Hybrid traffic state estimation method based on vehicle-road integration

Abstract

The embodiment of the invention discloses a mixed traffic state estimation method based on vehicle-road integration, and relates to the technical field of intelligent traffic control. The method comprises the following steps: respectively acquiring vehicle end measurement data and roadside measurement data of a traffic state of a previous period of an expressway, and substituting the vehicle end measurement data and the roadside measurement data into a vehicle end state observer and a roadside state observer to obtain vehicle end estimation and roadside estimation of the traffic state of the current period; vehicle end estimation and road side estimation are fused to serve as priori estimation of a vehicle end Kalman filter and a road side Kalman filter on the traffic state of the current period, and posteriori estimation is conducted on the traffic state of the current period; fusing the two posteriori estimations to obtain mixed estimation of the traffic state of the current period; and fusing the hybrid estimation with the vehicle end measurement data and the road side measurement data of the traffic state of the current period, substituting the fused data into a state observer to obtain the vehicle end estimation and the road side estimation of the traffic state of the next period, and performing the estimation of the subsequent period. According to the embodiment, the accuracy of traffic state estimation can be improved.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of intelligent traffic control, and in particular, to a method for estimating the mixed traffic state based on vehicle-road integration. Background Art

[0002] With the continuous development of autonomous driving technology, the autonomous ability of vehicles has been gradually improved. Vehicles are gradually used as mobile sensors to collect road section information and obtain more accurate perception information. However, due to the existence of blind spots in the distribution of sensors, it is difficult to accurately and real-time obtain the information of some road sections at present.

[0003] In the prior art, the patent application CN117496706A provides a method and device for designing a mixed traffic flow state estimator in an intelligent connected environment, and the patent application CN113034904A provides a method and device for estimating traffic state based on ETC data. However, these methods do not utilize the information of vehicle-mounted sensors and do not fully consider the natural defects of various state estimators, resulting in limited accuracy of traffic state estimation.

[0004] In view of this, the present invention is proposed. Summary of the Invention

[0005] The embodiments of the present invention provide a method for estimating the mixed traffic state based on vehicle-road integration, which is used to solve the technical problems described in the background art.

[0006] In a first aspect, the embodiments of the present invention provide a method for estimating the mixed traffic state based on vehicle-road integration, including: S10. Respectively obtain the vehicle-end measurement data and roadside measurement data of the traffic state of a highway cycle by a vehicle-end sensor and a roadside sensor; S20. Substitute the vehicle-end measurement data as the vehicle-end measurement output of the previous cycle into a vehicle-end state observer to obtain the vehicle-end estimate of the current cycle traffic state; substitute the roadside measurement data as the roadside measurement output of the previous cycle into a roadside state observer to obtain the roadside estimate of the current cycle traffic state; S30. Fuse the vehicle-end estimate and the roadside estimate as the prior estimate of the current cycle traffic state by a vehicle-end Kalman filter and a roadside Kalman filter, and respectively perform posterior estimation on the current cycle traffic state by the vehicle-end Kalman filter and the roadside Kalman filter; S40. Fuse the posterior estimates of the vehicle-end Kalman filter and the roadside Kalman filter to obtain the mixed estimate of the current cycle traffic state; S50. After fusing the mixed estimation with the on-vehicle measurement data of the current cycle traffic state, use it as the on-vehicle measurement output of the current cycle and substitute it into the on-vehicle state observer to obtain the on-vehicle estimation of the next cycle traffic state; after fusing the mixed estimation with the roadside measurement data of the current cycle traffic state, use it as the roadside measurement output of the current cycle and substitute it into the roadside state observer to obtain the roadside estimation of the next cycle traffic state. S60. Return to S30 according to the on-vehicle estimation and roadside estimation of the next cycle, and repeat the process to obtain the mixed estimation of the traffic state of each cycle in turn.

[0007] In a second aspect, an embodiment of the present invention provides an electronic device, which includes: One or more processors; A memory for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the vehicle-road integrated based hybrid traffic state estimation method described in the first aspect.

[0008] In a third aspect, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the vehicle-road integrated based hybrid traffic state estimation method described in the first aspect.

[0009] In summary, an embodiment of the present invention provides a vehicle-road integrated based hybrid traffic state estimation method, and its beneficial effects are as follows: it can collect traffic information jointly by using on-vehicle sensors and roadside sensors in the mixed traffic flow scenario of highways, and fuse the state observer and the Kalman filter to accurately estimate the real traffic state of highways. In particular, in the fusion of the state observer and the Kalman filter, this embodiment fully considers the problems existing in their state estimation, and overcomes their natural defects through the closed-loop feedback between the state observer and the Kalman filter cycle by cycle, improving the accuracy of state estimation. Specifically: The conventional Kalman filter performs a priori estimation through an open-loop model and cannot correct the a priori estimation by combining actual measurement values, resulting in a large initial estimation error and affecting the estimation accuracy and convergence speed. To address this defect, in this embodiment, the traffic state estimation of the state observer is used as the a priori estimation of the Kalman filter, and the measured information in the state observer is introduced into the Kalman filter, avoiding the error caused by the model a priori estimation. Compared with this embodiment, in patent applications CN117496706A and CN113034904A, the estimated value of the state observer is fused with the model a priori state estimation of the Kalman filter and used as the final pre-estimation value (i.e., the final a priori estimation) of the Kalman filter. The model a priori state estimation value of the Kalman filter is still obtained based on the mixed traffic flow model, and its error is still uncontrollable, interfering with the measured information in the state observer. In this embodiment, the estimation of the state observer after vehicle-road fusion is directly used as the a priori estimation of the Kalman filter, integrating the measured information at the vehicle end and the roadside, which can improve the accuracy of the a priori estimation and the final estimation of the Kalman filter and accelerate the convergence speed.

