A vehicle density estimation method based on perception message broadcasting and probability inference

Abstract

The present invention relates to the technical field of vehicle-mounted self-organizing networks, and provides a vehicle density estimation method based on perception message broadcasting and probability derivation. The density estimation is performed using the perception messages periodically broadcast by vehicles in a vehicle-mounted self-organizing network. First, each neighbor node is analyzed. i The average position of the nodes and the number of messages received are used to calculate the point-to-point packet reception probability. The PRP function that varies with the receiving distance is then obtained. The node perception rate and the average perception ratio (AAR) are calculated through probability deduction. Finally, the vehicle density is estimated by combining the number of perceived vehicles and the AAR. The whole process does not require additional communication overhead and only relies on the received perception messages. The impact of message loss on density estimation is considered through a probabilistic model, which is used to accurately evaluate vehicle density and facilitate congestion control, so as to improve the reliability and accuracy of vehicle communications and the safety of real-time traffic.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle-mounted self-organizing networks, and in particular to a vehicle density estimation method based on perception message broadcasting and probability derivation. Background Art

[0002] With the rapid increase in the number of cars, urban traffic problems have become increasingly prominent. Accurate estimation of vehicle density has become an important means to improve traffic management efficiency and ensure driving safety. In the existing technology, vehicle density estimation methods mainly include infrastructure-based methods and vehicular ad hoc network (VANET)-based methods.

[0003] Infrastructure-based methods rely on pre-installed cameras and sensors (such as induction loop detectors, roadside radars, etc.) to collect traffic status information and estimate vehicle density. However, these methods have obvious limitations: first, the coverage of infrastructure is limited, making it difficult to achieve comprehensive monitoring of the entire traffic network; second, the construction and maintenance of infrastructure is expensive and easily affected by weather and geographical location; in addition, such methods lack real-time performance and are difficult to reflect changes in traffic conditions in a timely manner.

[0004] VANET-based methods collect traffic status information through communication between vehicles, without the need for pre-installed infrastructure, and are therefore more flexible and timely. However, the communication channels in VANET are susceptible to interference and fading, resulting in the loss of perception messages, which in turn affects the accuracy of vehicle density estimation. Some existing VANET-based vehicle density estimation methods mainly rely on the number of received perception messages to estimate vehicle density, but due to the problem of message loss, this method may perform well in low-density areas, but tends to underestimate the actual vehicle density in high-density areas.

[0005] The prior art patent CN103269478B proposes a fast broadcast method based on location information in a vehicle-mounted self-organizing network, which aims to solve the problems of large delay and poor reliability faced by broadcast information transmission in VANET. This method uses multiple vehicles as candidate forwarding vehicles to reduce the number of retransmissions at the source point, and adaptively selects the broadcast mechanism according to the vehicle location. However, this method mainly focuses on the transmission efficiency and reliability of broadcast information, and does not directly optimize the accuracy of vehicle density estimation. Especially in high-density traffic scenarios, due to the increase in channel competition and message loss, the performance of this method in vehicle density estimation may be limited.

[0006] The prior art patent CN115565368A provides a method for quickly estimating traffic demand for urban traffic simulation. This method estimates traffic demand based on population density data of the target area, combined with road network information and gravity model. Although this method has certain accuracy and efficiency in traffic demand estimation, it is mainly aimed at urban traffic simulation rather than real-time vehicle density estimation. In addition, this method relies on a large amount of population density and road network data, and the calculation process is relatively complex, making it difficult to directly apply to real-time vehicle density estimation scenarios.

[0007] Existing technologies still face many challenges in vehicle density estimation, especially in terms of real-time performance, accuracy and flexibility. Infrastructure-based methods are limited by cost, coverage and real-time performance; while VANET-based methods face inaccurate estimation problems caused by channel interference and message loss. Therefore, a more efficient, accurate and flexible vehicle density estimation method is needed to adapt to the increasingly complex urban traffic environment. Summary of the invention

[0008] In view of this, the present invention aims to solve the problems of inaccurate vehicle density estimation, low communication efficiency and high cost in the prior art, and proposes a vehicle density estimation method based on perception message broadcasting and probability derivation, which only uses the received perception messages to determine the vehicle density, has no additional communication overhead, and has high evaluation accuracy.

