to the priority transmission parameters, obtains an execution state report, and transmits the execution state report back to the command center. According to the method, the communication mapping model is constructed and the priority transmission parameters are dynamically generated, so that the accuracy of communication state evaluation is improved, and efficient cooperative processing and intelligent priority distribution of multi-source alarm information are realized.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle-road cooperative alarm technology, and in particular to an intelligent road system for vehicle-road cooperative safety alarm. Background Art

[0002] Road traffic safety has always been a global concern. As the country with the most highway mileage in the world, road traffic safety is extremely important to China. Currently, when a road accident occurs, information is typically transmitted through various alarms, or secondary accidents are prevented by placing triangular signs at a specified safety distance. In recent years, vehicle-road collaboration methods based on 5G, edge computing, and artificial intelligence have further optimized real-time data integration and decision-making capabilities, such as dynamic traffic signal control and collision warning systems.

[0003] However, existing vehicle-road collaborative safety alarm systems suffer from two shortcomings: First, traditional risk assessment models struggle to effectively integrate vehicle dynamic data, road environment parameters, and historical accident characteristics, resulting in limited risk prediction accuracy and difficulty adapting to the real-time safety assessment needs of complex traffic scenarios. Second, existing alarm strategies often rely on fixed thresholds or simple priority settings, lacking dynamic coupled analysis of key safety indicators such as risk propagation paths and vehicle movement trajectories. This makes it impossible to achieve coordinated processing of multi-source alarm information and intelligent generation of priority strategies, which in turn affects the targeted nature of warning information and traffic diversion efficiency. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides an intelligent road system for vehicle-road collaborative safety alarm to solve the problems of insufficient risk prediction accuracy and poor dynamic adaptability of alarm strategies.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: The present invention provides an intelligent road system for vehicle-road cooperative safety alarm, which includes: The signal generation module receives and analyzes danger signals from vehicle lights and manual triggering devices in real time through visible light communication, simultaneously collects surrounding environment data, and generates an initial alarm signal; The path planning module triggers the road spike nodes based on the initial alarm signal, obtains the road spike network status, and performs node screening on the road spike network status to generate the optimal communication path; The communication evaluation module inputs the optimal communication path into the communication mapping model. The spatial matching layer integrates the road geometry and communication topology. The state mapping layer marks the signal strength and the state of the road spike node to generate a communication quality heat map. The risk decision module evaluates the risk level of each area based on the communication quality heat map and outputs priority transmission parameters; The information transmission module activates the sound and light alarm according to the priority transmission parameters, obtains the execution status report, and transmits the execution status report back to the command center.

[0007] As a preferred solution of the intelligent road system for vehicle-road cooperative safety alarm of the present invention, wherein: the danger signal includes the triggering event type, accident classification identifier, precise positioning data, accident severity level, and timestamp information; The surrounding environment data includes road spike node communication status parameters, road geometry feature data, real-time traffic flow data, and ambient light intensity parameters.

[0008] As a preferred solution of the intelligent road system for vehicle-road cooperative safety alarm of the present invention, wherein: the initial alarm signal is generated, the specific steps are as follows: Fuse danger signals with surrounding environment data to obtain dynamic risk feature data, quantify the risk of the dynamic risk feature data, and generate an initial alarm signal.

[0009] As a preferred solution of the intelligent road system for vehicle-road cooperative safety alarm of the present invention, the specific steps of obtaining the road stud network status are as follows: The initial alarm signal triggers the spike node via a fixed-frequency optical pulse, which then performs adjacency detection and establishes a spike network connection link. Detect the signal strength and connection quality of each spike node in the spike network connection link, calculate the effective response range and distribution, and output the spike network status.

[0010] As a preferred solution of the intelligent road system for vehicle-road cooperative safety alarm described in the present invention, the specific steps of generating the optimal communication path are as follows: Valid candidate spike nodes are screened based on the spike network status, and communication quality assessment is performed on the valid candidate spike nodes to generate spike node transfer probabilities. Dynamic path optimization is performed on the spike node transfer probabilities to obtain the optimal communication path.

[0011] As a preferred solution of the intelligent road system for vehicle-road cooperative safety alarm described in the present invention, the specific steps of generating the communication quality heat map are as follows: A multimodal feature fusion method is used to integrate cross-domain information and extract hierarchical features from the spatial geometry layer and the communication status layer to construct a communication mapping model. The optimal communication path is input into the communication mapping model. The spatial matching layer uses the geometric topology association algorithm to fuse the road geometry and communication connection relationship and establish the spatial position correspondence. The state mapping layer uses the signal feature extraction method to mark the signal strength and connection quality parameters of each spike node to obtain the link state parameters; The spatial position correspondence and link status parameters are integrated to generate a communication quality heat map.

[0012] As a preferred solution of the intelligent road system for vehicle-road cooperative safety alarm described in the present invention, the specific steps of evaluating the risk level of each area are as follows: Divide the communication quality heat map into multiple sub-region units and obtain the communication quality parameters in each sub-region unit; The communication quality parameters within each sub-area unit are divided into risk levels.

