Vehicle-road cooperation information interaction management system

Abstract

The invention discloses a vehicle-road cooperation information interaction management system, and relates to the technical field of vehicle-road cooperation information interaction management, and the system comprises a data collection module which collects vehicle sensor data, infrastructure data and vehicle networking data; the preprocessing module is used for preprocessing the collected data; the data fusion module is used for fusing the data; the environment perception module is used for carrying out environment perception on the fused data and generating a global environment perception graph; the behavior prediction module is used for performing behavior prediction based on the global environment perception graph; the risk assessment module is used for performing risk assessment according to the behavior prediction result; the early warning module is used for generating an early warning signal based on the risk assessment result; the method has the beneficial effects that efficient fusion of multi-source heterogeneous data is realized, an accurate global environment perception graph is generated, the perception capability of dynamic changes in a complex traffic environment is improved, and the reliability of behavior prediction and risk assessment is remarkably enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle-road collaborative information interaction management, and particularly to a vehicle-road collaborative information interaction management system. Background Art

[0002] The vehicle-road collaborative system realizes information interaction between vehicles and road infrastructure, other vehicles and pedestrians by integrating in-vehicle sensors, roadside infrastructure and vehicle networking technology, thereby improving traffic efficiency and enhancing driving safety. Existing vehicle-road collaborative systems still face many challenges in data fusion, environmental perception, behavior prediction and risk assessment, especially in the effective integration of multi-source heterogeneous data and real-time requirements. The existing technologies have not fully met the high standards required for the development of autonomous driving and intelligent transportation.

[0003] Existing behavior prediction models often rely on single or a small number of input variables, and do not fully consider the interaction of multi-dimensional information, resulting in low reliability of prediction results, especially insufficient performance in dealing with emergencies. Existing risk assessment and early warning mechanisms usually cannot respond in a timely manner in the face of a traffic environment with high real-time and high dynamics. There are delays in the transmission and feedback of early warning signals, which greatly affects the overall effectiveness and safety of the system. Summary of the Invention

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

[0005] Therefore, the present invention provides a vehicle-road collaborative information interaction management system to solve the problems of difficult multi-source heterogeneous data fusion and insufficient real-time performance and accuracy in the vehicle-road collaborative system.

[0006] To solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, an embodiment of the present invention provides a vehicle-road collaborative information interaction management system, which includes

[0008] A data acquisition module: collecting vehicle sensor data, infrastructure data and vehicle networking data;

[0009] A preprocessing module: preprocessing the collected data;

[0010] A data fusion module: fusing the data;

[0011] An environmental perception module: performing environmental perception on the fused data to generate a global environmental perception map;

[0012] A behavior prediction module: performing behavior prediction based on the global environmental perception map;

[0013] A risk assessment module: performing risk assessment according to the behavior prediction result;

[0014] Warning module: Generate warning signals based on the risk assessment results;

[0015] Information interaction module: Send warning signals to the vehicle and the driver based on information interaction;

[0016] Feedback module: Feedback the execution operations of the vehicle and the driver to the central processing module;

[0017] Central processing module: Process the feedback execution operations of the vehicle and the driver.

[0018] As a preferred solution of the vehicle-road collaborative information interaction management system described in the present invention, wherein:

[0019] The vehicle sensor data includes GPS data, radar and lidar data, and camera data, the infrastructure data includes signal light status data, weather condition data, and road condition data, and the vehicle networking data includes vehicle-to-vehicle communication data and vehicle-to-road communication data.

[0020] Preprocess the collected data, and input the preprocessed data into the multi-source data fusion module.

[0021] As a preferred solution of the vehicle-road collaborative information interaction management system described in the present invention, wherein:

[0022] Fuse the data, perform environmental perception on the fused data, and generate a global environmental perception map, including,

[0023] Adopt the method of Bayesian inference to fuse the data from different sensors into a unified probability distribution;

[0024] Construct a global environmental perception map based on the fused data, where the nodes represent traffic participants and environmental elements, the edges represent the spatial relationships and interaction relationships between the nodes, and define the global environmental perception map as G=(V,E), where V is the set of nodes and E is the set of edges.

