Expressway lane variable speed limit control method, device and medium

Abstract

The application discloses a highway lane-divided variable speed limit control method, equipment and medium, wherein the method comprises the following steps: acquiring road section state data, judging whether a congestion wave is generated and positioning; for a dissipating congestion wave, according to the SPECIALIST theory, a first variable speed limit control scheme of each lane is calculated by using a flow-density graph and a time-position graph; meanwhile, considering the safety speed limit value under the conditions of weather and road alignment, a lane-divided control strategy is used to determine a second variable speed limit control scheme, which comprises a variable speed limit control range and control time of each road section, and a control speed limit value of each lane; finally, the second variable speed limit control scheme is optimized with the time-space continuity optimization as the target, and lane-divided variable speed limit control is performed according to the optimized scheme. The application is a lane-divided variable speed limit control method which comprehensively considers the time-space continuity of speed limit values and the safety speed limit value under the conditions of bad weather and road alignment, and has practical engineering application value.

Description

Technical Field

[0001] This invention relates to the fields of intelligent highway and traffic management technology, and in particular to a method, device and medium for lane-specific variable speed limit control on highways. Background Technology

[0002] Variable speed limit control on highway mainlines is currently one of the important management and control measures to solve highway congestion. Existing research shows that variable speed limit control can effectively alleviate highway congestion, reduce traffic accidents, and improve road efficiency. Variable speed limit control on highway mainlines involves setting variable speed limit signs at intervals along the roadside. The road control center optimizes and adjusts the speed limit values ​​for each section based on real-time traffic data (flow rate, road occupancy, weather, etc.) to achieve uniform vehicle speed, smooth traffic flow, improve traffic capacity, and reduce or avoid traffic accidents. Currently, many variable speed limit control methods have been proposed by scholars both domestically and internationally, but most of them only address single scenarios, such as fixed bottleneck sections upstream of ramp entrances, or variable speed limit control methods under adverse weather conditions such as ice, snow, and rain. Methods applied to practical engineering are rare. Furthermore, the theoretically feasible methods often rely on assumptions that are unsuitable for actual traffic conditions. Summary of the Invention

[0003] This invention provides a method, device, and medium for lane-specific variable speed limit control on highways to solve the problem that existing variable speed limit control schemes cannot adapt to actual traffic conditions.

[0004] Firstly, a lane-specific variable speed limit control method for highways is provided, employing the LLESC-VSL (Variable speed limit control method of lane-level expressway with spatiotemporal continuity) variable speed limit control method, including the following steps:

[0005] S1: Determine if the detection cycle has been reached;

[0006] S2: Obtain traffic flow, speed, and density information for each road segment, as well as weather information and road alignment information;

[0007] S3: Determine whether congestion waves are generated on each road segment; if they are generated, proceed to the next step; if they are not generated, return to step S1.

[0008] S4: Locate the road segment that generates the congestion wave and merge consecutively congested road segments as a single congestion wave;

[0009] S5: Determine whether each congestion wave meets the dissipation constraint conditions, that is, determine whether the range of the upstream free flow state at each congestion wave location can provide a sufficient variable speed limit control range for the congestion wave; if it cannot dissipate, proceed to step S6; if it can dissipate, proceed to step S7.

[0010] S6: Locate the location of the non-dissipable congestion wave, search upstream of the congestion wave for dissipable congestion waves and dissipate them, while the non-dissipable congestion wave waits for detection and control in the next detection cycle.

[0011] S7: Based on SPECIALIST theory, the first variable speed limit control scheme for each lane is calculated using flow-density maps and time-location maps. The first variable speed limit control scheme includes the variable speed limit control range, control time, and control speed limit value. Among them, SPECIALIST theory is an algorithm based on congestion wave (shock wave) theory that can effectively dissipate congestion waves.

[0012] S8: Calculate and optimize the first variable speed limit control scheme by taking into account the safe speed limit value under weather and road alignment conditions;

[0013] S9: Combining the optimized first variable speed limit control scheme, the second variable speed limit control scheme is determined by using a lane-specific control strategy. The second variable speed limit control scheme includes the variable speed limit control range and control time for each road segment, and the control speed limit value for each lane.

[0014] S10: Establish an online speed limit optimization model with the goal of achieving optimal spatiotemporal continuity of the control speed limit value, optimize the second variable speed limit control scheme, implement lane-specific variable speed limit control using the optimized second variable speed limit control scheme, and wait for the next detection cycle.

[0015] According to the first aspect, in one possible implementation, the process of determining whether congestion waves are generated in each road segment in step S3 includes:

[0016] The average density and average speed of each road segment are compared with the set thresholds. If the average density is higher than the corresponding threshold and the average speed is lower than the corresponding threshold, then the road segment is considered to have a congestion wave.