[0009] To achieve the above object, the technical solution of the present invention is achieved as follows:

[0010] A vehicle density estimation method based on perception message broadcasting and probability derivation includes the following steps: S101: for any observation node O , initialize the observation interval , data rate , Sending range R , Observation distance increment , observation distance parameters;

[0011] S102: Observation Node O Collect the perception messages sent by neighboring nodes and analyze each neighboring node based on the received perception messages i the average position and number of received sensing messages;

[0012] S103: Calculate observation nodes O Received from neighboring nodes i The acceptance rate of packets , ,

[0013] in For the observation node in the observation period Successfully received i The number of packets sent, For Node i During the observation period The total number of packets sent within

[0014] S104: For the k vehicle nodes, calculate this k The average PRP of the nodes is taken as the observation distance r PRP at

[0015] S105: Get m PRP at the observation distance, and the corresponding observation distance m The PRP function that changes with the receiving distance is obtained by linear fitting using the least squares method.

[0016] S106: Evaluate the application layer perception probability NAP according to the fitted PRP function, where NAP is the probability that the receiver successfully receives at least one data packet sent by the sender within the observation period;

[0017] S107: Calculate the average perception ratio AAR of the perceived vehicles according to the perception probability NAP, where the average perception ratio AAR is the ratio of the number of neighbors perceived by the marked node to the total number of neighbors;

[0018] S108: Estimate the vehicle density based on the number of sensed vehicles and the sensed vehicle ratio AAR. The vehicle density is calculated using the following formula: :

[0019] ;

[0020] in, , is the sending range, Refers to the number of perceived vehicles.

[0021] Furthermore, in step S101, the observation distance parameter includes a set of observation distances , the observation distance is the given receiving distance, where j ( ) is the index of the observation distance, , .

[0022] Furthermore, in step S102, the observation node O Neighbor nodes that receive at least one sensing message are numbered in sequence. i It means that each neighbor node is analyzed according to the perception message received during the observation interval. iThe average position and the average position from the node i The number of perception messages.

[0023] Furthermore, in step S102, the observation node O Neighboring nodes that receive multiple perception messages i , calculate the observation node O With neighboring nodes i The receiving distance between them is used to correspond the neighboring nodes to different observation distances according to the receiving distance.

[0024] Furthermore, in step S104, the observation node is calculated O With adjacent nodes i Receiving distance between , will have the same index j = The vehicle nodes are grouped into the same list, and the relationship between the observation distance and the receiving distance is:

[0025] ;

[0026] Observation distance r The calculation formula of PRP at is:

[0027] ;

[0028] in, For the j Observation distance, Neighbor node i With the observation node O The distance between is the observation distance increment.

[0029] Further, in step S105, each observation distance and the PRP under the observation distance constitute a sample point , a total of m sample points, among which, m≤n ,right m The sample points are smoothed and denoised using a linear smoothing filter, and then a polynomial fitting is performed to obtain the PRP function that varies with the receiving distance.

[0030] Furthermore, when fitting the PRP function, the order of the polynomial fitting is smaller than the predefined residual square sum, and the fitted PRP function is expressed as , the residual sum of squares S It is expressed as:

[0031] .

[0032] Furthermore, in step S106, NAP is calculated using a probability derivation method:

[0033] ;

[0034] In the formula is the data rate, r is the distance between the transmitter and the receiver, is the observation interval.

[0035] Furthermore, AAR is the and perception range The ratio of the number of sensed vehicles to all vehicles is:

[0036] ;

[0037] in, r is the distance between the transmitter and the receiver, is the observation interval.

[0038] Compared with the prior art, the vehicle density estimation method based on perception message broadcasting and probability derivation of the present invention has the following advantages:

[0039] (1) The vehicle density estimation method based on perception message broadcasting and probability derivation described in the present invention, by introducing a probability derivation model, comprehensively considers the message loss factor, significantly improves the accuracy of vehicle density estimation, can perform more robustly in complex and changeable traffic environments, and can estimate vehicle density in real time and accurately.