[0013] As a preferred solution of the intelligent road system for vehicle-road cooperative safety alarm described in the present invention, wherein: the output priority transmission parameter, the specific steps are as follows: Based on the regional risk distribution map, the optimized path to avoid high-risk areas is obtained, and the priority transfer parameters are output in combination with the risk level.

[0014] As a preferred solution of the intelligent road system for vehicle-road cooperative safety alarm of the present invention, wherein: the sound and light alarm is started according to the priority transmission parameter, the specific steps are as follows: Parse the risk level identifier and alarm condition field of the priority transfer parameter, verify the validity of the parameters and activate the sound and light alarm.

[0015] As a preferred solution of the intelligent road system for vehicle-road cooperative safety alarm of the present invention, wherein: the specific steps of obtaining the execution status report are as follows: Monitor the sound and light alarm status in real time, perform validity verification on the sound and light alarm status, generate an execution status report, and upload the execution status report to the command center using the optimal communication path.

[0016] The beneficial effects of the present invention are as follows: by constructing a communication mapping model, integrating road geometry and communication topology, marking signal strength and node status, generating a communication quality heat map, realizing visual evaluation of communication status, solving the problem of insufficient single-point detection accuracy, and providing spatial support for risk classification; by parsing priority parameters to start sound and light alarms, loading visible light configuration to transmit information and send back status, forming a "perception-decision-execution-feedback" closed loop, solving the problems of alarm response lag and status unknowability, realizing accurate early warning and dynamic resource scheduling in high-risk areas, and improving the real-time and reliability of vehicle-road collaborative safety alarms. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 Schematic diagram of an intelligent road system for vehicle-road cooperative safety alarm.

[0019] Figure 2 Flowchart generated for the initial alarm signal.

[0020] Figure 3 This is a flowchart of the spike network status output.

[0021] Figure 4 This is a flow chart of sound and light alarm and information feedback. DETAILED DESCRIPTION

[0022] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0023] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0025] Reference Figures 1 to 4 , is an embodiment of the present invention, which provides an intelligent road system for vehicle-road cooperative safety alarm, including the following steps: The signal generation module receives and analyzes danger signals from vehicle lights and manual triggering devices in real time through visible light communication, simultaneously collects surrounding environment data, and generates an initial alarm signal; Danger signals include trigger event type, accident classification identifier, precise location data, accident severity level, and timestamp information; It should be noted that when a vehicle is involved in an accident, the light signals emitted by the headlight LED and the manual triggering device are received in real time through a high-sensitivity photoelectric detector. The received light signal is first demodulated, and the triggering event type and accident classification identifier are parsed according to the pulse position modulation rules; then the precise positioning data encoded in the subsequent light pulse sequence is extracted, and the severity of the accident is identified by decoding the changes in the pulse interval; finally, the timestamp information is parsed to complete the collection and verification of all danger signal fields.

[0026] It should also be noted that visible light communication is a wireless communication technology that uses the visible light band for data transmission. By modulating the brightness, flashing frequency or pulse timing of LED lights, the data to be transmitted is encoded into a light signal, which is received and demodulated by a photodetector to restore the data, thereby achieving high-speed and secure two-way communication while providing lighting.

[0027] Surrounding environment data includes road stud node communication status parameters, road geometry data, real-time traffic flow data, and ambient light intensity parameters; It should be noted that the photoelectric detector monitors the intensity and frequency changes of the received light signal to evaluate the communication status parameters of the road spike node; the inertial measurement unit detects the geometric features such as the slope and curvature of the road surface and records the road geometric feature data; the infrared sensor uses the number and speed of passing vehicles to obtain real-time traffic flow data; and the ambient light sensor measures the light intensity parameters in real time.

[0028] Kalman filtering is used to fuse danger signals with surrounding environment data to obtain dynamic risk feature data, and edge computing is used to quantify risks and generate initial alarm signals; It should be noted that first, the Kalman filtering algorithm is used to temporally and spatially align the danger signal with the surrounding environment data and suppress noise to obtain a multi-source fusion data set; and the obtained multi-source fusion data set is dynamically fused through real-time incremental correction. Then, the real-time data association analysis method is used to obtain dynamic risk feature data containing characteristic indicators such as real-time collision risk probability value, risk diffusion rate and accident impact area range. At the same time, each characteristic indicator of the dynamic risk feature data is compared with the preset risk threshold one by one, and the risk level is defined according to the characteristic indicators that exceed the preset risk threshold, and the current impact range and risk level are encapsulated as an initial alarm signal.