[0010] When there is a lack of traffic sensors or sensor failures in some sections of the state observer, the measured value is empty, and the lack of observed data limits the accuracy of the state estimation. There are also problems of inaccurate estimation and slow convergence speed. To address this defect, in this embodiment, the hybrid estimation after vehicle-end and roadside fusion is fed back to the two types of state observers as a supplement to the traffic sensor measurement values. In sections where sensor data is missing, the missing measurement values are filled by the hybrid estimation; in sections where sensor data exists, the accuracy of the measurement information is further improved through the fusion of the hybrid estimation and the sensor data.

[0011] In this way, through the closed-loop feedback of the state observer and the Kalman filter in each cycle, the accuracy of the state estimation and the convergence speed are continuously improved, realizing the hybrid traffic state estimation of vehicle-road integration. Brief Description of the Drawings

[0012] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0013] Figure 1 It is a schematic diagram of the cell division of a highway provided by an embodiment of the present invention; Figure 2 It is a schematic diagram of the layout of various sensors on a highway provided by an embodiment of the present invention; Figure 3It is a flowchart of a method for estimating a mixed traffic state based on vehicle-road integration provided by an embodiment of the present invention; Figure 4 It is a schematic structural diagram of an electronic device provided by an embodiment of the present invention; Reference numerals: 1 - Video sensing device, 2 - ETC gantry, 3 - Roadside unit, 4 - Roadside computing facility, 5 - Autonomous vehicle. Specific embodiments

[0014] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope protected by the present invention.

[0015] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus cannot be understood as a limitation of the present invention. In addition, the terms "first", "second", and "third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0016] In the description of the present invention, it should also be noted that 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 or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0017] An embodiment of the present invention provides a method for estimating the mixed traffic state based on vehicle-road integration, which uses vehicle-mounted sensors and roadside sensors to jointly collect road section information, and integrates a state observer and a Kalman filter to jointly estimate the traffic state of highways. To illustrate this method, the mixed state estimator that supports the implementation of this method is described first. In the field of traffic state estimation, both the state observer and the Kalman filter are commonly used state estimators. The former has a relatively simple structure, and the latter has relatively higher accuracy. Therefore, this embodiment combines the advantages of both to construct a mixed state estimator to achieve accurate estimation of the mixed traffic flow state on highways under different penetration rates of autonomous vehicles.

[0018] In a specific embodiment, the construction method of the mixed state estimator may include the following steps: S110, Highway cell division.

[0019] This step divides the highway into several road sections according to the positions of entrance and exit ramps, the positions of curvature radius changes, the positions of lane number changes, etc. Each road section is called a cell, and the divided cells are sequentially numbered to facilitate marking the sensor layout positions. Optionally, combined with Figure 1 , where there are entrance and exit ramps can be used as the boundary between two adjacent cells, and the position of lane number change can also be used as the boundary between two adjacent cells.

[0020] Furthermore, various sensors are usually deployed on highways to measure the traffic state in real time. Taking Figure 2 as an example, these sensors may include: a video sensing device 1 for collecting the traffic volume on the road and automatically converting and calculating the traffic flow density; a roadside unit 3 installed on the ETC gantry 2, and the roadside unit 3 is a device in the ETC system that communicates with the on-vehicle unit to realize vehicle identity recognition. Through this device, information such as vehicle speed and traffic volume can be obtained; and an autonomous vehicle 5 for collecting vehicle speed.

[0021] In addition, roadside computing facilities 4 are also deployed on highways. The roadside computing facilities 4 can store the information collected by different types of sensors, can also execute the operations in the subsequent mixed state observer and Kalman filter, and transmit the final estimation result to the general control center.

[0022] S120, Modeling of highway mixed traffic flow.