[0040] (2) The vehicle density estimation method based on perception message broadcasting and probability derivation described in the present invention only uses the original perception message for estimation without adding additional fields. It is easy to implement and promote. It can not only provide reliable data support for network optimization and congestion control, but also improve the reliability of vehicle communication and the safety of real-time traffic. It provides a more reliable and efficient vehicle density estimation solution for intelligent transportation systems, especially suitable for highway application scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The accompanying drawings constituting a part of the present invention are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the accompanying drawings:

[0042] Figure 1 This is a logical block diagram of a vehicle density estimation method based on perception message broadcasting and probability derivation according to an embodiment of the present invention. DETAILED DESCRIPTION

[0043] In order to make the technical means, objectives and effects of the present invention easy to understand, the embodiments of the present invention are described in detail below with reference to specific drawings.

[0044] It should be noted that all the terms used in the present invention for directional and positional indications, such as "upper", "lower", "left", "right", "front", "back", "vertical", "horizontal", "inside", "outside", "top", "lower", "lateral", "longitudinal", "center", etc., are only used to explain the relative positional relationship and connection status between the components in a certain state (as shown in the accompanying drawings), and are only for the convenience of describing the present invention, rather than requiring the present invention to be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the descriptions of "first", "second", etc. in the present invention are only used for descriptive purposes, and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features.

[0045] In the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0046] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0047] In intelligent transportation systems (ITS), vehicle density estimation is a key step in improving traffic management efficiency and driving safety. Traditional infrastructure-based methods, such as using cameras and sensors to collect traffic information, are effective but have limitations, such as limited coverage, high maintenance costs, and poor real-time performance. Factors such as weather and infrastructure location may also affect the estimation accuracy. With the development of vehicular ad hoc networks (VANETs), vehicles (also called observers) have become O ) can estimate the local vehicle density based on the received perception messages. These messages are regularly exchanged between vehicles and include information such as location, speed, and driving direction. Vehicles can directly communicate and exchange traffic status information without relying on fixed infrastructure, which improves the flexibility and real-time performance of density estimation. However, the communication channels in VANET are susceptible to interference, which leads to the loss of perception messages, thus affecting the accuracy of density estimation.

[0048] To address the above issues, this patent proposes an innovative vehicle density estimation method based on perception message broadcasting and probability derivation, which includes the following steps:

[0049] S101: For any observation node O , initialize the observation interval , data rate , Sending range R , Observation distance increment , observation distance and other parameters;

[0050] S102: Observation Node O Collect the perception messages sent by neighboring nodes and analyze each neighboring node based on the received perception messages i the average position and number of received sensing messages;

[0051] S103: Calculate observation nodes O Received from neighboring nodes i The acceptance rate of data packets;

[0052] Among them, the observation node O According to the number of received perception messages and the total number of sent perception messages, the number of observation nodes is approximately calculated. O Received from neighboring nodes i The acceptance rate of packets , that is, the probability of receiving a point-to-point data packet:

[0053] ;

[0054] in For the observation node in the observation period Successfully received i The number of packets sent, Is a node i During the observation period The total number of packets sent within is the data rate, i.e. the perceived message rate, the data rate It is a variable parameter, indicating how many times the message can be sent per second, in Hz. This parameter is configurable, and the default value is 10, but it can also be adjusted to other integer values.

[0055] S104: For the k vehicle nodes, calculate this k The average PRP of the nodes is taken as the observation distance r PRP at

[0056] S105: Get mPRP at the observation distance, and the corresponding observation distance m The PRP function that changes with the receiving distance is obtained by linear fitting using the least squares method.

[0057] S106: Evaluate the application layer awareness probability NAP according to the fitted PRP function, where NAP (Node Awareness Probability) refers to the probability that the receiver successfully receives at least one data packet sent by the sender within the observation period;

[0058] S107: Calculate the average awareness ratio AAR of the perceived vehicles according to the perception probability NAP. The average awareness ratio (Average Awareness Ratio, AAR) refers to the ratio of the number of neighbors perceived by the marked node to the total number of neighbors.

[0059] S108: Estimate the vehicle density based on the number of sensed vehicles and the sensed vehicle ratio AAR. The vehicle density is calculated using the following formula: :

[0060] ;

[0061] in, , is the transmission range, i.e. the transmission range of the vehicle, Refers to the number of perceived vehicles.