[0029] It should also be noted that the preset risk threshold is determined by the distribution law of the characteristic indicators output by the Kalman filter fusion, and the high percentile values ​​of the real-time collision risk probability value, risk diffusion rate and accident impact area range are used as the judgment critical value; Dynamic fusion refers to the continuous access to the latest collected danger signal and environmental data, and the updating of the existing fused data set through incremental algorithms; The real-time data association analysis method collects data such as the real-time relative speed and distance between the vehicle and the obstacle, the obstacle's movement speed, the road friction state, and the direction of movement. It assesses the real-time collision risk probability based on the correlation relationship that "fast speed and close distance indicate a high collision risk." It also calculates the risk diffusion rate based on the correlation relationship that "fast movement and slippery road surfaces indicate a rapid spread." The accident impact area is then delineated based on the "coverage of movement direction and duration." The Kalman filter is a recursive estimation algorithm that first predicts the current state based on the previous estimate (for example, using the vehicle's speed in the previous second to infer the current position). It then compares newly acquired sensor data with the predicted value and dynamically assigns weights based on the size of the difference and the reliability of the sensor data itself (data with small measurement errors and high stability are more reliable, while data with large errors or susceptibility to interference is less reliable). The algorithm ultimately fuses the data to generate the current optimal state estimate and updates the uncertainty range for the next round of predictions. Edge computing is a distributed computing paradigm that enables localized data processing and real-time analysis by deploying computing resources at the edge of the network, close to the data source. Its operational mechanism involves three key steps: First, data collected by terminal devices is preprocessed at the edge node, including data cleaning, format conversion, and feature extraction to reduce data redundancy. Second, the lightweight computing units built into the edge node execute preset data analysis algorithms, such as anomaly detection, pattern recognition, and real-time decision-making, directly extracting key information from the data and generating primary processing results. Finally, based on preset priority strategies, the edge node uploads high-value information or emergency event results to the cloud while discarding low-value data, thereby reducing data transmission latency and bandwidth usage, and achieving efficient and low-cost real-time response. The edge computing unit consists of a data processing unit, a storage unit, and a communication unit. The data processing unit performs real-time computing tasks, the storage unit temporarily stores collected data and intermediate calculation results, and the communication unit is responsible for data transmission between local devices and with the outside world.

[0030] The path planning module triggers the road spike network based on the initial alarm signal, obtains the road spike network status, and performs node screening on the road spike network status to generate the optimal communication path; The initial alarm signal triggers surrounding road spike nodes via fixed-frequency optical pulses and initiates the communication protocol. Adjacent road spike nodes are then activated synchronously and perform adjacency detection to establish a road spike network connection link. It should be noted that the initial alarm signal first triggers the optical signal transmitter built into the road spike node, generating a periodic sequence of light pulses at a preset frequency, which is then transmitted to surrounding road spike nodes as a communication trigger signal. Adjacent road spike nodes monitor changes in ambient light intensity in real time through photoelectric detectors. When an optical pulse signal that meets the preset frequency is detected, the internal communication protocol parsing program is immediately initiated to complete signal feature extraction and protocol matching verification. After verification, adjacent road spike nodes are activated and point-to-point connections are established in sequence, forming a chain communication network centered on the road spike node that generates the periodic light pulse sequence and transmitting information step by step. This ensures that the alarm information spreads quickly and orderly along the road to the following vehicles and facilities.

[0031] It should also be noted that the parameters of the fixed-frequency optical pulses are determined by the PPM encoding protocol, and the frequency range is 1Hz to 2Hz; The communication protocol is a standardized system of rules for information exchange between road spike nodes. It includes optical pulse encoding methods, timing synchronization mechanisms, data format definitions, and error detection methods (such as duplicate check). It ensures that different road spike nodes can accurately identify, interpret, and respond to trigger signals, achieving reliable transmission of alarm information and link synchronization.

[0032] The system detects the signal strength and connection quality of each spike node in the spike network connection link through optical signal interaction, calculates the effective response range and distribution, and outputs the spike network status data. It should be noted that each road spike node transmits uncoded raw light pulses to adjacent road spike nodes at a preset fixed period and synchronously records the transmission time. After the adjacent road spike node captures the light pulse through a photoelectric sensor, it immediately responds with a response signal at the same optical frequency, while also detecting the received light intensity and signal duration. The transmitter determines the communication delay based on the round-trip time difference of the pulse, and assesses the degree of signal attenuation based on the received light intensity (the light intensity attenuation threshold is set between 10% and 90% of the initial transmission intensity). Through multiple consecutive tests, a database of connection quality in each direction is established. The number of effective responding nodes, the fluctuation range of signal intensity, and the distribution of delay time are continuously tracked to ultimately generate road spike network status data.

[0033] It should also be noted that optical signal interaction is a physical communication method between road spike nodes that uses light and light sensors to transmit data. The transmitting road spike node converts the electrical signal into a sequence of light pulses of specific frequency and intensity, which are then directed into the air through an optical lens. The receiving road spike node uses a photodetector to capture the light pulses, and after photoelectric conversion, filtering and amplification, decodes the original information, achieving state synchronization between nodes, data transmission, and connection quality assessment. The preset fixed cycle duration is 10 milliseconds ± 0.5 milliseconds, and the value is determined by the hardware timer accuracy and communication protocol requirements.

[0034] Based on the spike network status data, a belief propagation algorithm is used to screen valid candidate spike nodes, and a dynamic programming algorithm is used to build a spike node pool. It should be noted that based on the spike network status data, a belief propagation algorithm is first used to obtain the comprehensive confidence score of each spike node. Spike nodes with comprehensive confidence scores above a preset confidence threshold are screened as valid candidate spike nodes. A dynamic programming algorithm is then used to transform the spike network into a graph structure of spike nodes and connection relationships. The cost of each path is determined according to communication quality indicators (taking into account signal strength and latency). Starting from the spike node that generates the periodic optical pulse sequence, the optimal path cost and forwarding order to reach adjacent spike nodes are compared and recorded layer by layer. The spike node sequence with the lowest cost and highest connection quality is selected by backtracking the path, and a spike node pool is generated.

