Vehicle driving track deduction method for large-area monitoring blind area

Abstract

The invention provides a vehicle driving track deduction method for a large-area monitoring blind area, and belongs to the technical field of intelligent traffic, and the method comprises the steps: firstly positioning a starting point road section and a terminal point road section of a target vehicle, and then determining a structure of a Bayesian network by referring to a traffic road network topological relation; in combination with real-time traffic data such as vehicle speed, traffic flow and road width, influence factors are quantified, probability values of each father node and each child node are calculated, and then Bayesian network parameters are obtained. Then, effective connection of road sections in the road network structure is achieved by selecting the road sections corresponding to the nodes with the high probability values, and finally target vehicle driving track deduction from the starting point to the ending point is completed. According to the method, the real driving track of the vehicle in the large-area monitoring blind area is highly restored, and the collaboration degree of various traffic data is improved.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent transportation technology, and in particular relates to a vehicle driving trajectory deduction method for large-area monitoring blind spots. Background Art

[0002] With the advancement of smart transportation construction, a variety of advanced traffic perception technologies have been introduced and their layout density has been increased, providing technical support for the effective collection of dynamic targets. However, with the acceleration of urbanization, the structure of the traffic network has become increasingly complex. Although smart transportation construction has achieved many results, it is still difficult to achieve comprehensive coverage and accurate perception of the urban road network. In addition, the complex traffic environment and diverse driver intentions have led to a significant increase in vehicle driving paths in the traffic network, making it particularly difficult to accurately reproduce the driving trajectory of the target vehicle in a large-scale complex road network.

[0003] The prior art involves a vehicle driving trajectory deduction technology based on a vehicle re-identification algorithm. The technical solution first collects video data captured by a surveillance camera at a traffic intersection and obtains the corresponding traffic network data; then detects the vehicle and extracts vehicle information through a target detection model based on deep learning, and then uses a re-identification algorithm to detect the same vehicle that appears in different surveillance videos; then maps the position of the vehicle in the surveillance image to the traffic network, uses the shortest path algorithm and obtains the optimal path between two adjacent positions in time series, connects them in sequence to form a complete driving trajectory, and finally visualizes the tracking trajectory. The disadvantages are as follows: The vehicle trajectory deduction technology based on the vehicle re-identification algorithm is suitable for areas where surveillance cameras are densely deployed. When a vehicle travels to a traffic area with a low density of surveillance cameras and a complex road network structure, it will be extremely difficult to use the above technical solution to deduce the vehicle's real and continuous driving trajectory.

[0004] The prior art involves the dynamic target motion trajectory deduction technology based on Bayesian network. The Bayesian network model consists of nodes, directed edges and network parameters. In the path deduction technology, the directed edges represent the connection relationship between the various sections (from the starting section node to the section node connected to it), and the network parameters represent the conditional probability of selecting different sections. The inference result is used to describe the possible movement direction of the target at the next moment. It has the following disadvantages: The Bayesian network solution applied to vehicle trajectory deduction adopts a single influencing factor quantification method in the parameter solution process and ignores the influence degree of each factor, making it unable to reasonably reflect the driver's decision intention, resulting in low accuracy of this technical solution. Summary of the invention

[0005] In view of the above-mentioned deficiencies in the prior art, the present invention provides a vehicle driving trajectory deduction method for large-area monitoring blind spots, which solves the following problems: optimizes the deficiency of the target vehicle re-identification method in the prior art that cannot cope with large-area monitoring blind spots; improves the existing Bayesian network deduction method, performs vehicle driving trajectory deduction, and realizes continuous deduction and prediction of the trajectory of target vehicles in large-area monitoring blind spots.

[0006] In order to achieve the above objectives, the technical solution adopted by the present invention is: a vehicle driving trajectory deduction method for a large-area monitoring blind spot, comprising the following steps: S1. Obtaining a traffic information dataset of a large monitoring blind area; S2. Based on the traffic information dataset, construct the structure of the Bayesian network model; S3, determine the impact quantification factor of the Bayesian network model; S4. Using the weight coefficient estimation process, estimate the influence of the quantitative factors in different time periods to obtain the weight coefficients; S5. Calculate the weighted Bayesian network model parameters according to the influencing quantification factors and weight coefficients to obtain the vehicle driving trajectory deduction results.