[0062] The vehicle density estimation method based on perception message broadcast and probability derivation disclosed in the present application is to use the perception messages (vehicle safety messages containing information such as speed, position and driving direction) broadcast periodically by vehicles in the vehicle ad hoc network to perform density estimation, and observe the nodes O By receiving neighbor nodes i Perception message, first analyze each neighbor node i The average position of the vehicle and the number of messages received are used to calculate the point-to-point packet reception probability (PRP). Subsequently, the PRP at different observation distances is smoothed and polynomially fitted to obtain a PRP function that varies with the receiving distance. Based on this function, the node perception rate (NAP) and the average perception ratio (AAR) are calculated by probability deduction. Finally, the number of vehicles sensed is combined with the The whole process does not require additional communication overhead and only relies on the received perception messages. The impact of message loss on density estimation is considered through a probabilistic model.

[0063] Compared with the traditional method that relies on the number of single perception messages, the vehicle density estimation method based on perception message broadcasting and probability derivation described in this application introduces a probability derivation model, comprehensively considers the message loss factor, and significantly improves the accuracy of vehicle density estimation. It can be more robust in complex and changeable traffic environments and can estimate vehicle density in real time and accurately. In addition, the vehicle density estimation method based on perception message broadcasting and probability derivation described in this application only uses the original perception message for estimation, without adding additional fields, and is easy to implement and promote. It can not only provide reliable data support for network optimization and congestion control, but also improve the reliability of vehicle communications and the safety of real-time traffic, and provide a more reliable and efficient vehicle density estimation solution for intelligent transportation systems, especially suitable for highway application scenarios.

[0064] In the example of the present application, in step S101, the observation distance parameter includes a set of observation distances: , the observation distance is the given receiving distance, where j ( ) is the index of the observation distance, , .

[0065] This setting sets a receiving distance range, that is, the sending range R , and introduced the observation distance increment As a benchmark, then, using the index j (from 1 to n ) to determine each observation distance one by one ,in equal j Multiply half of , that is, each observation distance is incremental and evenly spaced. At the same time, by calculating R Divide by The integer part of the observation distance is used to determine the total number of n , ensuring that the entire distance range is reasonably divided and the observation distance is accurately set, covering the entire required range while maintaining an appropriate interval, so that the system can more carefully capture the characteristic changes of the target at different distances, thereby improving the accuracy and reliability of the data and laying a solid foundation for subsequent measurement and analysis. This setting optimizes the measurement process, simplifies the data processing process, reduces the errors caused by improper distance setting, and enhances the flexibility and adaptability of the system.

[0066] In the example of the present application, in step S102, the observation node O Neighbor nodes that receive at least one sensing message are numbered in sequence. i ( i is an integer greater than or equal to 1) indicates that each neighbor node is analyzed based on the perception messages received during the observation interval. i The average position and the average position from the node i The number of perception messages.

[0067] In step S102, observe the node O The observation node periodically broadcasts the perception messages to the surroundings and also receives the periodic perception messages sent by neighboring nodes. For each neighboring node that receives at least one perception message, the observation node will number it and use the number i To identify these neighbor nodes. The observation node analyzes the messages received from each neighbor node within the preset observation interval. i The perception messages are calculated and the average value of the vehicle positions contained in these messages is used as the neighbor node i The average position of each neighbor node is recorded at the same time i The number of perception messages sent by each neighbor node can be recorded so that these data can be used to calculate the probability of receiving point-to-point data packets later. This setting can estimate the location of neighbor nodes more accurately, and taking into account the frequency differences in the sending of perception messages by different neighbor nodes, it helps to reduce the problem of inaccurate position estimation caused by single message errors and improves the reliability of position information. In addition, recording the number of perception messages from each neighbor node provides key data support for the subsequent calculation of the probability of receiving point-to-point data packets, making the entire vehicle density estimation process more scientific and accurate.

[0068] Preferably, in step S102, the observation node O Neighboring nodes that receive multiple perception messages i , calculate the observation node O With neighboring nodes i The receiving distance between the neighbor nodes is used to correspond to different observation distances according to the receiving distance. O Successfully received node during the observation period i The number of packets is ,So The average value of the positions contained in the perception message packets is taken as the neighbor node i location.

[0069] This setting makes it possible for the observation node to O Received from the same neighbor node i When multiple sensing messages are received, the computer automatically measures and calculates the number of observation nodes. O With the neighbor nodei According to the preset observation distance range, the observation node will classify the neighbor node into the corresponding observation distance interval, so as to more carefully depict the actual distribution of vehicles on the road. This not only helps to reduce the error caused by inaccurate distance estimation, but also provides more accurate data support for subsequent traffic flow analysis, congestion warning and other applications, thereby optimizing traffic management and improving road traffic efficiency.