[0023] This step is based on the above-divided cells to establish a mixed traffic flow model for the highway. Taking the traffic flow density as the traffic state variable as an example, the specific structure of the model is as follows: (1) Wherein, represents a dimension of The traffic state vector (i.e., the traffic flow density vector), represents the number of cells on the highway. By arranging the traffic flow densities of each cell in the order of cell numbers, we can obtain ; In this embodiment, the traffic state is measured and estimated periodically. and respectively represent the traffic state vectors of the th cycle and the th cycle; represents the system control input with a dimension of , represents the number of external inputs of traffic flow density (for example, if there are 3 vehicle merging entrances on the highway, then ); represents the traffic sensor measurement output vector with a dimension of , represents the number of sensors. By arranging the traffic flow densities measured by each sensor in the order of cell numbers, we can obtain (where cells without sensors do not participate in the arrangement. For example, if there are sensors only in the 1st, 3rd, and 5th cells among cells, then these three data are arranged in the cell order to form a 3 - dimensional vector), represents the measurement output vector of the traffic sensors in the th cycle; represents the system matrix; represents the input matrix; represents the output matrix (its specific value is related to the number and location of the deployed traffic sensors); is a constant matrix; represents the penetration rate of autonomous vehicles in the th cycle, , , and respectively represent the system matrix, input matrix, output matrix, and constant matrix under different penetration rates of autonomous vehicles; represents the system noise in the th cycle, represents the measurement noise in the th cycle. Both the system noise and the measurement noise are Gaussian white noises.

[0024] Furthermore, in this embodiment, the traffic sensors on the highway are divided into two categories: in - vehicle sensors and roadside sensors. Then, there are 2 corresponding output equations, which are specifically as follows: (2) Among them, and are the output matrices corresponding to the vehicle-mounted sensor and the roadside sensor respectively, and the matrix dimension is , represents the measurement output vector of the vehicle-mounted sensor in the th cycle, represents the measurement output vector of the roadside sensor in the th cycle.

[0025] Since the traffic information collected by different types of sensors is different, but it can be uniformly converted into traffic flow density through the conversion relationship between traffic parameters. For example, autonomous vehicles and vehicle-mounted sensors, as mobile sensors, calculate the speed of the cell where they are located, and then convert it into the traffic flow density of the cell as a measurement value. Therefore, the output matrices corresponding to different types of sensors can adopt a unified representation method, and the specific form is as follows: (3) Among them, represents the element corresponding to the th cell in , that is, as long as the corresponding sensor is installed in the th cell, and the sorting of this sensor among the sensors is , then the element in the th row and the th column of is 1, otherwise it is 0; similarly, represents the element corresponding to the th cell in

[0026] S130. Determine the conversion between the measurement value and the traffic flow density.

[0027] First, introduce the conversion between the measurement value of the roadside sensor and the traffic flow density.

[0028] 1) For cells with ETC gantries at both the entrance and the exit: The traffic flow density of the cell can be calculated respectively through the vehicle flow in one cycle collected by the ETC at the entrance and the exit, that is: (4) Among them, represents the traffic flow density of the th cell, T represents the duration of each cycle (unit: second), and respectively represent the vehicle density of the th cell in the th cycle and the th cycle, represents the The length of a cell (unit: meter), represents the traffic flow entering the th cell within one cycle, represents the traffic flow flowing out of the th cell within one cycle.

[0029] 2) For roadside video surveillance devices: One set of high-definition roadside video surveillance devices for highways is deployed per kilometer on average. The traffic flow density of each cell is calculated as follows: (5) Among them, is the traffic flow density obtained by collecting from the video surveillance device located upstream of the th cell, is the traffic flow density obtained by collecting from the video surveillance device located downstream of the th cell.

[0030] The conversion between the measured values of in-vehicle sensors and traffic flow density is described below. For the cells with autonomous driving vehicles, the position and speed information of each vehicle can be obtained, and then the average speed of the cell can be calculated, that is: (6) Among them, represents the average speed of the th cell, b is the number of autonomous driving vehicles in the th cell, is the speed of the th autonomous driving vehicle.

[0031] Then, according to the relationship between speed and density, the traffic flow density of the cell is further calculated.

[0032] S140. Construction of a hybrid state observer.

[0033] After completing the highway modeling and output matrix design, this step constructs the corresponding hybrid state observers for the in-vehicle and roadside respectively. Among them, the hybrid state observer based on in-vehicle sensors (hereinafter referred to as the in-vehicle state observer) is as follows: (7) Among them, is the traffic state vector of the in-vehicle state observer, and respectively represent the traffic state vectors (also known as in-vehicle estimations) of the in-vehicle state observer in the th cycle and the th cycle; is the The output vector of the vehicle-end state observer for one cycle represents the gain matrix of the vehicle-end state observer.

[0034] The hybrid state observer based on roadside sensors (hereinafter referred to as the roadside state observer) is as follows: (8) Among them, represents the traffic state vector of the roadside state observer, and respectively represent the traffic state vectors (also known as roadside estimates) of the roadside state observer in the th cycle and the th cycle; is the output vector of the roadside state observer in the th cycle, represents the gain matrix of the roadside state observer.

[0035] Generally speaking, vehicle-end sensors can obtain accurate information about the cell where they are located, and then accurately estimate the traffic flow density of the current road section through the vehicle-end state observer. However, it is difficult to obtain information beyond the line of sight. Roadside sensors can obtain information about the current cell and transmit the information to vehicles in the upstream road section, thereby making up for the deficiencies of vehicle-end sensors. Combining the advantages of the two types of sensors, in this embodiment, the estimation results of the vehicle-end state observer and the roadside state observer are fused to obtain a more accurate traffic flow density, and the estimation result is transmitted to the upstream road section as the information beyond the line of sight of the upstream road section, so as to accurately implement the control strategy.