[0007] The beneficial effects of the present invention are as follows: the present invention first locates the starting and ending sections of the target vehicle, and then establishes the structure of the Bayesian network with reference to the topological relationship of the traffic network. Combined with real-time traffic data such as vehicle speed, traffic volume, road width, and quantified influencing factors, the probability values ​​of each parent node and child node are calculated to obtain the Bayesian network parameters. Subsequently, by selecting the sections corresponding to the nodes with higher probability values, the effective connection of the sections in the road network structure is achieved, and finally the driving trajectory of the target vehicle from the starting point to the end point is deduced. The present invention highly restores the actual driving trajectory of the vehicle and improves the coordination degree of various types of traffic data.

[0008] Furthermore, the S1 is specifically: Based on the large-scale monitoring blind area, the traffic network is divided into multiple independent sections and numbered; Extract the length and width of each traffic section; The traffic flow and average vehicle speed of each road section at different time periods are recorded using surveillance videos to construct a traffic information dataset.

[0009] The beneficial effect of the above further scheme is: the above scheme constructs a traffic information data set, including independent road sections divided based on traffic intersections and road section ends, recording the length, width, traffic flow, average vehicle speed, etc. of each road section, providing data support for constructing the Bayesian network model structure and quantifying influencing factors.

[0010] Furthermore, the S2 is specifically: Based on the road network topology structure in the traffic information data set, the starting section and the end section are used as the starting node and the target node of the Bayesian network respectively, and the nodes except the starting node and the target node are used as the intermediate nodes; Based on the connection relationship between the intermediate nodes assigned by the vehicle driving direction on each road section, the Bayesian network model structure is constructed.

[0011] The beneficial effect of the above further scheme is that the above scheme directly maps the road network topology relationship into the Bayesian network structure, so that the Bayesian network structure and the actual road network are mutually mapped, which can intuitively reflect the driving trajectory of vehicles in the road network and facilitate the understanding and analysis of traffic flow changes. At the same time, the above scheme can effectively avoid the defects of traditional structural learning algorithms such as high computational complexity and lack of prior knowledge when constructing the Bayesian network structure.

[0012] Furthermore, the impact quantification factors include: Road Grade:

[0013] in, and Respectively represent i The width of the road segment before and after normalization, and Respectively represent the maximum and minimum widths of all road sections within the measurement area; Congestion level:

[0014]

[0015] in, represents the normalized calculation of the road congestion index, , Respectively The endpoint of the traffic congestion determination interval, It represents the average driving speed of the road section. Indicates the number of measured vehicles, Indicates the length of the road section. Indicates vehicle Time to pass through the section; Motion Vector:

[0016] in, and Represents the motion vector of a road segment before and after normalization; Vehicle density:

[0017] in, and Respectively represent the road intersection The vehicle density values ​​before and after normalization on the road segment, and They respectively represent the maximum and minimum vehicle density in all connected sections of the same road intersection.

[0018] The beneficial effect of the above further scheme is that the above scheme estimates the influencing factors that affect the driver's decision using a mathematical model, and the larger the value of the influencing factor, the higher the possibility that the driver will choose the road section. Compared with the traditional hierarchical unified estimation scheme, the above scheme uses corresponding mathematical models to estimate different influencing factors, which can more accurately measure the driver's decision.

[0019] Furthermore, the S4 comprises the following steps: S401, dividing the vehicle operation time into morning peak, afternoon peak, evening peak and non-peak periods; S402, initializing the weight coefficients of the quantization factors affecting each time period; S403, calculating weighted Bayesian network parameters based on the impact quantification factors and the weight coefficients corresponding to the vehicle operation time periods; S404, performing Bayesian network deduction based on the constructed Bayesian network model structure and weighted Bayesian network parameters; S405. Determine whether the deduced path is consistent with the actual path. If so, obtain the optimal weight coefficient for the time period. Otherwise, adjust the weight coefficient affecting the quantization factor for the time period and return to S403.