[0017] The threshold is determined as follows: historical data is used to draw a flow-density map of each road segment within the variable speed limit control area, and the average, maximum, or minimum value of the critical density and critical speed of each lane is taken as the corresponding threshold.

[0018] According to the first aspect, in one possible implementation, in step S7, a flow-density map is drawn by calculating flow and density using a method of doubling the traffic demand.

[0019] According to the first aspect, in one possible implementation, when drawing the traffic flow-density map, the traffic flow data is converted into an equivalent number of cars based on the different proportions of vehicle types in different lanes.

[0020] According to the first aspect, in one possible implementation, the application of a lane-specific control strategy to determine the second variable speed limit control scheme includes:

[0021] The optimized first variable speed limit control scheme for each lane is processed as follows: If the control speed limit value is differentiated by lane, no adjustment is made; if the variable speed limit control range and control time are not differentiated by lane, they are unified as follows: control is performed on a segment-by-segment basis, and the maximum, minimum, or average value of the variable speed limit control range and control time for all lanes is taken and unified.

[0022] According to the first aspect, in one possible implementation, the online speed limit optimization model is represented as follows:

[0023]

[0024]

[0025]

[0026] v i，j (k)≤v w (k)

[0027] |v i，j (k)-v i，j (k-1)|≤Δv

[0028] |v i，j (k)-v i-1，j (k)|≤Δv

[0029] |v i，j (k-1)-v i-1，j (k)|≤Δv

[0030] v i，j (k)≤v i，j+1 (k)

[0031] Among them, V ctrl (k) is the decision variable. This represents the speed limit value for each lane and each road segment in the kth detection cycle; This represents the speed limit value for the j-th lane of the i-th road segment in the k-th detection cycle; This represents the ideal speed limit for the j-th lane of the i-th road segment in the k-th detection period; α and β are weighting coefficients; Δv represents the speed limit variation range; v i，j(k) represents the detected speed value of the j-th lane in the i-th road segment during the k-th detection cycle; This represents the safe speed limit calculated based on the current road alignment information; v w (k) represents the safe speed limit calculated based on current weather information; J represents the total number of lanes; N represents the total number of road segments.

[0032] According to the first aspect, one possible implementation also includes:

[0033] S01: Construct a simulation experiment scenario for variable speed limit control based on highway information;

[0034] S02: Considering the randomness of traffic demand and driver compliance, the dissipated state parameters in the SPECIALIST theory are used as decision variables. The dissipated state parameters include the flow and density of states 4 and 5 in the SPECIALIST theory. The total travel delay on the main line of the highway and the number of traffic accidents on the main line of the highway are used as dual objective evaluation indicators to construct an offline optimization model for the variable speed limit control of the main line of the highway.

[0035] S03: Obtain historical traffic data of the highway and input it into the simulation experiment scenario of variable speed limit control. During the simulation, the LLESC-VSL variable speed limit control method is used to carry out lane-specific variable speed limit control and collect simulation evaluation data. The process is repeated until the congestion wave dissipates or the upper limit of the detection cycle is reached.

[0036] S04: Based on all collected simulation evaluation data, solve the offline optimization model of variable speed limit control for highway mainline to obtain the optimal dissipation state parameters, and apply the obtained optimal dissipation state parameters to step S7 of the LLESC-VSL variable speed limit control method.

[0037] According to the first aspect, in one possible implementation, the offline optimization model for the variable speed limit control of the highway mainline is represented as follows:

[0038]

[0039]

[0040]

[0041] d4>d5

[0042]

[0043]

[0044] q5 > q1

[0045] d5>d1

[0046] Where x is the decision variable, x = {d4, d5, q4, q5}, d4 and q4 represent the density and flux of state 4 in SPECIALIST theory, respectively; d5 and q5 represent the density and flux of state 5 in SPECIALIST theory, respectively; D r C refers to the relative indicator of total travel delay. r This refers to the relative indicator of the number of traffic accidents. and Let V represent the traffic demand randomness disturbance coefficient and the driver compliance disturbance coefficient, respectively; Ω represents the value space of the decision variable x; E(·) represents the expected value; Φ(·) represents the simulation evaluation, and → represents the output; LLESC-VSL(·) represents the variable speed limit control scheme obtained by using the LLESC-VSL variable speed limit control method; V ideal This represents the set of ideal speed limits under traffic conditions for each detection cycle; V ctrl U represents the set of controlled speed limit values ​​under traffic conditions in each detection cycle; T represents the set of variable speed limit control ranges in each detection cycle; v represents the set of variable speed limit control times in each detection cycle; free d1 represents the free flow velocity; q1 and d1 represent the density and flow rate of state 1 in SPECIALIST theory, respectively.