[0070] As a preferred example of the present application, in step S104, the observation node is calculated O With adjacent nodes i Receiving distance between , will have the same index j = The vehicle nodes are grouped into the same list, where The observation distance increment is preset. The relationship between the observation distance and the receiving distance is:

[0071] ;

[0072] Observation distance r The calculation formula of PRP at is:

[0073] ;

[0074] in, For the j Observation distance, Neighbor node i ( i is an integer greater than or equal to 1) and the observation node O The distance between is the increment of observation distance.

[0075] The setup first accurately measures the observation nodes O Its neighboring nodes i Receiving distance between , and then increment according to the preset observation distance , the vehicle nodes are divided into receiving distance Observation distance (satisfy The above setting introduces the observation distance increment , effectively refined the grouping criteria, reduced the misgrouping caused by small distance differences, and at the same time, calculated the specific observation distance rWhen calculating the performance evaluation indicator PRP at the node, the method of averaging the PRP values ​​of multiple nodes is adopted, which effectively weakens the impact of the measurement error of a single node on the overall evaluation result, thereby enhancing the stability and reliability of the evaluation.

[0076] As a preferred example of the present application, in step S105, each observation distance and the PRP under the observation distance constitute a sample point , a total of m There are sample points, because there are no neighbor nodes within a certain distance, so m≤ n , for this m The sample points are first smoothed and denoised using a linear smoothing filter (such as a Savitzky-Golay filter), and then a polynomial fitting is performed to obtain a PRP function that varies with the receiving distance. The order of the polynomial fitting is based on the standard of being less than the predefined residual square sum.

[0077] In the example of this application, the fitted PRP function is expressed as , the residual sum of squares S It is expressed as:

[0078] ;

[0079] in For observation distance.

[0080] In this application, in view of the association between observation distance and PRP value, each observation distance and its corresponding PRP value are combined into sample points, and then these sample points are finely smoothed and denoised using the Savitzky-Golay filter to eliminate random fluctuations in the data. Next, a PRP function model that changes with the observation distance is constructed through polynomial fitting technology. The application of the polynomial fitting method further ensures that the PRP function model can smoothly and accurately reflect the dynamic relationship between the receiving distance and the PRP value. More importantly, by setting the residual sum of squares S The threshold is used to control the order of the polynomial fitting, which ensures the fitting accuracy while avoiding the risk of overfitting, thereby enhancing the overall stability and prediction ability of the system.

[0081] As a preferred example of the present application, in step S106, NAP is calculated using a probability derivation method:

[0082] ;

[0083] In the formula is the data rate, r is the distance between the transmitter and the receiver, is the observation interval.

[0084] As a preferred example of this application, AAR is and perception range The ratio of the number of sensed vehicles to all vehicles in the .

[0085] ;

[0086] in, r is the distance between the transmitter and the receiver, is the observation interval.

[0087] In this example, is the assumed vehicle density, which is eliminated during the calculation process and refers to the vehicle density (vels / m).

[0088] Through innovative probability derivation methods, the key parameters such as the sensing message rate, the distance between the sender and the receiver, and the observation interval are cleverly integrated to build an efficient calculation model to accurately calculate the location-related probability in network-assisted positioning (NAP). This setting can more accurately reflect the actual location information of the vehicle or device through refined probability derivation, effectively reducing the positioning error. At the same time, the statistical analysis of the number of vehicles in the sensing cycle not only enhances the system's adaptability to different environments, but also improves its positioning ability in complex scenarios.

[0089] The vehicle density estimation method based on perception message broadcasting and probability derivation disclosed in the present application, first, the vehicle nodes in the vehicle-mounted ad hoc network (VANET) periodically broadcast perception messages containing information such as vehicle speed and position, so as to realize real-time sharing of information between vehicles; then, the probability derivation method is used to combine the number of perception messages received and the total number of messages sent to calculate the point-to-point packet reception probability (PRP), and the PRP function that varies with the receiving distance is obtained by polynomial fitting; based on this function, the point-to-point node perception rate (NAP) and the average perception ratio (AAR) are further evaluated to accurately estimate the vehicle density. The vehicle density estimation method based on perception message broadcasting and probability derivation described in the present application, by introducing probability derivation and polynomial fitting methods, can more accurately calculate the communication probability and vehicle density between vehicles, and can maintain a high estimation accuracy even in a complex and changeable traffic environment, significantly improve the accuracy and communication efficiency of vehicle density estimation in many aspects, and dynamically adjust the strategy of the minimum contention window value, effectively alleviate the problem of broadcast frame collision, especially when the vehicle density is high, significantly reduce channel congestion, and improve the successful sending rate and receiving rate of broadcast information. In addition, the rapid broadcast method based on location information further shortens the information propagation delay, ensures the real-time nature of the information, and provides drivers with more timely feedback on traffic conditions, which helps improve road safety and driving efficiency, and is particularly suitable for highway scenarios.