[0020] The beneficial effects of the above further scheme are: the above scheme quantitatively analyzes the influence of different influencing factors on driving decisions and improves the coordination of various traffic data. At the same time, the optimized Bayesian network model improves the model accuracy and enhances the prediction ability of driver behavior decision-making.

[0021] Furthermore, the S5 is specifically: Substitute the influencing quantization factor and weight coefficient into the expression of weighted Bayesian network parameters to solve the network parameters and obtain the selection probability of the road section corresponding to each node; According to the structure of the Bayesian network model, the road sections with the highest probability of being selected are connected in sequence to obtain the vehicle driving trajectory deduction result.

[0022] The beneficial effect of the above further scheme is that the vehicle driving trajectory deduction result obtained by the above scheme comprehensively considers various influencing quantitative factors such as traffic conditions, road attributes, traffic flow, etc. Compared with traditional deduction methods, it can reflect the vehicle's driving path selection in actual traffic scenarios and more accurately restore the vehicle's driving trajectory under different traffic conditions, providing a powerful tool for traffic management and analysis.

[0023] Furthermore, the expression of the selected probability is as follows:

[0024] in, Indicates selection i The probability value of the road segment, , , , and Both said , , , and The weight coefficient that affects the quantization factor, Indicates i Factors other than the quantitative factors that affect the road section.

[0025] The beneficial effect of the above further scheme is: the above scheme is the core of the weighted Bayesian network parameter calculation, and introduces weight coefficients on the basis of the traditional Bayesian network model, which effectively improves the model accuracy and the synergistic relationship of various influencing quantitative factors. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 The present invention is a flow chart of the method.

[0027] Figure 2 This is a schematic diagram of the experimental measurement area and section numbering.

[0028] Figure 3 Schematic diagram of the road network structure and its corresponding Bayesian network.

[0029] Figure 4 Schematic diagram of the deduction results using this invention. DETAILED DESCRIPTION

[0030] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0031] Example like Figure 1 As shown, the present invention provides a vehicle driving trajectory deduction method for a large-area monitoring blind spot, and its implementation method is as follows: S1. Obtain a traffic information dataset of a large-area monitoring blind spot, which is specifically: Based on the large-scale monitoring blind area, the traffic network is divided into multiple independent sections and numbered; Extract the length and width of each traffic section; The traffic flow and average vehicle speed of each road section at different time periods are recorded using surveillance videos to construct a traffic information dataset.

[0032] S2. Based on the traffic information data set, the structure of the Bayesian network model is constructed, which is specifically: Based on the road network topology structure in the traffic information data set, the starting section and the end section are used as the starting node and the target node of the Bayesian network respectively, and the nodes except the starting node and the target node are used as the intermediate nodes; Based on the connection relationship between the intermediate nodes assigned by the vehicle driving direction on each road section, the Bayesian network model structure is constructed.

[0033] S3, determine the impact quantification factor of the Bayesian network model; In this embodiment, the influencing quantification factors constitute the key parameters in the Bayesian network model. Based on the behavior patterns of conventional drivers, the present invention selects factors such as road grade, congestion level, motion vector and vehicle density as key influencing factors for drivers to choose the optimal road section in an urban traffic environment.

[0034] (1) Road grade: Select road section width data As the basis for judging the road grade, the road grade of the present invention is obtained by normalizing the road section width using formula (1): .when The larger it is, the wider the road is, and the greater the probability that the driver prefers this road section.

[0035] (1) in, and Respectively represent i The width of the road segment before and after normalization, and They respectively represent the maximum and minimum widths of all road sections within the measurement area.

[0036] (2) Congestion level: Congestion index It is an important indicator that comprehensively reflects road congestion. Its value range is 0-10. The higher the value, the more serious the traffic congestion. Therefore, the congestion index selected in this invention is As the basis for judging the degree of congestion. The average driving speed of the road section is calculated using formula (2), and the urban traffic congestion degree and congestion index are obtained by comparing with Table 1. Table 1 is the corresponding relationship between the average driving speed of urban road sections and the traffic congestion index.