[0047] Secondly, an electronic device is provided, comprising:

[0048] A memory that stores computer programs;

[0049] The processor, when calling and executing the computer program, implements the highway lane-specific variable speed limit control method as described above.

[0050] Thirdly, a computer-readable storage medium is provided that stores a computer program, which, when executed by a processor, implements the highway lane-specific variable speed limit control method as described above.

[0051] This invention proposes a lane-specific variable speed limit control method, equipment, and medium for highways. Based on SPECIALIST theory, it constructs an LLESC-VSL variable speed limit control method that comprehensively considers safe speed limits under adverse weather and road alignment conditions, while also taking into account the spatiotemporal continuity of speed limits. This lane-level variable speed limit control scheme can simultaneously dissipate multiple congestion waves and is applicable to real-world traffic conditions. Furthermore, it proposes using two states in the congestion wave change process from SPECIALIST theory as decision variables, with the dual objectives of minimizing total travel delay and the total number of traffic accidents within the highway's variable speed limit control range. An offline optimization model for highway mainline variable speed limit control under uncertain traffic demand and driver compliance conditions is established. Optimal state dissipation parameters are obtained using offline simulation and optimization techniques and then applied to the LLESC-VSL variable speed limit control method to obtain the optimal variable speed limit control scheme, which can resist environmental disturbances and improve robustness. This invention overcomes the shortcomings of existing variable speed limit controls that only consider a single influencing factor, effectively improving road traffic efficiency while reducing road traffic accidents, and has practical engineering application value. Attached Figure Description

[0052] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0053] Figure 1 This is a flowchart of the LLESC-VSL variable speed limiting control method provided in an embodiment of the present invention;

[0054] Figure 2 This is a schematic diagram of a congestion wave provided in an embodiment of the present invention;

[0055] Figure 3 This is a framework diagram of the highway lane-specific variable speed limit control method provided in an embodiment of the present invention;

[0056] Figure 4 This is a schematic diagram of data interaction in a variable speed limit control simulation experiment scenario provided in an embodiment of the present invention. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be described in detail below. Obviously, the described embodiments are merely some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other implementation methods obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0058] In the description of this invention, it should be understood that the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or order.

[0059] This invention provides a lane-specific variable speed limit control method for highways, employing the LLESC-VSL (Variable speed limit control method of lane-level expressway with spatiotemporal continuity) variable speed limit control method, such as... Figure 1 As shown, it includes the following steps:

[0060] S1: Determine whether the detection cycle has been reached. In this embodiment, the detection cycle is 120s. In other embodiments, the detection cycle can also be selected as 30s, 60s, 5min, etc., according to actual needs.

[0061] S2: Obtain traffic flow, speed, and density information for each road segment from the detectors. Each road segment is equipped with a detector to obtain the above traffic data information for the corresponding road segment; and collect weather information and road alignment information.

[0062] S3: Determine whether congestion waves are generated on each road segment; if they are generated, proceed to the next step; if they are not generated, return to step S1.

[0063] Specifically, the process of determining whether congestion waves are occurring on each road segment includes:

[0064] The average density and average speed of each road segment are compared with the set thresholds. If the average density is higher than the corresponding threshold and the average speed is lower than the corresponding threshold, then the road segment is considered to have a congestion wave.

[0065] The threshold is determined as follows: Historical data is used to create a flow-density map for each road segment within the variable speed limit control area, and the average value of the critical density and critical speed for each lane is taken as the corresponding threshold. Alternatively, in other embodiments, the maximum or minimum value of the critical density and critical speed for each lane can be used as the corresponding threshold. A smaller threshold indicates that the road segment is more likely to be identified as having a congestion wave, thus triggering variable speed limit control; conversely, a larger threshold makes it less likely for variable speed limit control to be activated.

[0066] S4: Locate the road segment that generates the congestion wave and merge consecutive congested road segments as a single congestion wave.

[0067] S5: Determine whether each congestion wave meets the dissipation constraint conditions, that is, determine whether the range of the upstream free flow state at each congestion wave location can provide a sufficient variable speed limit control range for the congestion wave; if it cannot dissipate, proceed to step S6, and if it can dissipate, proceed to step S7.

[0068] The method for determining whether the range of upstream free flow at each congestion wave location can provide a sufficient variable speed control range for the congestion wave is as follows: the required control range is calculated based on the time-position fundamental diagram in SPECIALIST theory.