[0036] In a specific embodiment, the fusion of the estimation results of the two state observers may include the following steps: Step 1: Assume that there are two weights and for the vehicle-end estimate and the roadside estimate respectively, then a weight combination pair can be generated, and the weighted summation result of the traffic flow density is: (9) Step 2: Calculate the differences and between the vehicle-end estimate and the roadside estimate and respectively: (10) Step 3: Perform loop calculations in each cycle to find the weight combination pair corresponding to the minimum standard deviation as the final weight combination: (11) Step 4: Substitute the final weight combination into formula (9) to calculate the estimated value of traffic flow .

[0037] S150. Construction of the hybrid Kalman filter

[0038] While constructing the hybrid state observer, a hybrid Kalman filter based on vehicle-mounted sensors (hereinafter referred to as the vehicle-mounted Kalman filter) and a hybrid Kalman filter based on roadside sensors (hereinafter referred to as the roadside Kalman filter) can be constructed synchronously. The specific steps are as follows: 1) In a conventional Kalman filter, the traffic state of cells is estimated a priori according to the constructed hybrid traffic flow model. However, since the hybrid traffic flow model itself is an open-loop system and lacks real-time feedback of information, it is prone to large estimation errors caused by external interference. Therefore, in this embodiment, the result after fusing the above vehicle-mounted estimation and roadside estimation is used as the a priori estimation of the Kalman filter, that is: (12) Wherein, represents the a priori estimation of the traffic state of the th cycle by the Kalman filter.

[0039] 2) Calculate the a priori error covariance matrices and of the vehicle-mounted Kalman filter and the roadside Kalman filter respectively: (13) Wherein, represents the posterior error covariance matrix of the vehicle-mounted Kalman filter in the th cycle (derived from the update in the th cycle), represents the a priori error covariance matrix of the vehicle-mounted Kalman filter in the th cycle; represents the posterior error covariance matrix of the roadside Kalman filter in the th cycle, represents the a priori error covariance matrix of the roadside Kalman filter in the th cycle, Q represents the system noise.

[0040] 3) Calculate the gain matrices and of the vehicle-mounted Kalman filter and the roadside Kalman filter respectively: (13-1) Wherein, and represent the gain matrices of the on-vehicle Kalman filter and the roadside Kalman filter at the th cycle, and represent the output matrices of the on-vehicle Kalman filter and the roadside Kalman filter respectively. In this embodiment, , , R represents the measurement noise.

[0041] 4) Update the posterior state estimates respectively according to the prior estimates and of the two types of Kalman filters, the prior error covariance matrices , , and the gain matrices , to obtain the final estimates of the two types of Kalman filters: (14) where, and represent the prior estimate and the posterior estimate of the traffic state of the on-vehicle Kalman filter at the th cycle respectively, represents the on-vehicle measurement output of the on-vehicle Kalman filter at the th cycle; and represent the prior estimate and the posterior estimate of the traffic state of the roadside Kalman filter at the th cycle respectively, represents the roadside measurement output of the roadside Kalman filter at the th cycle. In this embodiment, the measurement outputs of the state observer and the Kalman filter are the same, that is , .

[0042] 5) Update the posterior error covariance matrices and : (15) where, and represent the posterior error covariance matrices of the on-vehicle Kalman filter and the roadside Kalman filter at the th cycle respectively; represents the identity matrix, that is, the diagonal elements are 1 and the other elements are 0.

[0043] To obtain a more accurate estimated density, this embodiment uses a fusion algorithm to perform a weighted sum of the two posterior estimates obtained by the Kalman filter. Specifically as follows: Step 1: Assume that the posterior estimates of the on-vehicle Kalman filter and the posterior estimates of the roadside Kalman filter each have a weight and , then a weight combination pair can be generated, and the weighted summation result of the traffic flow density is: (16) Step 2: Calculate the differences between , and respectively, that is: and , namely: (17) Step 3: Perform loop calculations in each cycle to find the value combination pair corresponding to the minimum value of the standard deviation , as the final weight combination: (18) Step 4: Substitute the final weight combination into formula (16) to calculate the final estimate of the traffic flow in one cycle : (19) This estimated value is the final estimate of the true traffic state of the current cycle by the entire hybrid state estimator. For the sake of easy distinction, this embodiment refers to it as the hybrid estimate.

[0044] In practical applications, the estimation of the hybrid traffic flow density is carried out in chronological order, and the state observer and the Kalman filter need to perform iterative calculations. In order to obtain a more accurate traffic flow density, in subsequent state estimations, the calculation result of formula (19) can be fused with the measurement data collected by the on-vehicle and roadside sensors as the feedback value of the state observer in the next cycle, that is, perform the calculation of the next cycle based on formulas (7) and (8), so as to correct the estimation result of the state observer, and then obtain more accurate state values of all cells. Specifically as follows: The output value corresponding to in formula (19) can be calculated from the output matrix: (20) where represents the measurement output matrix corresponding to formula (19), and this matrix can be designed according to needs to satisfy the corresponding observable judgment matrix ( , ) It suffices to be full rank. For convenience, the identity matrix can be taken.