[0025] As a preferred solution of the vehicle-road collaborative information interaction management system described in the present invention, wherein:

[0026] Predict the behaviors of traffic participants based on the global environmental perception map, including,

[0027] Obtain traffic participants and their surrounding environmental information according to the environmental perception map;

[0028] Extract the position information of traffic participants based on GPS data Speed And direction

[0029] Extracting obstacle detection data relative to traffic participants based on radar and lidar data and relative speed

[0030] Extracting object detection information in front of the vehicle based on camera data through vision processing algorithms

[0031] Extracting the current traffic signal state based on the traffic signal state

[0032] Extracting the traffic density of the current road based on road condition data and weather information

[0033] Extracting communication data with surrounding vehicles based on vehicle-to-vehicle communication data, including relative position and relative speed

[0034] Extracting road information obtained from the traffic management center based on vehicle-to-road communication data

[0035] Normalizing the extracted information to form a unified feature vector Expressed as:

[0036]

[0037] According to the feature vector Using linear transformation and non-linear activation function, calculate the intermediate feature, expressed as:

[0038]

[0039] where, F (t) is the intermediate feature, W1 is the weight matrix, b1 is the bias vector, and tanh(·) is the hyperbolic tangent activation function;

[0040] Extracting time series features based on features of multiple time steps to generate a new feature vector H(t), expressed as:

[0041]

[0042] where, H (t) is the newly generated feature vector, λ k is the decay factor, and K is the number of historical time steps;

[0043] Using linear transformation to map the time series feature vector to the predicted output Ŷ(t) of the future trajectory, expressed as:

[0044]

[0045] Among them, is the predicted result of the behavior trajectory at time t, W1 is the weight matrix of the input layer, b1 is the bias vector of the input layer, W2 is the weight matrix of the prediction layer, and b2 is the bias vector of the prediction layer. is the input data sequence at time t, and K is the number of historical time steps.

[0046] As a preferred solution of the vehicle-road collaborative information interaction management system described in the present invention, wherein:

[0047] For the behavior prediction, risk assessment is carried out according to the behavior prediction result, and a warning signal is generated, including

[0048] Based on the predicted trajectory of the host vehicle and the predicted trajectories of surrounding traffic participants, select Δt per second as the time step, and calculate the Euclidean distance between the host vehicle and other traffic participants at each time step Δt, expressed as:

[0049]

[0050] Wherein: is the Euclidean distance between the host vehicle and traffic participant j at time point t + τ; and are the predicted position coordinates of the host vehicle in the x and y directions at time t + τ, respectively; and are the predicted position coordinates of other participants in the x and y directions at time t + τ.

[0051] Detect the collision risk based on the Euclidean distance between the host vehicle and other traffic participants, and the expression is:

[0052]

[0053] Set the safety distance threshold δ safe , if the Euclidean distance between the host vehicle and other traffic participants is less than this threshold, there is a collision risk;

[0054] Combine the speed difference, acceleration difference and road conditions to calculate the risk factor;

[0055] Define the speed difference between the host vehicle and other traffic participants, expressed as:

[0056]

[0057] Wherein: is the speed difference between the host vehicle and traffic participant j at time point t + τ; is the speed vector of the host vehicle at time point t + τ; is the speed vector of traffic participant j at time point t + τ;

[0058] Define the acceleration difference between the host vehicle and other traffic participants, expressed as:

[0059]

[0060] Where: is the acceleration difference between the host vehicle and traffic participant j at time point t + τ; is the acceleration vector of the host vehicle at time point t + τ; is the acceleration vector of traffic participant j at time point t + τ;

[0061] Calculate the impact of road conditions on risk based on road infrastructure data, expressed as:

[0062]

[0063] Where: is the road condition impact factor at time point t + τ; is the road condition coefficient at time point t + τ, is the weather condition at time point t + τ; T max is the maximum allowable temperature or humidity value;

[0064] Construct an analysis and evaluation model based on the risk factor calculation formula, expressed as:

[0065]

[0066] Where: is the comprehensive risk value at time point t + τ; ∈ is a small positive number to prevent division by zero; v max is the maximum speed of the host vehicle; a max is the maximum acceleration of the host vehicle; N is the total number of surrounding traffic participants; is the collision detection indicator function at time point t + τ.