[0069] S6: Locate the location of the non-dissipable congestion wave, search upstream of the congestion wave for dissipable congestion waves and dissipate them, while the non-dissipable congestion wave waits for detection and control in the next detection cycle.

[0070] There are several ways to handle congestion waves: dissipable congestion waves can be dissipated in the original way; when encountering an undissipable congestion wave, start from the undissipable congestion wave and search upstream for dissipable congestion waves in sequence. By dissipating the upstream dissipable congestion waves of this undissipable congestion wave, the degree of congestion (the speed and range of congestion waves propagating upstream) is reduced, and the next opportunity for dissipation is waited for.

[0071] like Figure 2 As shown, in the case of two congestion waves, road segments 6 and 8 generate congestion waves simultaneously. If the calculated length of the upstream free-flow area required for the congestion wave to dissipate in road segment 8 is greater than that in road segment 7, then under the current conditions, the congestion wave in road segment 8 cannot dissipate. Therefore, by sequentially examining the dissipable congestion waves upstream from road segment 8, the dissipable congestion wave in road segment 6 is controlled first. The dissipation of the congestion wave in road segment 6 can alleviate the congestion situation in its downstream segments to some extent.

[0072] S7: Based on SPECIALIST theory, the first variable speed limit control scheme for each lane is calculated using flow-density maps and time-location maps. The first variable speed limit control scheme includes the variable speed limit control range, control time, and control speed limit value.

[0073] In this embodiment, the flow-density map is drawn by progressively increasing traffic demand to calculate flow and density. Furthermore, when drawing the flow-density map, the flow data is converted into an equivalent number of cars based on the different vehicle types in each lane.

[0074] S8: Calculate and optimize the first variable speed limit control scheme by taking into account the safe speed limit value under weather and road alignment conditions.

[0075] Among them, the safe speed limit value v under adverse weather conditions w (k) Visibility under adverse weather conditions is taken into consideration. The maximum legal speed limit of 120 km / h is taken when the weather is good, referring to the safe speed limit given in the Highway Engineering Technical Standard (JTG B01-2014) (as shown in Table 1).

[0076] Table 1

[0077]

[0078] Safe speed limit under poor road alignment conditions The calculation mainly considers the lateral force coefficient and superelevation of the road. When the road alignment is good, the speed limit is taken as the maximum legal speed limit of 120 km / h. The calculation formula is as follows:

[0079]

[0080]

[0081] In the formula, R i θ represents the curve radius. i Indicates the lateral force coefficient. This represents the average speed of vehicles traveling in the outermost lane of the i-th road segment during the k-th detection period. It has an extremely high cross slope.

[0082] S9: Combining the optimized first variable speed limit control scheme with the lane-specific control strategy, determine the second variable speed limit control scheme. The second variable speed limit control scheme includes the variable speed limit control range and control time for each road segment, and the speed limit value for each lane.

[0083] In this embodiment, lane-specific control is mainly reflected in two aspects: First, based on the different vehicle types in different lanes (different vehicle types have different probabilities of choosing different lanes), the traffic flow data is converted into the equivalent number of vehicles according to the equivalent car conversion factor, and a traffic flow-density map is drawn for each lane; Second, the optimized first variable speed limit control scheme (variable speed limit control range, control time, and control speed limit value) for each lane is processed as follows: If the control speed limit value is differentiated by lane, no adjustment is made; if the variable speed limit control range and control time are not differentiated by lane, they are unified. The unification method is: control is performed on a segment-by-segment basis, and the maximum, minimum, or average value of the variable speed limit control range and control time for all lanes is taken and unified. The larger the value, the wider the influence range (time and space) of the variable speed limit control.

[0084] As shown in Tables 2 and 3 (retaining two significant figures), the equivalent car conversion factor for different car models and the probability statistics of different car models choosing different lanes are presented respectively.

[0085] Table 2

[0086]

[0087] Table 3

[0088]

[0089] S10: Establish an online speed limit optimization model with the goal of achieving optimal spatiotemporal continuity of the control speed limit (meaning minimizing the change in the control speed limit over time and space), optimize the second variable speed limit control scheme, implement lane-specific variable speed limit control using the optimized second variable speed limit control scheme, and wait for the next detection cycle.