[0045] Then, equation (20) is respectively combined with the measured values of the vehicle-mounted sensors shown in equation (2) and the measured values of the roadside sensors for fusion calculation to obtain new output values. Optionally, due to the absence of traffic sensors in some cells, two methods are adopted in this embodiment for fusion with the measured values: taking the vehicle-mounted sensors as an example, if the vehicle-mounted measured value of the traffic state in the th cell of the highway in the current cycle is missing (no sensor is deployed, or the sensor fails, etc.), then the output value in of this cell is directly used as the new output value of this cell: if the vehicle-mounted measured value of the traffic state in the th cell of the highway in the current cycle exists, then the output value in of this cell is weighted and averaged with to be used as the new output value of this cell. Finally, the new output values of each cell are arranged in order to obtain the new output vector . The fusion calculation of the roadside sensors is similar.

[0046] Exemplarily, the overall fusion calculation of all cells can be expressed as: (21) And substituting the fused and into equations (7) and (8) to correct the state observer for the next cycle. Among them, the vehicle-mounted state observer is corrected to: (22) where represents the output corresponding to the hybrid estimation in the th cycle, represents the new output after fusing the output corresponding to the hybrid estimation in the th cycle with the vehicle-mounted measurement data; represents the fusion function, such as the fusion operation shown in formula (21), and all the data in this fusion operation are from the same cycle.

[0047] The roadside state observer is corrected to: (23) where represents the The output corresponding to the hybrid estimation of one cycle is fused with the roadside measurement data to obtain a new output. In subsequent cycles, both the vehicle-end estimation and the roadside estimation are based on the state observers shown in Equations (22) and (23). The entire above process represents a complete hybrid state estimator.

[0048] Based on the above hybrid state estimator, Figure 3 is a flowchart of a vehicle-road integrated hybrid traffic state estimation method provided by an embodiment of the present invention. This method is applicable to the hybrid traffic flow scenario of expressways and can be executed by the above roadside computing facilities or other electronic devices. As Figure 3 shown, the method specifically includes: S10. Respectively obtain the vehicle-end measurement data and roadside measurement data of the traffic state on the expressway in the previous cycle from the vehicle-end sensors and roadside sensors.

[0049] Exemplarily, still taking the traffic flow density as the traffic state, various sensors collect the traffic data of each cell on the expressway in real time, and the collected data can be converted into the traffic flow density through Equations (4), (5), (6), etc. Among them, the traffic flow density from the vehicle-end sensors is called the vehicle-end measurement data, and the traffic flow density from the roadside sensors is called the roadside measurement data.

[0050] The state estimation in this embodiment is performed periodically. In each estimation, the currently ongoing cycle is called the current cycle, and the cycle where the previous estimation was located is called the previous cycle. For example, the state estimation is performed every 10 minutes. If the current time is 8:10 and the traffic state between 8:00 and 8:10 is estimated, then the period from 8:00 to 8:10 is called the current cycle, and the period from 7:50 to 8:00 is called the previous cycle.

[0051] In each estimation, it is necessary to obtain the vehicle-end measurement data and roadside measurement data of the previous cycle and the current cycle for estimating the real traffic state of the current cycle.

[0052] S20. Take the vehicle-end measurement data of the previous cycle as the vehicle-end measurement output of the previous cycle, substitute it into the vehicle-end state observer, and obtain the vehicle-end estimation of the traffic state in the current cycle; take the roadside measurement data of the previous cycle as the roadside measurement output of the previous cycle, substitute it into the roadside state observer, and obtain the roadside estimation of the traffic state in the current cycle.

[0053] Specifically, arrange the vehicle-end measurement values in each cell on the expressway in the previous cycle in sequence according to the cell numbers (where the cells with sensors participate in the arrangement, and the cells without sensors do not participate in the arrangement), and the measurement vector of the vehicle-end sensors in the previous cycle can be obtained. Take this vector as Substitute it into Equation (7) for calculation, and the obtained is the vehicle-end estimation of the traffic state in the current cycle.

[0054] Similarly, arrange the roadside measurement values in each cell of the highway in the previous cycle in sequence according to the cell numbers (where the cells with sensors participate in the arrangement, and the cells without sensors do not participate in the arrangement), and the measurement vector of the roadside sensors in the previous cycle can be obtained. Take this vector as Substitute it into formula (8), and the obtained is the roadside estimate of the traffic state in the current cycle.

[0055] S30. Fuse the vehicle-end estimate and the roadside estimate as the prior estimate of the traffic state in the current cycle by the vehicle-end Kalman filter and the roadside Kalman filter, and the vehicle-end Kalman filter and the roadside Kalman filter respectively perform posterior estimation on the traffic state in the current cycle.