[0037] (2) in, It represents the average driving speed of the road section. Indicates the number of measured vehicles, Indicates the length of the road section. Indicates vehicle The time it takes to pass through the section.

[0038] Table 1

[0039] The present invention uses formula (3) to realize normalized calculation of road congestion index based on the average driving speed of the road section and the traffic congestion determination interval in which it is located. The specific calculation is shown in formula 3: (3) in, represents the normalized calculation of the road congestion index, , Respectively The endpoint of the traffic congestion determination interval, It represents the average driving speed of the road section. Indicates the number of measured vehicles, Indicates the length of the road section. Indicates vehicle Time to pass through the section; The larger the value, the lower the congestion level, and the greater the probability that the driver will prefer this road section. Set to 1.

[0040] (3) Motion vector: At a road intersection, when drivers choose the next driving section, they usually prefer to choose a route close to the destination. In order to evaluate the correlation between the vehicle's driving direction and the destination direction, the present invention uses the concept of "motion vector" to analyze whether the vehicle is moving towards the destination. By quantifying the angle between the vehicle's driving direction and the target direction (the direction of the line connecting the starting point and the destination), the motion vector is used to analyze whether the vehicle is moving towards the destination. (in ), can effectively judge the driving intention of the vehicle. When it is smaller, it indicates that the target vehicle is closer to the destination; otherwise, it indicates that the vehicle may deviate from the destination or take a detour.

[0041] According to the inverse relationship between the motion direction angle and the driving intention, the present invention uses a cosine function to normalize the angle to obtain the normalized motion vector of a certain road section. , see formula (4) for the specific method. When it is close to 0, it indicates that the road section selected by the target vehicle deviates significantly from the destination direction; When it is close to 1, it means that the target vehicle is moving closer to the destination.

[0042] (4) in, and Respectively represent the motion vector of a road section before and after normalization.

[0043] (4) Vehicle density: Vehicle density refers to the density of vehicles on a lane. Generally speaking, drivers tend to choose roads with low vehicle density, which not only helps to improve driving comfort, but also provides more space to deal with danger in an emergency, and can effectively avoid road congestion during peak hours. See formula (2). Vehicle density can be calculated by the ratio of the traffic flow of a known road section to the average driving speed of the section.

[0044] At a road intersection, drivers usually make a quick decision on the driving route by comparing the vehicle density values ​​of different road sections. In order to reflect this decision-making process, the present invention first calculates the vehicle density values ​​of each road section connected to the intersection. Then, the vehicle density of different road sections at the same intersection is normalized, as shown in formula (5): (5) in, and Respectively represent the road intersection The vehicle density values ​​before and after normalization on the road segment, and They represent the maximum and minimum vehicle density in all connected sections of the same road intersection. The closer it is to 1, the more drivers are willing to choose this road section.

[0045] S4. Use the weight coefficient estimation process to estimate the influence of the quantitative factors in different time periods and obtain the weight coefficient. The implementation method is as follows: S401, dividing the vehicle operation time into morning peak, afternoon peak, evening peak and non-peak periods; S402, initializing the weight coefficients of the quantization factors affecting each time period; S403, calculating weighted Bayesian network parameters based on the impact quantification factors and the weight coefficients corresponding to the vehicle operation time periods; S404, performing Bayesian network deduction based on the constructed Bayesian network model structure and weighted Bayesian network parameters; S405. Determine whether the deduced path is consistent with the actual path. If so, obtain the optimal weight coefficient for the time period. Otherwise, adjust the weight coefficient affecting the quantization factor for the time period and return to S403.

[0046] S5. Calculate the weighted Bayesian network model parameters according to the influencing quantification factors and weight coefficients to obtain the vehicle driving trajectory deduction results, which are specifically: Substitute the influencing quantization factor and weight coefficient into the expression of weighted Bayesian network parameters to solve the network parameters and obtain the selection probability of the road section corresponding to each node; According to the structure of the Bayesian network model, the road sections with the highest probability of being selected are connected in sequence to obtain the vehicle driving trajectory deduction result.