[0067] Set the risk threshold T. If exceeds the threshold, there is a potential risk event and the warning mechanism needs to be triggered immediately;

[0068] Based on the range of values of the risk assessment function divide the warning information into three categories: low-level warning, medium-level warning, and high-level warning. Define the low-level warning risk threshold T low and the high-level warning risk threshold T high . When is less than T low , it is a low-level warning. When is greater than T low and less than T high , it is a medium-level warning. When Rrisk(t) is greater than T high, it is a high - level warning;

[0069] Based on the warning level, warning information is sent to the driver in different ways:

[0070] Low - level warnings are issued through visual cues such as HUD display and dashboard display;

[0071] Medium - level warnings are issued through audio cues such as alarm sounds;

[0072] High - level warnings are issued not only through visual and audio cues, but also through steering wheel or seat vibration feedback.

[0073] As a preferred embodiment of the vehicle - road collaborative information interaction management system described in the present invention, wherein:

[0074] The warning signal is sent to the vehicle and the driver based on information interaction, including,

[0075] The information interaction includes information communication between vehicles, between vehicles and infrastructure, and between vehicles and the cloud. Each vehicle continuously broadcasts its current position and speed information through vehicle - to - vehicle communication protocols;

[0076] When the risk event is detected, an emergency notice is sent through vehicle - to - vehicle communication;

[0077] Through vehicle - to - infrastructure communication, the current traffic signal status and road closure information are obtained from traffic lights and dynamic road sign infrastructure. The vehicle and the driver adjust their driving strategies according to the received information;

[0078] By communicating with the cloud through vehicle - to - cloud communication, the vehicle connects to the cloud platform to obtain global traffic information such as traffic flow distribution and weather conditions within the city. The information is used for long - term route planning and strategy adjustment, and the driving data is uploaded to the cloud for subsequent data analysis and system optimization.

[0079] As a preferred embodiment of the vehicle - road collaborative information interaction management system described in the present invention, wherein:

[0080] The execution operations of the vehicle and the driver are fed back to the central processing module, including,

[0081] The operations of the driver are monitored in real - time. The operations include steering wheel rotation, braking force, and throttle operation, to determine whether the driver responds to the warning signal;

[0082] The feedback data after each decision execution, including driver operations and environmental changes, is sent back to the central processing module for analysis.

[0083] In a second aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and: when the computer program is executed by the processor, any step of the vehicle-road collaborative information interaction management system described in the first aspect of the present invention is implemented.

[0084] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, and: when the computer program is executed by the processor, any step of the vehicle-road collaborative information interaction management system described in the first aspect of the present invention is implemented.

[0085] The beneficial effects of the present invention are as follows: 1. It realizes the efficient fusion of multi-source heterogeneous data, generates an accurate global environment perception map, improves the perception ability of dynamic changes in complex traffic environments, and significantly enhances the reliability of behavior prediction and risk assessment;

[0086] 2. By comprehensively considering the feature extraction information of multiple time steps, it can achieve high-precision behavior prediction, thereby effectively predicting possible dangerous situations and providing a reliable basis for subsequent risk assessment and early warning mechanisms;

[0087] 3. Through real-time risk monitoring, it can flexibly adjust the early warning method according to different risk levels, maximize the guarantee of driving safety, and prevent accidents from occurring. Description of the Drawings

[0088] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0089] Figure 1 It is a flowchart of the vehicle-road collaborative information interaction management system in Embodiment 1. Detailed Embodiments

[0090] To make the above objects, features, and advantages of the present invention more obvious and understandable, the specific embodiments of the present invention will be described in detail below with reference to the drawings in the specification.

[0091] Many specific details are set forth in the following description to facilitate a thorough understanding of the present invention. However, the present invention may be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0092] Second, the "one embodiment" or "embodiment" referred to herein means a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it an individual or alternative embodiment that is mutually exclusive with other embodiments.