[0056] Specifically, substitute the final weight combination determined by formulas (9)-(11) into formula (9) to calculate and The fusion value of . Obtain the Kalman filter in the current cycle through formulas (13)(13-1)(14). Take as and substitute it into formula (14). At the same time, after arranging the vehicle-end measurement values in each cell of the highway in the current cycle in sequence, take it as Substitute it into formula (14), and the calculated is the posterior estimate of the traffic state in the current cycle by the vehicle-end Kalman filter; similarly, take as Substitute it into formula (14). At the same time, after arranging the roadside measurement values in each cell of the highway in the current cycle in sequence, take it as Substitute it into formula (14), and the calculated is the posterior estimate of the traffic state in the current cycle by the roadside Kalman filter.

[0057] S40. Fuse the posterior estimates of the vehicle-end Kalman filter and the roadside Kalman filter to obtain the hybrid estimate of the traffic state in the current cycle.

[0058] Specifically, calculate the fusion result of and according to formula (19) , as the hybrid estimate of the traffic state in the current cycle. Thus, the first estimation of the true traffic state of the highway by the entire hybrid state estimator is completed.

[0059] S50. After fusing the mixed estimation with the on-vehicle measurement data of the current cycle traffic state, use it as the on-vehicle measurement output of the current cycle and substitute it into the on-vehicle state observer to obtain the on-vehicle estimation of the next cycle traffic state; after fusing the mixed estimation with the roadside measurement data of the current cycle traffic state, use it as the roadside measurement output of the current cycle and substitute it into the roadside state observer to obtain the roadside estimation of the next cycle traffic state.

[0060] From this step, enter the next estimation of the hybrid state estimator, that is, estimate the true traffic state of the highway in the next cycle. Specifically, first calculate the corresponding output value according to Equation (20) from the mixed estimation ; At the same time, respectively obtain the on-vehicle measurement data of the on-vehicle sensors for the traffic state of the highway in the next cycle ; and the roadside measurement data of the roadside sensors for the traffic state of the highway in the next cycle ; .

[0061] Then, according to Equation (21), fuse the of each cell with to obtain the new output of the on-vehicle state observer; Substitute as into Equation (22) to obtain the new , that is, the on-vehicle estimation of the next cycle traffic state.

[0062] At the same time, according to Equation (21), fuse the of each cell with to obtain the new output of the roadside state observer. Substitute as into Equation (23) to obtain the new , that is, the roadside estimation of the next cycle traffic state.

[0063] S60. Return to S30 according to the on-vehicle estimation and roadside estimation of the next cycle, and so on in a loop to obtain the mixed estimation of the traffic state of each cycle in turn.

[0064] Specifically, return to S30 and repeat the operations from S30 to S60, that is, fuse the on-vehicle estimation of the next cycle traffic state and the roadside estimation of the next cycle traffic state as the prior estimation of the next cycle traffic state by the on-vehicle Kalman filter and the roadside Kalman filter; the two Kalman filters respectively perform posterior estimation on the traffic state of the next cycle.

[0065] Fuse the posterior estimations of the two Kalman filters to obtain the mixed estimation of the next cycle traffic state. Thus, the current estimation of the hybrid state estimator is completed.

[0066] Then continue with the subsequent estimation of the hybrid state estimator. After fusing the hybrid estimation of the next cycle with the vehicle-side measurement data of the next cycle, it is used as the vehicle-side measurement output of the next cycle and substituted into the vehicle-side state observer to obtain the vehicle-side estimation of the cycle after the next; after fusing the hybrid estimation of the next cycle with the roadside measurement data of the next cycle, it is used as the roadside measurement output of the next cycle and substituted into the roadside state observer to obtain the roadside estimation of the cycle after the next.

[0067] Return to S30 based on the vehicle-side estimation and roadside estimation of the cycle after the next, and repeat the operations from S30 to S60 again. In this way, the hybrid estimations of each cycle are obtained in turn. It can be seen that S10 and S20 are only executed in the first state estimation, while S30 to S60 are executed in the first state estimation and each subsequent state estimation.

[0068] In summary, this embodiment provides a method for estimating hybrid traffic states based on vehicle-road integration. In the scenario of hybrid traffic flow on highways, traffic information is jointly collected using vehicle-side sensors and roadside sensors, and a state observer and a Kalman filter are integrated to estimate the real traffic states of highways. In particular, in the integration of the state observer and the Kalman filter, this embodiment fully considers the problems existing in their state estimations, and overcomes their natural defects and improves the accuracy of state estimation through the closed-loop feedback between the state observer and the Kalman filter on a per-cycle basis. Specifically: The usual Kalman filter performs prior estimation through an open-loop model and cannot correct the prior estimation by combining actual measurement values, resulting in a large initial estimation error and affecting the estimation accuracy and convergence speed. To address this defect, this embodiment uses the traffic state estimation of the state observer as the prior estimation of the Kalman filter, and introduces the measured information in the state observer into the Kalman filter, avoiding the error caused by the model prior estimation. Compared with this embodiment, in Patent Applications CN117496706A and CN113034904A, the estimated value of the state observer is fused with the model prior state estimation of the Kalman filter and used as the final pre-estimation value of the Kalman filter (i.e., the final prior estimation). Among them, the model prior state estimation value of the Kalman filter is still obtained based on the hybrid traffic flow model, and its error is still uncontrollable, interfering with the measured information in the state observer. However, this embodiment directly uses the state observer estimation after vehicle-road integration as the prior estimation of the Kalman filter, integrating the vehicle-side measured information and the roadside measured information, which can improve the accuracy of the prior estimation and the final estimation of the Kalman filter and accelerate the convergence speed.