[0047] In this embodiment, the present invention first locates the starting and ending sections of the target vehicle based on the vehicle re-identification algorithm, and then establishes the structure of the Bayesian network with reference to the topological relationship of the traffic network. Combined with real-time traffic data such as vehicle speed, traffic volume, road width, and based on the quantitative influencing factors, the probability values ​​of each parent node and child node are calculated using formula (6), and the Bayesian network parameters are obtained. Subsequently, by selecting the sections corresponding to the nodes with higher probability values, the effective connection of the sections in the road network structure is achieved, and finally the driving trajectory of the target vehicle from the starting point to the end point is deduced.

[0048] (6) in, i Indicates i road section, , Indicates n road section, Indicates i The number of factors that affect the road section other than the quantitative factor, , , , and Both said , , , and The estimation process of the weight coefficients affecting the quantitative factors refers to the idea of ​​sensitivity analysis and is gradually iteratively optimized during the model deduction process.

[0049] The present invention is subjected to a simulation experiment as follows.

[0050] (1) Data source In this embodiment, the campus road network is selected as the experimental sample area, and a campus road network vector map is constructed based on satellite remote sensing images. The traffic road network in the experimental area is divided into multiple independent sections and numbered based on traffic intersections and road ends. Figure 2 As shown, the black arrow indicates the direction in which the road is passable.

[0051] In this embodiment, the length and width of each traffic section are extracted using high-precision remote sensing images, and the traffic flow and average speed of each section at different time periods are recorded using surveillance video data, so as to construct the traffic information data set required for the experiment in this paper. The data record is taken as an example in Table 1.

[0052] Table 1

[0053] (2) Construction of Bayesian network model structure The experiment is based on the vehicle re-identification technology extracted from the campus road network. For example, a vehicle starts driving westward from section 01 one night and finally leaves southward from section 12. The experiment uses the starting section and the end section as the starting nodes of the Bayesian network. With the target node , and the remaining independent road sections in the experimental area are used as intermediate nodes Based on the driving direction of vehicles on each road section, the connection relationship between each intermediate node is assigned, and the Bayesian network structure constructed is as follows Figure 3 As shown, Figure 3 (a) is a schematic diagram of the road network structure. Figure 3 (b) is a schematic diagram of the Bayesian network structure.

[0054] (3) Construction of Bayesian network model structure The experiment selected four influencing factors, namely road grade, congestion level, motion vector and vehicle density, to reflect the driver's decision intention. Based on the road width, road length, traffic flow and average speed data of the traffic information data set in Table 1, the above influencing factors were quantified by combining the above method to obtain the quantification results of the influencing factors of each node in Table 2.

[0055] Table 2

[0056] (4) Network parameter calculation In this embodiment, in order to measure the influence of different influencing factors on the driver's decision, the weight coefficient estimation process is used to obtain the weight coefficient of the influencing factor for each time period as shown in Table 3, and then the weighted Bayesian network parameter calculation formula for each time period is obtained, as shown in formula (7).

[0057] Table 3

[0058] (7) (5) Derivation results Substitute the quantified impact factor into Equation 7 (late) to solve the network parameters and obtain Figure 3 The probability distribution of the road sections corresponding to each node in (b) is shown in Table 4, which is a table of vehicle driving trajectory deduction results. The road sections with the highest probability of being selected are connected in sequence according to the Bayesian network structure, and the deduction path result of the present invention is obtained as follows: .

[0059] Table 4

[0060] (6) Result verification This experiment collected surveillance video data taken by fixed cameras on campus. There were no blind spots in the surveillance video in the measurement area. Based on the vehicle re-identification technology, the same vehicle appearing in different videos can be extracted, and the video clips of the target vehicle appearing in the surveillance can be recorded in sequence, thus forming a total of 100 complete driving trajectories for deduction experiments and result verification.

[0061] The starting and ending sections of the actual path are set as known conditions, and the sections in the gray area are assumed to be monitoring blind spots. The present invention is used to conduct experiments, and the deduction experimental results are compared with the actual trajectory in the video clip to verify the accuracy of the model results. Taking the above target vehicle as an example, the deduction results obtained based on the present invention are as follows: Figure 4 As shown in the figure. The red arrow is the driving trajectory with the highest probability deduced by the scheme in this paper; the percentage after the road section number is the probability distribution of the road section being selected; the video clip is the verification point of whether the vehicle passes through the road section. Surveillance video recognition, appearing in , , Verification point, finally at the end of the road section The surveillance video recognition and deduction results are completely consistent with the actual path in the video clip.