[0090] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A vehicle density estimation method based on perception message broadcasting and probability derivation, characterized in that: The steps include: S101: For any observation node O , initialize the observation interval , data rate , Sending range R , Observation distance increment , observation distance parameters; S102: Observation Node O Collect the perception messages sent by neighboring nodes and analyze each neighboring node based on the received perception messages i the average position and number of received sensing messages; S103: Calculate observation nodes O Received from neighboring nodes i The acceptance rate of packets , , in For the observation node in the observation period Successfully received i The number of packets sent, For Node i During the observation period The total number of packets sent within S104: For the k vehicle nodes, calculate this k The average PRP of the nodes is taken as the observation distance r PRP at S105: Get m PRP at the observation distance, and the corresponding observation distance m The PRP function that changes with the receiving distance is obtained by linear fitting using the least squares method. S106: Evaluate the application layer perception probability NAP according to the fitted PRP function, where NAP is the probability that the receiver successfully receives at least one data packet sent by the sender within the observation period; When fitting the PRP function, the polynomial fitting order is smaller than the predefined residual square sum. The fitted PRP function is expressed as , the residual sum of squares S It is expressed as: ； Observation distance r The calculation formula of PRP at is: ； in, For the j Observation distance, Neighbor node i With the observation node O The distance between is the observation distance increment; NAP is calculated using a probabilistic approach: ； In the formula is the data rate, r is the distance between the transmitter and the receiver, is the observation interval; S107: Calculate the average perception ratio AAR of the perceived vehicles according to the perception probability NAP. The average perception ratio AAR is the ratio of the number of neighbors perceived by the marked node to the total number of neighbors. AAR is the ratio of the number of neighbors perceived by the marked node to the total number of neighbors during the perception period. and perception range The ratio of the number of sensed vehicles to all vehicles is: ； S108: Estimate the vehicle density based on the number of sensed vehicles and the sensed vehicle ratio AAR. The vehicle density is calculated using the following formula: : ； in, , is the sending range, Refers to the number of perceived vehicles.

2. The vehicle density estimation method based on perception message broadcasting and probability derivation according to claim 1 is characterized in that: In step S101, the observation distance parameter includes a set of observation distances , the observation distance is the given receiving distance, where j ( ) is the index of the observation distance, , .

3. The vehicle density estimation method based on perception message broadcasting and probability derivation according to claim 1 or 2, characterized in that: In step S102, observe the node O Neighbor nodes that receive at least one sensing message are numbered in sequence. i It means that each neighbor node is analyzed according to the perception message received during the observation interval. i The average position and the average position from the node i The number of perception messages.

4. The vehicle density estimation method based on perception message broadcasting and probability derivation according to claim 3 is characterized in that: In step S102, observe the node O Neighboring nodes that receive multiple perception messages i , calculate the observation node O With neighboring nodes i The receiving distance between them is used to correspond the neighboring nodes to different observation distances according to the receiving distance.

5. The vehicle density estimation method based on perception message broadcasting and probability derivation according to claim 4 is characterized in that: In step S104, the observation node is calculated O With adjacent nodes i Receiving distance between , will have the same index j = The vehicle nodes are grouped into the same list, and the relationship between the observation distance and the receiving distance is: 。 6. The vehicle density estimation method based on perception message broadcasting and probability derivation according to claim 5 is characterized in that: In step S105, each observation distance and the PRP under the observation distance constitute a sample point , a total of m sample points, among which, m≤n ,right m The sample points are smoothed and denoised using a linear smoothing filter, and then a polynomial fitting is performed to obtain the PRP function that varies with the receiving distance.

Images