[0093] Embodiment 1, referring to Figure 1 , is the first embodiment of the present invention. This embodiment provides a vehicle-road collaborative information interaction management system, including the following steps: Data acquisition module: Collect vehicle sensor data, infrastructure data, and vehicle network data;

[0094] Preprocessing module: Preprocess the collected data;

[0095] Data fusion module: Fuse the data;

[0096] Environment perception module: Perform environment perception on the fused data to generate a global environment perception map;

[0097] Behavior prediction module: Perform behavior prediction based on the global environment perception map;

[0098] Risk assessment module: Perform risk assessment based on the behavior prediction results;

[0099] Warning module: Generate a warning signal based on the risk assessment results;

[0100] Information interaction module: Send the warning signal to the vehicle and the driver based on information interaction;

[0101] Feedback module: Feed back the execution operations of the vehicle and the driver to the central processing module;

[0102] Central processing module: Process the feedback execution operations of the vehicle and the driver.

[0103] The vehicle sensor data includes GPS data, radar and lidar data, and camera data. The infrastructure data includes signal light status data, weather condition data, and road condition data. The vehicle network data includes vehicle-to-vehicle communication data and vehicle-to-road communication data.

[0104] Preprocess the collected data. The preprocessed data is input into the multi-source data fusion module. The input data includes vehicle status information, environment perception information, and infrastructure information with a unified timestamp.

[0105] The step of fusing the data and performing environment perception on the fused data to generate a global environment perception map includes,

[0106] Using the method of Bayesian inference, the data from different sensors are fused into a unified probability distribution;

[0107] Define multiple observations z1, z2, …, z from different sensors n , and the fused estimated value is, and the expression is:

[0108]

[0109] where w i is the weight of each sensor data, and the weight is determined by factors such as the signal-to-noise ratio and historical accuracy of the data, is the independent estimated value of each sensor, and the calculation formula of the weight is:

[0110]

[0111] where is the observation error variance of sensor i.

[0112] Construct a global environment perception map based on the fused data. The nodes represent traffic participants and environmental elements, and the edges represent the spatial relationships and interaction relationships between the nodes. Define the global environment perception map as G = (V, E), where V is the set of nodes and E is the set of edges.

[0113] Label corresponding information for each node and edge in the topological map, such as dynamic information like position, speed, acceleration, etc., and static information like signal light status, road conditions, etc.

[0114] The behavior prediction of traffic participants based on the global environment perception map includes,

[0115] Obtain traffic participant and its surrounding environment information according to the environment perception map;

[0116] Extract traffic participant position information based on GPS data speed and direction

[0117] Extract obstacle detection data relative to the traffic participant based on radar and lidar data and relative speed

[0118] Extract object detection information in front of the vehicle based on camera data through visual processing algorithms

[0119] Extract the current signal light status based on the signal light status

[0120] Extract the traffic density of the current road based on road condition data and weather information

[0121] Extract communication data with surrounding vehicles based on vehicle-to-vehicle communication data, including relative position and relative speed

[0122] Extract road information obtained from the traffic management center based on vehicle-to-road communication data

[0123] Normalize the extracted information to form a unified feature vector Denoted as:

[0124]

[0125] According to the feature vector Adopt a linear transformation and a non-linear activation function to calculate the intermediate feature, denoted as:

[0126]

[0127] where, F (t) is the intermediate feature, W1 is the weight matrix, b1 is the bias vector, and tanh(·) is the hyperbolic tangent activation function;

[0128] During the process of behavior prediction, the input environmental perception data and vehicle state data have high dimensions and heterogeneity. Different types of data have different feature spaces and different measurement criteria. The speed and acceleration of the vehicle belong to physical quantities, while the road conditions and signal light states are categorical data or discrete data. Therefore, directly using these data for trajectory prediction may not be able to fully explore their potential correlations and complex time-dependent relationships.

[0129] To effectively process heterogeneous data and extract useful information for trajectory prediction, through a step of representing intermediate features, the original data is reduced in dimension, normalized, and feature extracted so that the subsequent deep learning model can better capture the mutual relationships and temporal characteristics between the data;

[0130] The position, direction, speed, and acceleration of the ego vehicle: directly affect the motion state of the vehicle and are the core variables for trajectory prediction. Through intermediate feature extraction, the relationship between these variables and the future trajectory can be understood.

[0131] The position, speed, and acceleration of other traffic participants: used to predict possible interaction behaviors, possible collision risks, or avoidance behaviors. Intermediate feature extraction can help the model understand the complexity of these interactions.