Claims

1. A vehicle driving trajectory deduction method for large-area monitoring blind spots, characterized in that: The following steps are involved: S1. Obtain a traffic information dataset of a large monitoring blind area; S2. Based on the traffic information dataset, construct the structure of the Bayesian network model; S3, determine the quantitative factors affecting the Bayesian network model; S4, using the weight coefficient estimation process, estimate the influence of the quantitative factors in different time periods to obtain the weight coefficient; S5. Calculate the weighted Bayesian network model parameters according to the influencing quantification factors and weight coefficients to obtain the vehicle driving trajectory deduction results.

2. The vehicle driving trajectory deduction method for large-area monitoring blind spots according to claim 1 is characterized in that: The S1 is specifically: Based on the large-scale monitoring blind area, the traffic network is divided into multiple independent sections and numbered; Extract the length and width of each traffic section; The traffic flow and average vehicle speed of each road section at different time periods are recorded using surveillance videos to construct a traffic information dataset.

3. The vehicle driving trajectory deduction method for large-area monitoring blind spots according to claim 1 is characterized in that: The S2 is specifically: Based on the road network topology structure in the traffic information data set, the starting section and the end section are used as the starting node and the target node of the Bayesian network respectively, and the nodes except the starting node and the target node are used as the intermediate nodes; Based on the connection relationship between the intermediate nodes assigned by the vehicle driving direction on each road section, the Bayesian network model structure is constructed.

4. The vehicle driving trajectory deduction method for large-area monitoring blind spots according to claim 1 is characterized in that: The impact quantification factor include: Road Grade: in, and Respectively represent i The width of the road segment before and after normalization, and Respectively represent the maximum and minimum widths of all road sections within the measurement area; Congestion level: in, represents the normalized calculation of the road congestion index, , Respectively The endpoint of the traffic congestion determination interval, It represents the average driving speed of the road section. Indicates the number of measured vehicles, Indicates the length of the road section. Indicates vehicle Time to pass the section; Motion Vector: in, and Represents the motion vector of a road segment before and after normalization; Vehicle density: in, and Respectively represent the road intersection The vehicle density values ​​before and after normalization on the road segment, and They respectively represent the maximum and minimum vehicle density in all connected sections of the same road intersection.

5. The vehicle driving trajectory deduction method for large-area monitoring blind spots according to claim 4 is characterized in that: The S4 comprises the following steps: S401, dividing the vehicle operation time into morning peak, afternoon peak, evening peak and non-peak periods; S402, initializing the weight coefficients of the quantization factors affecting each time period; S403, calculating weighted Bayesian network parameters based on the impact quantification factors and the weight coefficients corresponding to the vehicle operation time periods; S404, performing Bayesian network deduction based on the constructed Bayesian network model structure and weighted Bayesian network parameters; S405. Determine whether the deduced path is consistent with the actual path. If so, obtain the optimal weight coefficient for the time period. Otherwise, adjust the weight coefficient affecting the quantization factor for the time period and return to S403.

6. The vehicle driving trajectory deduction method for large-area monitoring blind spots according to claim 5 is characterized in that: The S5 is specifically: Substitute the influencing quantization factor and weight coefficient into the expression of weighted Bayesian network parameters to solve the network parameters and obtain the selection probability of the road section corresponding to each node; According to the structure of the Bayesian network model, the road sections with the highest probability of being selected are connected in sequence to obtain the vehicle driving trajectory deduction result.

7. The vehicle driving trajectory deduction method for large-area monitoring blind spots according to claim 6 is characterized in that: The expression of the probability of being selected is as follows: in, Indicates selection i The probability value of the road segment, , , , and Both said , , , and The weight coefficient affecting the quantization factor, Indicates i Factors other than the quantitative factors that affect the road section.

Images