[0090] The optimization index is set to improve the temporal and spatial continuity of the speed limit, thereby reducing acceleration and deceleration behaviors during driving. On the one hand, this improves the homogeneity of vehicle speed, reducing the probability of traffic accidents; on the other hand, from an engineering perspective, it increases drivers' acceptance of speed limits and their changes, thus enhancing the effectiveness of the variable speed limit control scheme. In this embodiment, the online speed limit optimization model is represented as follows:

[0091]

[0092]

[0093]

[0094] v i，j (k)≤v w (k) (2-4)

[0095] |v i，j (k)-v i，j (k-1)|≤Δv (2-5)

[0096] |v i，j (k)-v i-1，j (k)|≤Δv (2-6)

[0097] |v i，j (k-1)-v i-1，j (k)|≤Δv (2-7)

[0098] v i，j (k)≤v i，j+1 (k) (2-8)

[0099] In formula (2-1) V ctrl (k) is the decision variable. This represents the speed limit value for each lane and each road segment in the kth detection cycle; This represents the speed limit value for the j-th lane of the i-th road segment in the k-th detection cycle; Let represent the ideal speed limit value for the j-th lane of the i-th road segment in the k-th detection period; i represents the road segment number (numbered sequentially from downstream to upstream), N represents the total number of road segments; j represents the lane number, numbered sequentially from the rightmost to the innermost based on the traffic flow direction, and J represents the total number of lanes. This online speed limit optimization model uses the average of the sum of the squares of the speed differences between the current and previous times for each lane and each road segment; the squares of the speed differences between the current and downstream road segments at the same time; and the squares of the speed differences between the previous and current times for the current and downstream road segments as the first optimization objective. This represents minimizing the spatiotemporal difference of the speed limit value, i.e., maximizing spatiotemporal continuity. The second optimization objective is the square of the difference between the controlled speed limit value and the ideal speed limit value, representing the desire to dissipate congestion with as few speed changes as possible. The weighted sum of the two objectives is taken as the overall optimization objective, where α and β are weighting coefficients. Formula (2-2) aims to improve driver compliance and driving safety by setting the variable speed limit value of the speed limit sign as a discrete value, where Δv represents the range of speed limit change, and can take values ​​such as 5 km / h, 10 km / h, and 20 km / h; Formula (2-3) states that the speed limit value for each lane and road segment should be less than the safe speed limit value considering the road alignment, where... This represents the safe speed limit calculated based on the current road alignment information, v i，j (k) represents the detected speed value of the j-th lane in the ith segment during the k-th detection cycle; Formula (2-4) states that the speed limit for each lane and each segment should be less than the safe speed limit considering weather conditions, where v w (k) represents the safe speed limit calculated based on current weather information; Formula (2-5) indicates that the absolute value of the difference between the current speed limit and the next speed limit for the same lane and the same road segment should be less than the speed limit change range considering time continuity, where Δv represents the speed limit change range considering time continuity; Formula (2-6) indicates that the absolute value of the difference between the current speed limit and the downstream speed limit for the same lane and the same road segment at the same time should be less than the speed limit change range considering spatial continuity, where Δv represents the speed limit change range considering spatial continuity; Formula (2-7) indicates that the absolute value of the difference between the previous speed limit and the current speed limit for the current road segment and the downstream speed limit for the same lane should be less than the speed limit change range considering both spatial and temporal continuity, where Δv represents the speed limit change range considering both time and spatial continuity; Formula (2-8) indicates that the speed limit for the inner lane should be higher than the speed limit for the outer lane.

[0100] The LLESC-VSL variable speed limit control method calculates the variable speed limit control range, control time, and control speed limit value based on SPECIALIST theory. The lane-specific strategy of the LLESC-VSL variable speed limit control method adjusts the control distance and control time required for congestion wave dissipation at different locations on each lane segment. Simultaneously, considering the spatiotemporal continuity of the control speed limit value, an online speed limit value optimization model is used to adjust the control speed limit value. The main contents include: 1) converting the lane-level control range into a segment-level control range and rounding it according to the segment interval length; converting the lane-level control time into a segment-level control time and rounding it according to the detection cycle. 2) Adjust the control time and control distance based on the control scheme of the previous detection cycle and the traffic conditions of the current detection cycle: For congestion waves detected in the previous detection cycle that have not yet dissipated in the current detection cycle, their control range remains the control range calculated in the previous detection cycle, and their control time is the control time calculated in the previous detection cycle minus the duration of the previous detection cycle; the condition for further updating / resetting the congestion wave is: the congestion wave dissipates or its control duration is less than or equal to 0. 3) Optimize the control speed limit value based on the constructed online speed limit value optimization model that considers spatiotemporal continuity.

[0101] It should be noted that SPECIALIST theory is an algorithm based on congestion wave (shock wave) theory that can effectively dissipate congestion waves. It is an existing technology, and for details, please refer to "Hegyi A, Hoogendoom SP, Schreuder M, et al. SPECIALIST: A dynamic speed limit control algorithm based on shock wave theory [C] / / 2008 11th international IEEE conference on intelligent transportation systems. IEEE, 2008: 827-832.", which will not be elaborated here.