[0132] Infrastructure data, including signal light status, road conditions, and weather information: Usually categorical or discrete data, through feature extraction, it can be transformed into continuous feature vectors for easy fusion with other physical quantity data.

[0133] Connected vehicle data: Reflects traffic dynamics in a larger area. Through feature extraction, it can help the model capture the relationship between local and global traffic states.

[0134] Extract time series features based on multiple time steps to generate a new feature vector H(t), expressed as:

[0135]

[0136] Where H (t) is the newly generated feature vector, λ k is the decay factor, and K is the number of historical time steps;

[0137] Use linear transformation to map the time series feature vector to the predicted output Ŷ(t) of the future trajectory, expressed as:

[0138]

[0139] Where, is the predicted result of the behavior trajectory at time t, W1 is the weight matrix of the input layer, b1 is the bias vector of the input layer, W2 is the weight matrix of the prediction layer, and b2 is the bias vector of the prediction layer, is the input data sequence at time t, and K is the number of historical time steps.

[0140] For the said behavior prediction, conduct risk assessment based on the behavior prediction result and generate a warning signal, including,

[0141] Based on the predicted trajectory of the ego vehicle and the predicted trajectories of surrounding traffic participants, select Δt seconds as the time step, and for each time step Δt, calculate the Euclidean distance between the ego vehicle and other traffic participants, expressed as:

[0142]

[0143] Where: is the Euclidean distance between the ego vehicle and traffic participant j at time t+τ; and are the predicted position coordinates of the ego vehicle in the x and y directions at time t+τ respectively; and are the predicted position coordinates of other participants in the x and y directions at time t+τ.

[0144] Detect collision risks based on the Euclidean distance between the ego vehicle and other traffic participants. The expression is:

[0145]

[0146] Set the safety distance threshold δ safe , if the Euclidean distance between the host vehicle and other traffic participants is less than this threshold, there is a collision risk;

[0147] Calculate the risk factor by combining speed difference, acceleration difference and road conditions;

[0148] Define the speed difference between the host vehicle and other traffic participants, expressed as:

[0149]

[0150] Where: is the speed difference between the host vehicle and traffic participant j at time point t + τ; is the speed vector of the host vehicle at time point t + τ; is the speed vector of traffic participant j at time point t + τ;

[0151] Define the acceleration difference between the host vehicle and other traffic participants, expressed as:

[0152]

[0153] Where: is the acceleration difference between the host vehicle and traffic participant j at time point t + τ; is the acceleration vector of the host vehicle at time point t + τ; is the acceleration vector of traffic participant j at time point t + τ;

[0154] Calculate the impact of road conditions on risk based on road infrastructure data, expressed as:

[0155]

[0156] Where: is the road condition impact factor at time point t + τ; is the road condition coefficient at time point t + τ, is the weather condition at time point t + τ; T max is the maximum allowable temperature or humidity value;

[0157] Based on the risk factor calculation formula, construct an analysis and evaluation model, expressed as:

[0158]

[0159] Where: is the comprehensive risk value at time point t + τ; ∈ is a small positive number to prevent division by zero; v maxis the maximum speed of the host vehicle; a max is the maximum acceleration of the host vehicle; N is the total number of surrounding traffic participants; is the collision detection indication function at time point t + τ.

[0160] Set a risk threshold T. If exceeds the threshold, there is a potential risk event, and the warning mechanism needs to be triggered immediately;

[0161] Based on the value range of the risk assessment function divide the warning information into three categories: low - level warning, medium - level warning, and high - level warning. Define the low - level warning risk threshold T low and the high - level warning risk threshold T high When is less than T low it is a low - level warning. When is greater than T low and less than T high it is a medium - level warning. When Rrisk(t) is greater than T high it is a high - level warning;

[0162] Based on the warning level, send warning information to the driver in different ways:

[0163] The low - level warning is issued through visual cues such as HUD display and dashboard display;

[0164] The medium - level warning is issued through audio cues such as alarm sounds;

[0165] The high - level warning is issued not only through visual and audio cues but also through steering wheel or seat vibration feedback.