[0035] It should also be noted that the belief propagation algorithm is a method for evaluating the reliability of network nodes through information transmission. It simulates the process of messages being passed back and forth between adjacent network nodes. Each network node collects feedback information from adjacent network nodes and then calculates a confidence score by weightedly averaging the network node's real-time status data with the feedback information. The confidence score is continuously updated and adjusted as the latest feedback from adjacent network nodes is continuously received. After multiple rounds of information exchange, each network node can obtain a relatively accurate confidence value, thereby screening the most reliable connection path in the network. The preset confidence threshold is defined based on the basic reliability requirements of the spike node communication function, and the value range is usually in the middle and high end of the scoring range; The dynamic programming algorithm is a mathematical method that seeks optimal solutions in stages. It breaks down complex problems into several interconnected subproblems and avoids repeated calculations by recording the optimal solution to each subproblem. In a spike network, the dynamic programming algorithm transforms the path selection process from the alarm source to all adjacent spike nodes into a multi-stage decision-making problem. It first determines the next closest spike node, then gradually plans the transmission sequence of subsequent spike nodes. At each step, the comprehensive cost is calculated by comprehensively considering indicators such as signal strength and communication delay, and finally backtracking to find a complete transmission path with the lowest overall cost (i.e., high reliability and low latency).

[0036] Based on the spike node pool, the artificial slime mold algorithm is used to dynamically calculate the spike node transfer probability, and dynamic path optimization is performed on the spike node transfer probability to obtain the optimal communication path; The expression for calculating the spike node transfer probability is: ; Among them, P ij represents the probability of the current spike node i transferring to the neighboring node j, represents the pheromone concentration from spike node i to neighbor node j, α represents the influencing factor of pheromone concentration, and its value range is 0.5≤α≤3; represents the heuristic factor from spike node i to neighbor node j, β represents the influence factor of the heuristic factor, and its value range is 0.5≤β≤2.5; represents the pheromone concentration from spike node i to candidate node k, α represents the influencing factor of pheromone concentration, and its value range is 0.5≤α≤3; represents the heuristic factor from spike node i to candidate node k, β represents the influence factor of the heuristic factor, and its value range is 0.5≤β≤2.5; d ik represents the physical distance from spike node i to candidate node k, d ij represents the physical distance from spike node i to neighbor node j, represents the average distance of all possible paths in the spike node pool, γ represents the distance sensitivity coefficient, and its value range is 0.1≤γ≤1.5; N(i) represents the set of neighboring nodes of the current spike node i, k represents the sum index, and e represents the exponential function; It should be noted that a static spike node pool is first constructed using the ant colony algorithm. During initialization, multiple virtual ants synchronously depart from the alarm source spike node and select a communication path based on a comprehensive evaluation index of signal strength and communication delay. The higher the signal strength and the lower the communication delay, the greater the probability of selection. During their movement, the virtual ants release pheromones on the communication path to mark high-quality communication paths. After multiple iterations, spike nodes with pheromone concentrations reaching a preset threshold (defined based on historical network performance statistics and engineering experience, and the value range is set in the high percentile interval of the pheromone concentration distribution) are selected to form the initial spike node pool. Then, dynamic path optimization is achieved through the artificial slime mold algorithm, which monitors the signal quality and communication status of each node in real time. When a node's signal strength drops or its communication delay increases, the probability of the spike node being selected is reduced, and the transfer priority of neighboring spike nodes in good condition is increased. Through multiple continuous iterative evaluations and adjustments to the path selection strategy, an optimal communication path that can adapt to the current network status is ultimately determined.

[0037] It should also be noted that the ant colony algorithm is a meta-heuristic optimization algorithm that simulates the foraging behavior of ant colonies in nature. It uses multiple virtual ants to explore paths in parallel in the solution space, dynamically releases pheromones to mark high-quality solutions based on the quality of the paths, and uses a positive feedback mechanism to give better paths a higher probability of selection. After multiple generations of iterations, pheromones accumulate on the optimal path to form positive reinforcement, ultimately guiding the colony to converge to the global optimal or near-optimal solution. The artificial slime mold algorithm is a dynamic path optimization algorithm that simulates the behavioral mechanisms of slime mold, such as pulsating contraction, extension, and morphological reconstruction in the process of searching for food, and dynamically adjusts the search strategy to balance global exploration and local development capabilities.