[0102] The lane-specific variable speed limit control method for highways provided in the above embodiments is based on the SPECIALIST theory to construct the LLESC-VSL variable speed limit control method. It comprehensively considers the safe speed limit value under adverse weather conditions and poor road alignment conditions, and also considers the spatiotemporal continuity of the speed limit value. It is a lane-level variable speed limit control scheme that can dissipate multiple congestion waves at the same time and is applicable to actual traffic conditions.

[0103] In SPECIALIST theory, the spatiotemporal evolution of a congestion wave mainly includes six states: State 1 is the free-flow state downstream of the congestion wave; State 2 is the congested state; State 3 is the state where the density remains constant and the speed and flow are low after implementing variable speed limit control on the congested traffic flow; State 4 is the state after the traffic flow adapts to the variable speed limit control; State 5 is the state where the traffic flow accelerates after the variable speed limit control is lifted; and State 6 is the free-flow state upstream of the congestion wave origin. In step S7, according to SPECIALIST theory, in the process of calculating the first variable speed limit control scheme for each lane using the flow-density map and time-location map, this embodiment fixes the values ​​of the dissipation state parameters (density d4 in state 4, density d5 in state 5, flow q4 in state 4, and flow q5 in state 5) and then uses them to calculate the first variable speed limit control scheme. However, considering that the traffic conditions on highways vary at different times, such as the significant difference between weekdays and holidays, and the obvious differences between different times of day (such as daytime and nighttime), if the above state quantities are all taken as the same fixed values ​​at different times, and the uncertainty of traffic demand and driver compliance is not taken into account, the control accuracy will also be affected.

[0104] Based on the above, another embodiment of the present invention provides a lane-specific variable speed limit control method for highways. The difference between this method and the previous embodiment lies in the introduction of offline simulation technology to obtain optimal dissipation state parameters more suitable for different time periods. These parameters are then applied to the LLESC-VSL variable speed limit control method to obtain the optimal variable speed limit control scheme. Specifically, the method includes:

[0105] S01: Constructing a simulation experiment scenario for variable speed limit control based on highway information.

[0106] Specifically, based on the number of lanes and ramp configuration, a basic road network model is established in the microscopic traffic simulation software SUMO using the OpenStreetMap open-source urban map. The road network model is then configured based on real OD (Original Distance) data obtained from historical data to construct a variable speed limit control simulation experiment scenario. In this example, the microscopic simulation software used is SUMO (Simulation of Urban Mobility). The experimental scenario road is divided into equally spaced segments, each 2km long, numbered i = 1, 2, ..., N. Simultaneously, detectors are deployed at equal intervals along each segment, with each detector number matching the segment number. Furthermore, a data interaction channel is established between the simulation scenario and the simulation optimization algorithm through the SUMO interface tool TraCI. Specific data interaction details are as follows: Figure 4 As shown.

[0107] S02: Considering the randomness of traffic demand and driver compliance, the dissipative state parameters in SPECIALIST theory are used as decision variables. The dissipative state parameters include the flow and density of states 4 and 5 in SPECIALIST theory. The total travel delay on the main line of the highway and the number of traffic accidents on the main line of the highway are used as dual objective evaluation indicators to construct an offline optimization model for variable speed limit control on the main line of the highway.

[0108] The offline optimization model for the variable speed limit control of the main line of the highway is expressed as follows:

[0109]

[0110]

[0111]

[0112] d4>d5 (3-4)

[0113]

[0114]

[0115] q s >q1 (3-7)

[0116] d5>d1 (3-8)