[0166] The sending of the warning signal to the vehicle and the driver based on information interaction includes,

[0167] The information interaction includes information communication between vehicles, between vehicles and infrastructure, and between vehicles and the cloud. Each vehicle continuously broadcasts its current position and speed information through vehicle - to - vehicle communication protocols;

[0168] When the risk event is detected, send an emergency notice through vehicle - to - vehicle communication, and the content includes the event type, occurrence location, and timestamp;

[0169] Through vehicle - to - infrastructure communication, obtain the current traffic signal status and road closure information from traffic lights, dynamic road sign infrastructure, and the vehicle and the driver adjust the driving strategy according to the received information;

[0170] The vehicle is connected to the cloud platform through communication with the cloud, obtains global traffic information such as traffic flow distribution and weather conditions within the city, uses the information for long-term route planning and strategy adjustment, and uploads driving data to the cloud for subsequent data analysis and system optimization.

[0171] The feedback of the vehicle and driver's execution operations to the central processing module includes:

[0172] Real-time monitoring of the driver's operations, including steering wheel rotation, braking force, and throttle operation, to determine whether the driver responds to the warning signal;

[0173] The feedback data after each decision execution, including driver operations and environmental changes, is transmitted back to the central processing module for analysis.

[0174] This embodiment also provides a computer device applicable to the vehicle-road collaborative information interaction management system, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the vehicle-road collaborative information interaction management system proposed in the above embodiment.

[0175] This computer device can be a terminal, which includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of this computer device is used to provide computing and control capabilities. The memory of this computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The communication interface of this computer device is used to communicate with external terminals in a wired or wireless manner, and the wireless manner can be implemented through WIFI, operator networks, NFC (Near Field Communication), or other technologies. The display screen of this computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of this computer device can be a touch layer covered on the display screen, or a button, trackball, or touchpad set on the computer device housing, or an external keyboard, touchpad, or mouse, etc.

[0176] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the vehicle-road collaborative information interaction management system proposed in the above embodiment. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, a magnetic disk, or an optical disc.

[0177] In summary, the present invention is through:

[0178] Embodiment 2, referring to Table 1, is the second embodiment of the present invention. To further verify the technical solution of the present invention, experimental simulation data of the vehicle-road collaborative information interaction management system is given.

[0179] To verify the performance of the vehicle-road collaborative information interaction management system, this test was conducted in a simulated urban traffic environment. The experimental scenario covered various complex traffic conditions, including high-density traffic flow, unpredictable weather conditions, sudden situations such as traffic signal failures, etc. The test objects included the intelligent vehicle system 1 equipped with the system of the present invention and the traditional vehicle system 2 not equipped with the system. The collection of experimental data included data from sensors such as GPS, radar, lidar, and cameras, as well as traffic signal status data, weather condition data, road condition data, vehicle-to-vehicle communication data, and vehicle-to-road communication data.

[0180] In the specific implementation process, first, the data acquisition module collects the vehicle sensor data, infrastructure data, and vehicle networking data in the experiment in real time. Then, the preprocessing module filters, denoises, and synchronizes the time of these data to ensure the accuracy and timeliness of the data. After that, the data fusion module uses the Bayesian inference method to fuse multi-source data into a unified probability distribution and generates a global environment perception map based on this. The perception map comprehensively displays traffic participants, environmental elements, and their spatial and interaction relationships.

[0181] In the behavior prediction module, based on the environmental perception map and multi-dimensional sensor data, features such as the location information, speed, and direction of traffic participants are extracted, and the future behavior trajectories are predicted through a time series model. The risk assessment module analyzes these prediction results, calculates the Euclidean distance, speed difference, and acceleration difference between the host vehicle and other traffic participants, and quantitatively evaluates the collision risk in combination with road condition data. Once potential risks are detected, the warning module generates corresponding warning signals according to the risk level and sends these signals to the vehicle driver through the information interaction module to timely remind the driver to take necessary operations.

[0182] During the test process, the vehicle driver of System 1 responds to the warning signals. Operations such as steering wheel rotation, braking force, and throttle operation are monitored in real time and transmitted back to the central processing module through the feedback module for analysis. The central processing module further optimizes the decision-making logic of the system and applies new strategies in the next test. Through this closed-loop system, the vehicle can maintain higher safety and response speed in complex traffic environments.