[0038] The communication evaluation module inputs the optimal communication path into the communication mapping model. The spatial matching layer integrates the road geometry and communication topology. The state mapping layer marks the signal strength and the state of the road spike node to generate a communication quality heat map. A multimodal feature fusion method is used to integrate cross-domain information and extract hierarchical features from the spatial geometry layer and the communication status layer to construct a communication mapping model. It should be noted that in the PyTorch deep learning framework, the spatial feature extraction network is called through the nn.Module instruction, and a three-level progressive spatial unit is constructed for the spatial feature extraction network, including coarse-grained units, medium-grained units and fine-grained units. At the same time, different spatial analysis operators are configured for each spatial unit. The coarse-grained unit uses graph convolution to capture the global topological relationship of the road and outputs the initial spatial feature map. The medium-grained unit uses the attention mechanism to obtain the structural importance score of each region in the initial spatial feature map, enhances the feature weights of high-importance regions, and suppresses redundant information of low-importance regions, thereby refining the initial spatial feature map and generating a regional structural feature map. The fine-grained unit uses dynamic edge weight adjustment to capture local detail changes, further optimizes the intermediate spatial feature map and outputs the complete spatial feature map to complete the Construction of the spatial geometry layer; calling the time series state evolution network through the torchdiffeq.odeint parameter, and embedding the state transition solver for the time series state evolution network through the ODESolver class, synchronously setting the time step parameters and convergence threshold for the state transition solver: the time step is set to 0.01 to ensure time series continuity, and the convergence threshold is set to 1e-5 to ensure state stability. The time series state evolution network receives time series communication data (such as channel quality and signal strength series) and simulates the change of state over time through the dynamic function parameterized by the internal neural network; using torchdiffeq.odeint to call ODESolver, gradually obtain the state evolution with a time step of 0.01 seconds, and finally fuse the states of each time step to generate a communication feature map, completing the construction of the communication state layer.

[0039] The spatial geometry layer and the communication state layer are integrated across domains and hierarchical features are extracted through a multimodal feature fusion method. Specifically, the bimodal feature concatenation unit is first called through the torch.cat parameter to concatenate the complete spatial feature map output by the spatial geometry layer with the communication feature map output by the communication state layer in the channel dimension. The multi-scale attention mechanism is then embedded through the CrossModalFusion class, and the initial parameters for the attention mechanism are set simultaneously: the spatial feature weight is set to 0.6, and the communication feature weight is set to 0.4. Finally, the communication mapping model is constructed. Next, the constructed communication mapping model is trained. In the specific operation, the multimodal feature dataset is first divided into a sample set, a training set, and a validation set in a ratio of 7:1:2; on the sample set, the StandardScaler is used to normalize the features, and the stratified sampling method is used to balance the data distribution to generate standardized training samples; on the training set, the AdamW optimizer is used to backpropagate the standardized training samples, and the focus loss function is applied for gradient optimization to obtain the optimized communication mapping model parameters; on the validation set, the optimized communication mapping model parameters are forward inferenced to obtain the validation loss; when the validation loss reaches the convergence threshold, the training is terminated, and the trained communication mapping model is output using the torch.jit.script parameter.

[0040] It should also be noted that the multimodal feature fusion method is a collaborative processing technology for multi-source heterogeneous feature information. The core lies in deeply fusing feature information from different sources and with different forms of expression through specific information association and complementary mechanisms.

[0041] The optimal communication path is input into the communication mapping model. The spatial matching layer uses the geometric topology association algorithm to fuse the road geometry and communication connection relationship and establish the spatial position correspondence. It should be noted that the spatial coordinate sequence of the optimal communication path and the road topology data of the road geometry layer are first extracted and unified into the same geographic coordinate system; then, based on the geometric topology association algorithm, by analyzing the similarity of spatial positions and the compatibility of topological structures, the communication path nodes and the road geometry nodes are bidirectionally matched to establish a preliminary space-communication correspondence relationship; finally, for the erroneous matching caused by coordinate errors, the matching results are adjusted through topology verification, and finally a spatial position correspondence table containing the one-to-one binding of road position identifiers and communication path nodes is output.

[0042] It should also be noted that the geometric topology association algorithm is a computational method used to establish the association relationship between spatial geometric information and topological structure. The core lies in deeply integrating and associating the two by analyzing the geometric features and topological properties of spatial elements. The spatial coordinate sequence contains geometric information such as the longitude and latitude of path nodes, direction angles, etc. The road topology data of the road geometry layer contains geometric features such as the coordinates of road section nodes, lane connection relationships, and traffic sign locations.

[0043] The state mapping layer uses the signal feature extraction method to mark the signal strength and connection quality parameters of each spike node to obtain the link state parameters; It should be noted that the original signal data is first collected in real time through the radio frequency unit of the communication node, covering basic indicators such as the received RSSI (signal strength indicator), SNR (signal-to-noise ratio), end-to-end transmission delay and packet loss rate; the collected original signal data is then smoothed using a sliding window, and by obtaining the mean and standard deviation of the original signal data within the window, instantaneous noise interference is eliminated to obtain a stable signal feature sequence; the overall signal strength is then characterized by the mean RSSI within the statistical window, and the connection quality is quantified by combining the mean SNR, mean delay and packet loss rate; finally, the signal strength value and connection quality parameters are bound to the unique identifier of the corresponding spike node (such as the node ID) to generate link status parameters that dynamically reflect the real-time communication status of each node.