[0069] When the state observer lacks traffic sensors or the sensors malfunction in some sections, the measured values are empty, and the lack of observed data limits the accuracy of state estimation. There are also problems of inaccurate estimation and slow convergence speed. To address this defect, in this embodiment, the hybrid estimation after fusing the vehicle end and roadside is fed back to the two types of state observers as a supplement to the traffic sensor measured values. In sections where sensor data is missing, the hybrid estimation is used to fill the gap in the measured values; in sections where sensor data exists, the accuracy of the measurement information is further improved through the fusion of the hybrid estimation and the sensor data.

[0070] In this way, through the closed-loop feedback of the state observer and the Kalman filter in each cycle, the accuracy and convergence speed of state estimation are continuously improved, realizing the hybrid traffic state estimation of vehicle-road integration.

[0071] Figure 4 The following is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. As Figure 4 shown, the device includes a processor 60, a memory 61, an input device 62, and an output device 63; the number of processors 60 in the device can be one or more. Figure 4 Here, one processor 60 is taken as an example; the processor 60, the memory 61, the input device 62, and the output device 63 in the device can be connected through a bus or other means. Figure 4 Here, the connection through the bus is taken as an example.

[0072] The memory 61, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and modules, such as program instructions / modules corresponding to the hybrid traffic state estimation method based on vehicle-road integration in the embodiments of the present invention. The processor 60 executes various functional applications and data processing of the device by running the software programs, instructions, and modules stored in the memory 61, that is, implements the above-mentioned hybrid traffic state estimation method based on vehicle-road integration.

[0073] The memory 61 mainly includes a program storage area and a data storage area. Among them, the program storage area can store an operating system and application programs required for at least one function; the data storage area can store data created according to the use of the terminal, etc. In addition, the memory 61 can include a high-speed random access memory, and can also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state storage devices. In some instances, the memory 61 can further include a memory remotely set relative to the processor 60, and these remote memories can be connected to the device through a network. Examples of the above network include but are not limited to the Internet, an enterprise internal network, a local area network, a mobile communication network, and their combinations.

[0074] The input device 62 can be used to receive input numerical or character information and generate key signal inputs related to the user settings and function controls of the device. The output device 63 can include display devices such as a display screen.

[0075] An embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the vehicle-road integrated hybrid traffic state estimation method according to any one of the embodiments.

[0076] The computer storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (non-exhaustive list) of the computer-readable storage medium include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, the computer-readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, apparatus, or device.

[0077] The computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries the computer-readable program code. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium can also be any computer-readable medium other than the computer-readable storage medium, and this computer-readable medium can send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device.

[0078] The program code included on the computer-readable medium can be transmitted by any appropriate medium, including but not limited to wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0079] Computer program code for performing the operations of the present invention may be written in one or more programming languages or combinations thereof, including object-oriented programming languages such as Java, Smalltalk, C++, and also including conventional procedural programming languages such as the C language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0080] Finally, 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 them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the technical solutions of the embodiments of the present invention.

Claims

1. A method for estimating the mixed traffic state based on vehicle-road integration, characterized in that, Including: S10. Respectively obtain the vehicle - end measurement data and roadside measurement data of the traffic state in one cycle on the highway by the vehicle - end sensors and roadside sensors; S20. Substitute the vehicle - end measurement data as the vehicle - end measurement output of the previous cycle into the vehicle - end state observer to obtain the vehicle - end estimation of the traffic state in the current cycle; Substitute the roadside measurement data as the roadside measurement output of the previous cycle into the roadside state observer to obtain the roadside estimation of the traffic state in the current cycle; S30. Fuse the vehicle - end estimation and the roadside estimation as the prior estimations of the traffic state in the current cycle by the vehicle - end Kalman filter and the roadside Kalman filter, and respectively perform posterior estimations of the traffic state in the current cycle by the vehicle - end Kalman filter and the roadside Kalman filter; S40. Fuse the posterior estimations of the vehicle - end Kalman filter and the roadside Kalman filter to obtain the hybrid estimation of the traffic state in the current cycle; S50. After fusing the hybrid estimation with the vehicle - end measurement data of the traffic state in the current cycle, use it as the vehicle - end measurement output of the current cycle and substitute it into the vehicle - end state observer to obtain the vehicle - end estimation of the traffic state in the next cycle; after fusing the hybrid estimation with the roadside measurement data of the traffic state in the current cycle, use it as the roadside measurement output of the current cycle and substitute it into the roadside state observer to obtain the roadside estimation of the traffic state in the next cycle; S60. Return to S30 according to the vehicle - end estimation and the roadside estimation in the next cycle, and repeat the process to sequentially obtain the hybrid estimations of the traffic state in each cycle.