[0183] Specifically, it is shown in Table 1 below:

[0184] Table 1: Parameter comparison table between System 1 and System 2

[0185]

[0186] Based on data analysis, the performance of the system of the present invention in different environments can be seen. From the perspective of the average response time, the response times of System 1 in rainy days and sunny days are 0.8 seconds and 0.7 seconds respectively, while the response times of System 2 are 1.2 seconds and 1.0 seconds. This shows that System 1 has significant advantages in data fusion and real-time risk assessment, can analyze and process complex traffic information faster, and thus timely send warning signals to the driver, which is particularly important in high-risk environments and greatly reduces the probability of accidents.

[0187] In terms of the collision risk assessment value, the assessment values of System 1 in rainy days and sunny days are 0.3 and 0.2 respectively, which are significantly lower than 0.6 and 0.4 of System 2. This shows that System 1 can more accurately judge potential collision risks by comprehensively considering multi-dimensional data and take corresponding warning measures. Due to the lack of efficient data fusion and environmental perception capabilities, System 2 has a higher risk assessment value and is difficult to provide reliable safety guarantees in complex environments.

[0188] Data on the driver operation error rate also shows that the operation error rates of System 1 in rainy and sunny days are 2.5% and 1.8% respectively, significantly lower than 4.6% and 3.2% of System 2. This indicates that through real-time warning and information interaction, System 1 can effectively assist the driver in making correct decisions, reducing operation errors caused by information lag or insufficiency. However, due to untimely or inaccurate warning information, System 2 may lead the driver to make wrong judgments at critical moments, thus increasing the risk of operation errors.

[0189] From the perspective of the total number of warnings and the number of successful risk avoidance events, System 1 issued 10 and 8 warnings in rainy and sunny days respectively, and the number of successful risk avoidance events was 9 and 7. While the number of warnings of System 2 was 15 and 12, and the number of successful risk avoidance events was only 10 and 8. This shows that the warning mechanism of System 1 is better, which can issue warnings when necessary and accurately guide the driver to avoid risks, avoiding the situation of unnecessary alarms disturbing the driver's attention.

[0190] In summary, the vehicle-road collaborative information interaction management system of the present invention significantly improves the safety and response ability of vehicles in complex traffic environments through advanced multi-source data fusion, accurate environment perception and real-time behavior prediction. Compared with the prior art, this system shows significant advantages in response speed, risk assessment accuracy and driver assistance, not only reducing the potential risk of traffic accidents, but also improving the operation accuracy and sense of security of the driver.

[0191] 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. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A vehicle-road collaborative information interaction management system, characterized in that: including, a data acquisition module: acquiring vehicle sensor data, infrastructure data, and vehicle networking data; a preprocessing module: preprocessing the acquired data; a data fusion module: fusing the data; an environment perception module: performing environment perception on the fused data to generate a global environment perception map; a behavior prediction module: performing behavior prediction based on the global environment perception map; a risk assessment module: performing risk assessment according to the behavior prediction results; a warning module: generating a warning signal based on the risk assessment results; an information interaction module: sending the warning signal to the vehicle and the driver based on information interaction; a feedback module: feeding back the execution operations of the vehicle and the driver to the central processing module; a central processing module: processing the fed-back execution operations of the vehicle and the driver.

2. The vehicle-road collaborative information interaction management system according to claim 1, characterized in that: The vehicle sensor data includes GPS data, radar and lidar data, and camera data. The infrastructure data includes signal light status data, weather condition data, and road condition data. The vehicle networking data includes vehicle-to-vehicle communication data and vehicle-to-road communication data; Preprocessing the acquired data, and inputting the preprocessed data into a multi-source data fusion module.

3. The vehicle-road collaborative information interaction management system according to claim 2, wherein: The fusing of the data and the performing of environment perception on the fused data to generate a global environment perception map includes: Using the method of Bayesian inference to fuse data from different sensors into a unified probability distribution; Constructing a global environment perception map based on the fused data, where the nodes represent traffic participants and environmental elements, and the edges represent the spatial relationships and interaction relationships between the nodes. Defining the global environment perception map as G=(V, E), where V is the set of nodes and E is the set of edges.