[0117] In formula (3-1), x is the decision variable, x = {d4, d5, q4, q5}, where d4 and q4 represent the density and flow rate in state 4 of the SPECIALIST theory, respectively, and d5 and q5 represent the density and flow rate in state 5 of the SPECIALIST theory, respectively. The unit of density is veh / km, and the unit of flow rate is veh / h / lane; D r Cr refers to the relative indicator of total travel delay, and Cr refers to the relative indicator of the number of traffic accidents. This variable speed limit control offline optimization model takes both as dual objectives; Ω represents the value space of the decision variable x; E(·) represents the expected value. This represents the random disturbance coefficient of traffic demand. Traffic demand randomness follows a normal distribution This random disturbance represents the change in highway travel traffic volume, meaning that the change in traffic demand is the change in traffic volume between each origin-destination (OD) and the traffic volume OD. m The product of its corresponding random disturbance coefficient, through which the traffic volume of each OD pair is located with a 95% probability at [OD]. m ×0.8, OD m Within [×1.2], where M represents the number of OD pairs, OD m This represents the value of the m-th OD pair. This represents the driver compliance disturbance coefficient. use This formula represents the degree of deviation between the speed a driver intends to drive and the given speed limit, expressed as: given a speed limit for driver i, the speed at which the driver intends to drive is equal to or greater than the speed limit. This indicates that the larger the value, the lower the driver i's compliance with the speed limit; where V veh V represents the actual speed of the vehicle. ctrl This indicates the speed displayed on the speed limit sign; The mean is 0, and the confidence interval (-0.1, 0.1) has a confidence level of 95%. The confidence interval setting indicates that the maximum possible execution deviation by the driver is 10%, and that 95% of all drivers have a compliance rate within 10%, meaning there is a 95% probability that the driver's driving speed is within [v...]. ctrl ×0.9, v ctrl Within [×1.1], given V ctrl The actual driving speed of the driver at that time was

[0118] In formula (3-2), Φ(·) represents simulation evaluation, and → represents simulation output, meaning that given decision variables and disturbances, each simulation can output the bi-objective function value corresponding to the decision variables; in formula (3-3), LLESC-VSL(·) represents the variable speed limit control scheme obtained by using the LLESC-VSL variable speed limit control method, and → represents the key parameter output of the variable speed limit control scheme. LLESC-VSL is the control method proposed in this invention. Here, the traffic conditions of each road segment and lane in each detection cycle are obtained through simulation output to determine whether congestion occurs. At the same time, based on the congestion wave dissipation principle, the variable speed limit control scheme of the previous detection cycle, the dissipation state parameters of the offline optimization part, and the online speed limit value optimization model, the control range, control time, and speed limit value required for the dissipation of each wave are calculated; V ideal V represents the set of ideal speed limits under traffic conditions for each detection cycle. ideal ={V ideal (1), ..., V ideal (k), ..., V ideal (K)}, V ctrl V represents the set of speed limit values ​​under traffic conditions in each detection cycle. ctrl ={V ctrl (1), ..., V ctrl (k), ..., V ctrl (K)}, U represents the set of variable speed limit control ranges for each detection cycle, U = {U(1), ..., u(k), ..., U(K)}; T represents the set of variable speed limit control times for each detection cycle, T = {T(1), ..., T(k), ..., T(K)}, K represents the total number of detection cycles; Formulas (3-4)-(3-8) are the constraints for congestion wave dissipation; v free d1 represents the free flow velocity; q1 and d1 represent the density and flow rate of state 1 in SPECIALIST theory, respectively.

[0119] S03: Obtain historical traffic data for the corresponding time period of the highway and input it into the variable speed limit control simulation experiment scenario for simulation. During the simulation, the LLESC-VSL variable speed limit control method is used for lane-specific variable speed limit control, and simulation evaluation data is collected. This process is repeated until the congestion wave dissipates or the upper limit of the detection cycle is reached. The dissipation state parameters are continuously updated, and then the simulation is repeated. The LLESC-VSL variable speed limit control method is used for lane-specific variable speed limit control until the congestion wave dissipates or the upper limit of the detection cycle is reached. Multiple sets of simulation evaluation data are collected in this way.

[0120] S04: Based on all collected simulation evaluation data, solve the offline optimization model of the variable speed limit control for the main line of the highway to obtain the optimal dissipation state parameters. When solving the offline optimization model of the variable speed limit control for the main line of the highway, a bi-objective stochastic simulation optimization algorithm (BOSPAS) can be used, employing fusion surrogate modeling technology, collapsed sphere noise filtering technology, and potential zone search technology.

[0121] S05: Apply the obtained optimal dissipation state parameters to the LLESC-VSL variable speed limit control method (specifically applied in step S7) to perform online lane-specific variable speed limit control on the highway during the corresponding time period.