[0044] It should also be explained that the signal feature extraction method is a process of extracting key parameters reflecting link quality from the original signal data collected by the communication node. The core is to remove noise interference through algorithm processing and extract characteristic values ​​that can effectively characterize the communication status. In specific operations, the original signal data including received signal strength, signal-to-noise ratio, end-to-end delay and packet loss rate are first obtained from the radio frequency unit, and then instantaneous fluctuations are eliminated through preprocessing methods such as sliding window smoothing. Then, based on statistical rules (such as mean and standard deviation), the signal strength (the overall strength is represented by the RSSI mean) and connection quality (the comprehensive SNR mean reflects the anti-interference ability, the delay mean reflects the transmission efficiency, and the packet loss rate reflects the reliability) are obtained. Finally, quantifiable node signal strength values ​​and connection quality parameters are generated, providing a direct basis for dynamically evaluating the node communication status.

[0045] The spatial position correspondence and link state parameters are integrated through DS evidence theory to generate a communication quality heat map; It should be noted that, first, spatial location correspondences and link state parameters are associated by geographic location to form a collection of evidence bodies based on specific locations. Subsequently, a basic probability allocation (BPA) is defined for each evidence body based on DS evidence theory, and the connection quality parameters are converted into initial confidence levels through normalization. This initial confidence level is then modified based on the stability of the spike node signal strength (e.g., the coefficient of fluctuation calculated from the RSSI mean and standard deviation) to obtain a preliminary BPA for each piece of evidence. Next, the Dempster synthesis rule is used to resolve conflicts and synthesize multiple evidence bodies in overlapping areas to obtain a comprehensive communication quality confidence value. Finally, this comprehensive communication quality confidence value is mapped to a geographic grid and visualized using a color gradient to generate a communication quality heat map that reflects the correlation between spatial location and communication quality, visually displaying the communication reliability distribution in different regions.

[0046] It should also be noted that the core of the DS evidence theory is to quantify the degree of trust in propositions through BPA, divide the evidence space into a mutually exclusive and exhaustive set of propositions, and assign a probability value (BPA) reflecting the degree of trust to each proposition. The DS evidence theory realizes the fusion of multi-source evidence through the "Dempster synthesis rule" - when there is no conflict between the evidence, it is directly weighted and integrated; when there is a conflict, the weight of the conflicting evidence is reduced, and the contribution of consistent evidence is increased, and finally a comprehensive trust is generated, providing a more reliable basis for decision-making. It is widely used in information fusion, target identification, risk assessment and other fields.

[0047] The risk decision module uses the DQN algorithm to evaluate the risk level of each area based on the communication quality heat map, and adopts a dynamic path planning algorithm to output priority transfer parameters; The communication quality heat map is divided into multiple sub-region units using the spatial rasterization method, and the communication quality parameters in each sub-region unit are obtained through spatial statistical analysis. It should be noted that the basic size of the grid cells is set, taking into account the level of spatial analysis precision (reducing the grid size to capture local changes and expanding it to focus on overall trends), regional spatial complexity (e.g., small grid cells are required in densely built-up urban areas, while larger grid cells can be used in open plains), and hardware processing capabilities (e.g., GPU memory limitations and real-time computing speed). Each pixel in the communication quality heat map is then assigned to the corresponding grid cell according to its latitude and longitude through geographic coordinate mapping. Nearest neighbor interpolation is used to process edge points to ensure that all grid cells are closely spaced and non-overlapping. After grid segmentation, spatial statistical analysis is used to extract the communication quality parameters of each grid cell. The communication quality heat map values ​​within each grid cell are aggregated and statistically analyzed, including mean, standard deviation, and maximum / minimum values. Furthermore, based on the spatial coordinates of the grid cells, the mean difference between adjacent grid cells is obtained to quantify the quality gradient change. The final output is a sub-regional communication quality parameter table consisting of "grid coordinates-mean-standard deviation-extreme value-gradient."

[0048] The DQN algorithm is used to classify the risk level of communication quality parameters in each sub-regional unit to form a regional risk distribution map; It should be noted that the communication quality parameters of each sub-region unit are first sorted as state input, and the risk level labels are annotated in combination with historical data. Then, the DQN algorithm environment is initialized, and the state of each sub-region unit (communication quality parameter vector) is mapped to a high-dimensional state space that can be processed by the DQN algorithm. The action space is defined as a risk level classification, and reward rules are set (such as giving a +1 reward when the predicted level is consistent with the true label, and a -0.5 penalty when it is inconsistent); then the state-action-reward interaction data is stored through the experience replay mechanism, and the greedy algorithm is used to select the policy network. The policy network is continuously trial-and-errored by interacting with the environment (including the task-related operating environment of the sub-region unit communication quality parameter state and the risk level label reward) to optimize the policy network. Finally, the optimized policy network is used to predict the risk level of the communication quality parameters of all sub-region units, and the risk level prediction results are associated with the sub-region space coordinates to form a regional risk distribution map.

[0049] Based on the regional risk distribution map, a dynamic path planning algorithm is used to obtain the optimal path to avoid high-risk areas. The risk level and the optimized path are combined to output the priority transfer parameters using a multi-objective decision-making method. It should be noted that, first, the regional risk distribution map is input into the dynamic path planning algorithm as a spatial constraint condition, and at the same time, combined with dynamic environmental factors such as real-time traffic flow and road connectivity, a comprehensive optimization objective function including risk avoidance objectives and path efficiency objectives is constructed; then, the dynamic path planning algorithm is used to search the path, and the heuristic function is used to guide the path to avoid high-risk areas while ensuring the global optimality of the path; after obtaining multiple candidate optimization paths, a multi-objective decision-making method is used to weigh the path risk exposure and path efficiency indicators (total mileage, estimated time) obtained by weighted summation of the risk levels of each sub-regional unit passed by the statistical candidate path, combined with the coverage of the candidate path in each sub-regional unit, as dual objectives to obtain the comprehensive utility value of each candidate optimization path; finally, the candidate optimization path with the highest comprehensive utility value is selected as the final recommended path, and the key transfer parameters including the path start-end coordinate sequence, estimated travel time and path risk exposure are extracted to form priority transfer parameters.