4. The vehicle-road collaborative information interaction management system according to claim 3, wherein: The performing of behavior prediction on traffic participants based on the global environment perception map includes: Obtaining traffic participants and their surrounding environment information according to the environment perception map; Extracting traffic participant location information based on GPS data Speed and direction Extracting obstacle detection data relative to traffic participants based on radar and lidar data and relative speed Extract object detection information in front of the vehicle based on camera data through vision processing algorithms Extract the current signal light state based on the signal light state Extract the traffic density of the current road based on road condition data and weather information Extract communication data with surrounding vehicles based on workshop communication data, including relative position and relative speed Road information obtained from a traffic management center based on vehicle-road communication data 5. The vehicle-road collaborative information interaction management system according to claim 4, wherein: For the extracted road information Perform normalization to form a unified feature vector According to the eigenvector Use linear transformation and non-linear activation function to calculate the intermediate features; Extracting time series features based on features of multiple time steps to generate a new feature vector H(t); Using a linear transformation to map the time series feature vector to the predicted output Ŷ(t) of the future trajectory.

6. The vehicle-road collaborative information interaction management system according to claim 5, characterized in that: The behavior prediction, performing risk assessment according to the behavior prediction results, and generating a warning signal, includes: Based on the predicted trajectory of the host vehicle and the predicted trajectories of surrounding traffic participants, selecting each second as the time step Δt, and calculating the Euclidean distance between the host vehicle and other traffic participants at each time step Δt; Detecting collision risks based on the Euclidean distance between the host vehicle and other traffic participants; Set the safety distance threshold δ safe , if the Euclidean distance between the host vehicle and other traffic participants is less than this threshold, there is a risk of collision.

7. The vehicle-road collaborative information interaction management system according to claim 6, wherein: Calculating risk factors by combining speed differences, acceleration differences, and road conditions; Defining the speed difference between the host vehicle and other traffic participants; Defining the acceleration difference between the host vehicle and other traffic participants; Calculating the impact of road conditions on risks based on road infrastructure data; Constructing an analysis and evaluation model based on the risk factor calculation formula.

8. The vehicle-road collaborative information interaction management system according to claim 7, wherein: Set the risk threshold T. If it exceeds the threshold, there is a potential risk event, and the early warning mechanism needs to be triggered immediately; Based on the value range of the risk assessment function the warning information is divided into three categories: low-level warning, medium-level warning, and high-level warning. Define the low-level warning risk threshold T low and the high-level warning risk threshold T high . When is less than T low , it is a low-level warning. When is greater than T low and less than T high , it is a medium-level warning. When Rrisk(t) is greater than T high , it is a high-level warning; Based on the warning level, issuing warning information to the driver in different ways: Low-level warnings are issued through visual cues on the HUD display and the dashboard display; Medium-level warnings are issued through audio cues of an alarm sound; High-level warnings are issued not only through visual cues and audio cues, but also through vibration feedback of the steering wheel or the seat.

9. The vehicle-road collaborative information interaction management system according to claim 8, characterized in that: The sending of the warning signal to the vehicle and the driver based on information interaction includes: The information interaction includes information communication between vehicles, between vehicles and infrastructure, and between vehicles and the cloud. Each vehicle continuously broadcasts its current position and speed information through vehicle-to-vehicle communication protocols. When the risk event is detected, an emergency notice is sent through vehicle-to-vehicle communication. Through vehicle-to-infrastructure communication, the current traffic signal status and road closure information are obtained from traffic lights and dynamic road sign infrastructure. Vehicles and drivers adjust their driving strategies according to the received information. Through vehicle-to-cloud communication, connect to the cloud platform to obtain global traffic information such as traffic flow distribution and weather conditions within the city. The information is used for long-term route planning and strategy adjustment, and the driving data is uploaded to the cloud for subsequent data analysis and system optimization.

10. The vehicle-road collaborative information interaction management system according to claim 9, wherein: The feedback of the execution operations of the vehicle and the driver to the central processing module includes Real-time monitoring of the driver's operations, which include steering wheel rotation, braking force, and throttle operation, to determine whether the driver responds to the warning signal. The feedback data after each decision execution, including driver operations and environmental changes, is transmitted back to the central processing module for analysis.

Images