2. The method according to claim 1, wherein Before S10, it further includes: Conduct cell division on the highway; Based on the divided cells, establish a hybrid traffic flow model of the highway; According to the number and positions of the vehicle - end sensors, and the number and positions of the roadside sensors, construct the output matrix in the hybrid traffic flow model; According to the hybrid traffic flow model and the output matrix, respectively construct a vehicle - end state observer based on the vehicle - end sensors and a roadside state observer based on the roadside sensors; According to the hybrid traffic flow model and the output matrix, respectively construct a vehicle - end Kalman filter based on the vehicle - end sensors and a roadside Kalman filter based on the roadside sensors.

3. The method according to claim 1, wherein The vehicle - end state observer is: Among them, and respectively represent the estimation of the vehicle end traffic state in the th cycle and the th cycle. represents the output of the vehicle end state observer in the th cycle. represents the vehicle end measurement output in the th cycle; represents the system control input in the th cycle; represents the penetration rate of autonomous vehicles in the th cycle. , , and respectively represent the corresponding system matrix, input matrix, constant matrix and gain matrix; represents the output matrix based on the output of vehicle end sensors. Correspondingly, the step of substituting the vehicle - end measurement data as the vehicle - end measurement output of the previous cycle into the vehicle - end state observer to obtain the vehicle - end estimation of the traffic state in the current cycle includes: After arranging the vehicle-end measurement values in each cell of the highway in sequence, as substitute them into the vehicle-end state observer, and the obtained is the vehicle-end estimate of the traffic state in the current cycle.

4. The method according to claim 1, wherein The vehicle - end Kalman filter is: Among them, and respectively represent the prior estimate and the posterior estimate of the traffic state in the th cycle by the on-vehicle Kalman filter, represents the gain matrix of the on-vehicle Kalman filter in the th cycle, represents the output matrix of the on-vehicle Kalman filter; represents the th cycle of the penetration rate of autonomous vehicles, represents the corresponding constant matrix; represents the on-vehicle measurement output in the th cycle of the on-vehicle Kalman filter; Correspondingly, S30 includes: Fuse the vehicle-end estimation and roadside estimation, and use the fusion result as Substitute it into the vehicle-end Kalman filter; After arranging the vehicle-end measurement values in each cell of the highway in the current cycle in order, it is used as substituted into the vehicle-end Kalman filter; Calculate according to the substituted formula , which is the posterior estimate of the current cycle traffic state by the car-end Kalman filter.

5. The method according to claim 1, characterized in that, The vehicle - end state observer is: Among them, and respectively represent the car-end estimations of the traffic states in the th cycle and the th cycle, represents the output of the car-end state observer in the th cycle, represents the car-end measurement output in the th cycle; represents the system control input in the th cycle; represents the penetration rate of autonomous vehicles in the th cycle, , , and respectively represent the corresponding system matrix, input matrix, constant matrix and gain matrix; represents the output matrix based on the output of the car-end sensors. Correspondingly, the step of, after fusing the hybrid estimation with the vehicle - end measurement data of the traffic state in the current cycle, using it as the vehicle - end measurement output of the current cycle and substituting it into the vehicle - end state observer to obtain the vehicle - end estimation of the traffic state in the next cycle includes: Calculate the hybrid estimation according to the output matrix The corresponding output ; The said output is fused with the on-vehicle measurement data of the current cycle traffic state to obtain a new output , where , represents a fusion function; Modify the vehicle - end state observer to: Among them, represents the output corresponding to the hybrid estimation of the th cycle, represents the new output after fusing the output corresponding to the hybrid estimation of the th cycle with the car body measurement data; Substitute the new output as into the corrected vehicle-end state observer, and the obtained is the vehicle-end estimate of the traffic state in the next cycle.

6. The method according to claim 5, wherein Said output is fused with the on-vehicle measurement data of the current cycle traffic state to obtain a new output , including: If there is no vehicle end sensor in the th cell of the highway, the output value of the th cell in is used as the new output value of the th cell ; ; If there is a vehicle end sensor in the th cell of the highway, the output value of the th cell in is weighted-averaged with the vehicle end measurement value of the traffic state of the th cell in the current cycle to obtain the new output value of the th cell ; ; Arrange the new output values of each cell in sequence to obtain the new output .

7. The method according to claim 1, characterized in that, The step of fusing the vehicle - end estimation and the roadside estimation includes: Respectively use multiple groups of weighting parameters to perform cyclic weighted fusion on the vehicle - end estimation and the roadside estimation to obtain multiple fusion result sequences; Take the weighting parameter combination corresponding to the fusion result sequence with the minimum standard deviation as the optimal weighting parameter combination; According to the optimal weighting parameter combination, perform the final weighted fusion on the vehicle - end estimation and the roadside estimation.

8. The method according to claim 1, wherein The traffic state is traffic flow density.

9. An electronic device, characterized in that, Comprising: One or more processors; A memory for storing one or more programs, When the one or more programs are executed by the one or more processors, enabling the one or more processors to implement the vehicle-road integrated hybrid traffic state estimation method according to any one of claims 1-8.

10. A computer-readable storage medium, characterized in that, On which a computer program is stored, and when the computer program is executed by a processor, it implements the vehicle-road integrated hybrid traffic state estimation method according to any one of claims 1-8.

Images