[0050] It should also be noted that dynamic path planning algorithms are a type of method that can obtain the optimal path in real time in a dynamically changing environment. The core is to continuously update and optimize path decisions by using state space search and heuristic guidance while taking into account the dynamic factors of the environment. The path start-end coordinate sequence is derived from the spatial trajectory of the final recommended path, and the estimated travel time is obtained from the total mileage of the path and the real-time traffic flow; Multi-objective decision-making method is a type of decision analysis technology used to seek a balance or optimal solution among multiple conflicting objectives. The core is to transform multiple objective functions into a comparable unified evaluation system, use Pareto frontier analysis to identify all non-dominated solutions, and thus determine the optimal decision solution in complex trade-offs.

[0051] In the information transmission module, the road stud network activates the sound and light alarm according to the priority transmission parameters, transmits the alarm information to the surrounding facilities through visible light communication, and transmits the execution status back to the command center.

[0052] The road spike network uses a data decoding algorithm to parse the risk level identifier and alarm condition field of the priority transmission parameter, and activates the sound and light alarm after verifying the validity of the parameters; It should be noted that the risk level identifier and alarm condition field of the priority transfer parameter are first parsed through a data decoding algorithm. The received priority transfer parameter is first error-checked to ensure correct data transmission. Then, according to the risk level binary coding comparison table clearly defined in the communication protocol (00 indicates low risk, 01 indicates medium risk, 10 indicates high risk, and 11 is a reserved value), the binary code of the risk level identifier is converted into the corresponding risk level value. At the same time, the code of the alarm condition field is converted into specific trigger parameters. The road stud network then matches the audible and visual alarm mode according to the risk level value, and finally activates the audible and visual alarm device. Through the combined changes of light color, flashing frequency and sound intensity, it sends a visual warning signal that matches the risk level to surrounding road users.

[0053] It should also be noted that the sound and light alarm modes are pre-defined by road safety standards: high risk: the red LED light flashes 5 times per second, and the buzzer sounds at a high frequency of 80 decibels; medium risk: the yellow LED light flashes 3 times per second, and the buzzer sounds at a medium frequency of 60 decibels; low risk: the blue LED light flashes 1 time per second, and the buzzer sounds at a low frequency of 40 decibels.

[0054] A priority scheduling algorithm is used to extract the visible light communication configuration field from the priority transfer parameters, and the configuration parameters are loaded through the embedded communication protocol parser, and the alarm information is transmitted to the surrounding facilities using visible light communication; It should be explained that, first, the priority scheduling algorithm is used to screen out the highest priority communication request from the received priority transfer parameters, and the visible light communication configuration field containing key parameters such as modulation mode, transmission rate, wavelength range, etc. is accurately located and extracted; then, the embedded communication protocol parser parses the extracted visible light communication configuration field, converts the binary-coded configuration parameters into specific communication control instructions, and verifies the validity of the parameters; and initializes the hardware device according to the parsed configuration parameters (adjusts the modulation frequency of the LED light source, sets the emission angle of the optical lens), and converts the alarm information into an optical signal code that meets the requirements of the communication protocol; finally, the modulated optical signal is sent to the surrounding facilities through the visible light emitting unit, and the surrounding receiving equipment captures the optical signal through the photoelectric sensor and parses the alarm content according to the same protocol.

[0055] The road spike network monitors the status of sound and light alarms in real time, generates execution status reports, and uploads them to the command center using the optimal communication path.

[0056] It should be noted that the photoelectric sensor built into the road spike node detects the flashing frequency and photoelectric intensity of the LED light in real time, while the sound sensor collects the decibel value of the buzzer and continuously records the duration of the alarm trigger. The collected data such as the light flashing frequency, sound decibel value and triggering duration are compared with the preset standard range (defined based on the road spike equipment specifications and communication protocol requirements) to determine whether the current sound and light alarm is working properly. The LED light flashing frequency detected by the road spike node through the photoelectric sensor, the buzzer sound decibel value collected by the sound sensor, the continuously recorded alarm triggering duration, the judgment result and the timestamp are integrated to generate a structured execution status report. Based on the optimal communication path, the execution status report is forwarded hop by hop to the road spike gateway node using short-range wireless communication. The road spike gateway node then uploads the execution status report to the command center via the cellular network, realizing real-time remote transmission and monitoring of the alarm execution status.

[0057] In summary, the present invention achieves visual evaluation of communication status by: constructing a communication mapping model, integrating road geometry and communication topology, marking signal strength and node status, generating a communication quality heat map, and realizing visual evaluation of communication status, thereby solving the problem of insufficient single-point detection accuracy and providing spatial support for risk classification; by parsing priority parameters to activate sound and light alarms, loading visible light configuration to transmit information and transmit back status, forming a "perception-decision-execution-feedback" closed loop, solving the problems of alarm response lag and status unknowability, realizing accurate early warning and dynamic resource scheduling in high-risk areas, and improving the real-time and reliability of vehicle-road collaborative safety alarms.

[0058] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. An intelligent road system for vehicle-road cooperative safety alarm, characterized by: include, The signal generation module receives and analyzes danger signals from vehicle lights and manual triggering devices in real time through visible light communication, simultaneously collects surrounding environment data, and generates an initial alarm signal; The path planning module triggers the road spike nodes based on the initial alarm signal, obtains the road spike network status, and performs node screening on the road spike network status to generate the optimal communication path; The communication evaluation module inputs the optimal communication path into the communication mapping model. The spatial matching layer integrates the road geometry and communication topology. The state mapping layer marks the signal strength and the state of the road spike node to generate a communication quality heat map. The risk decision module evaluates the risk level of each area based on the communication quality heat map and outputs priority transmission parameters; The information transmission module activates the sound and light alarm according to the priority transmission parameters, obtains the execution status report, and transmits the execution status report back to the command center.

2. The intelligent road system for vehicle-road cooperative safety alarm according to claim 1, characterized in that: The danger signal includes the trigger event type, accident classification identifier, precise positioning data, accident severity level and timestamp information; The surrounding environment data includes road spike node communication status parameters, road geometry feature data, real-time traffic flow data and ambient light intensity parameters.

3. The intelligent road system for vehicle-road cooperative safety alarm according to claim 2, characterized in that: The specific steps of generating the initial alarm signal are as follows: Fuse danger signals with surrounding environment data to obtain dynamic risk feature data, quantify the risk of the dynamic risk feature data, and generate an initial alarm signal.

4. The intelligent road system for vehicle-road cooperative safety alarm according to claim 1, characterized in that: The specific steps for obtaining the spike network status are as follows: The initial alarm signal triggers the spike node via a fixed-frequency optical pulse, which then performs adjacency detection and establishes a spike network connection link. Detect the signal strength and connection quality of each road spike node in the road spike network connection link, calculate the effective response range and distribution, and output the road spike network status.

5. The intelligent road system for vehicle-road cooperative safety alarm according to claim 4, characterized in that: The specific steps of generating the optimal communication path are as follows: Valid candidate spike nodes are screened based on the spike network status, and communication quality assessment is performed on the valid candidate spike nodes to generate spike node transfer probabilities. Dynamic path optimization is performed on the spike node transfer probabilities to obtain the optimal communication path.

6. The intelligent road system for vehicle-road cooperative safety alarm according to claim 1, characterized in that: The specific steps of generating the communication quality heat map are as follows: A multimodal feature fusion method is used to integrate cross-domain information and extract hierarchical features from the spatial geometry layer and the communication status layer to construct a communication mapping model. The optimal communication path is input into the communication mapping model. The spatial matching layer uses the geometric topology association algorithm to fuse the road geometry and communication connection relationship and establish the spatial position correspondence. The state mapping layer uses the signal feature extraction method to mark the signal strength and connection quality parameters of each spike node to obtain the link state parameters; The spatial position correspondence and link status parameters are integrated to generate a communication quality heat map.

7. The intelligent road system for vehicle-road cooperative safety alarm according to claim 6, characterized in that: The specific steps for assessing the risk level of each area are as follows: Divide the communication quality heat map into multiple sub-region units and obtain the communication quality parameters in each sub-region unit; The communication quality parameters within each sub-area unit are divided into risk levels.

8. The intelligent road system for vehicle-road cooperative safety alarm according to claim 1, characterized in that: Outputting the priority transfer parameter refers to obtaining an optimized path to avoid high-risk areas based on a regional risk distribution map, and outputting the priority transfer parameter in combination with the risk level.

9. The intelligent road system for vehicle-road cooperative safety alarm according to claim 8, characterized in that: The specific steps of starting the sound and light alarm according to the priority transmission parameter are as follows: Parse the risk level identifier and alarm condition field of the priority transfer parameter, verify the validity of the parameters and activate the sound and light alarm.

10. The intelligent road system for vehicle-road cooperative safety alarm according to claim 1, characterized in that: The specific steps for obtaining the execution status report are as follows: Monitor the sound and light alarm status in real time, perform validity verification on the sound and light alarm status, generate an execution status report, and upload the execution status report to the command center using the optimal communication path.

Citation Information

Patent Citations

  • Expressway intersection vehicle-road cooperation system based on intelligent spike and guidance method

    CN110213740A

  • Expressway automatic safety early warning system and method for preventing secondary accident

    CN116312043A

  • Interconnection test method and system for wireless communication module

    CN120128969A

  • Self -driving car traffic accident early warning system

    CN205910859U

  • Systems and methods for jointly optimizing wan and LAN network communications

    US20140321298A1

Cited By

  • Aircraft path planning method and device based on communication optimization, equipment and medium

    CN120848559A

  • Aircraft path planning method and device based on communication optimization, equipment and medium

    CN120848559B