[0122] Unlike previous methods that only considered single fixed congested road sections, moving congestion waves, and congestion scenarios under adverse weather conditions, this embodiment proposes a lane-level variable speed limit control method for highways that considers the randomness of traffic demand, the randomness of driver compliance, and the spatiotemporal continuity of speed limits. It also considers variable speed limit control under adverse weather conditions and different road alignments, combining offline optimization with online control. This invention overcomes the shortcomings of existing variable speed limit controls that only consider a single influencing factor, effectively improving road traffic efficiency while reducing road traffic accidents, and has practical engineering application value. Specifically, using two states (flow and density in states 4 and 5) during the congestion wave change process as decision variables, and minimizing the total travel delay and the total number of traffic accidents within the highway variable speed limit control range as dual objectives, a dual-objective offline optimization model for highway mainline variable speed limit control under conditions of uncertain traffic demand and uncertain driver compliance is established. Simultaneously, a dual-objective stochastic simulation optimization algorithm (BOSPAS) combining surrogate modeling, collapsed sphere noise filtering, and potential zone search techniques is used to solve the problem. Unlike existing variable speed limit control methods, this invention innovates and improves upon them in the following aspects: 1) It is a control method that considers the spatiotemporal continuity of speed limits; 2) It is a simulation optimization method that considers driver compliance and the randomness of traffic demand, capable of resisting environmental disturbances and improving the robustness of the method; 3) It is a lane-level variable speed limit control method; 4) It is a variable speed limit control method that comprehensively considers safe speed limits under adverse weather conditions and poor road alignment conditions; 5) It is a method that can simultaneously dissipate multiple congestion waves; 6) It is a variable speed limit control method that combines offline optimization and online control. Therefore, this embodiment provides a comprehensive, highly operable, and effective variable speed limit control method.

[0123] This invention also provides a simulation platform for lane-specific variable speed limit control on highways that considers the spatiotemporal continuity of speed limit values, such as... Figure 4 As shown, the system includes a Python main control engine and a SUMO traffic simulation model for variable speed limit control on the main line of a highway. The Python main control engine acts as an information exchange bridge, responsible for data exchange between the simulation optimization control algorithm, the simulation control scheme, the SUMO traffic simulation model, and the simulation data files. The SUMO traffic simulation model for variable speed limit control on the main line of the highway is built according to the experimental scenario and mainly includes experimental scenario road network files, routing files, additional files, configuration files, etc., which are mainly used to describe the road network conditions and traffic demand conditions of the experimental scenario. The microscopic traffic simulation software can also use VISSIM, PARAMICS, etc., and the control platform can be implemented using other programming languages ​​such as C# and Java. This invention does not limit the specific types of software used.

[0124] This invention also provides an electronic device, comprising:

[0125] A memory that stores computer programs;

[0126] The processor, when calling and executing the computer program, implements the highway lane-specific variable speed limit control method described in the above embodiments.

[0127] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the highway lane-specific variable speed limit control method described in the above embodiments.

[0128] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0129] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0130] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0131] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0132] It is understood that the same or similar parts in the above embodiments can be referred to each other, and the contents not described in detail in some embodiments can be referred to the same or similar contents in other embodiments.

[0133] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A lane-specific variable speed limit control method for highways, characterized in that, The LLESC-VSL variable speed limit control method is adopted. The LLESC-VSL variable speed limit control method is a lane-level variable speed limit control method for highways with spatiotemporal continuity, and includes the following steps: S1: Determine if the detection cycle has been reached; S2: Obtain traffic flow, speed, and density information for each road segment, as well as weather information and road alignment information; S3: Determine whether congestion waves are generated on each road segment; if they are generated, proceed to the next step; if they are not generated, return to step S1. S4: Locate the road segment that generates congestion waves; S5: Determine whether each congestion wave meets the dissipation constraint. If it cannot be dissipated, proceed to step S6; if it can be dissipated, proceed to step S7. S6: Locate the location of the non-dissipable congestion wave, search upstream of the congestion wave for dissipable congestion waves and dissipate them, while the non-dissipable congestion wave waits for detection and control in the next detection cycle. S7: Based on SPECIALIST theory, the first variable speed limit control scheme for each lane is calculated using flow-density map and time-location map. The first variable speed limit control scheme includes variable speed limit control range, control time, and control speed limit value. S8: Calculate and optimize the first variable speed limit control scheme by taking into account the safe speed limit value under weather and road alignment conditions; S9: Combining the optimized first variable speed limit control scheme, the second variable speed limit control scheme is determined by using a lane-specific control strategy. The second variable speed limit control scheme includes the variable speed limit control range and control time for each road segment, and the control speed limit value for each lane. The method of using lane-specific control to determine the second variable speed limit control scheme includes: The optimized first variable speed limit control scheme for each lane is processed as follows: If the control speed limit value is differentiated by lane, no adjustment is made; if the variable speed limit control range and control time are not differentiated by lane, they are unified. The unification method is: control is performed on a road segment basis, and the maximum, minimum or average value of the variable speed limit control range and control time of all lanes is taken and unified. S10: Establish an online speed limit optimization model with the goal of optimizing the spatiotemporal continuity of the control speed limit value, optimize the second variable speed limit control scheme, implement lane-specific variable speed limit control with the optimized second variable speed limit control scheme, and wait for the next detection cycle. Also includes: S01: Construct a simulation experiment scenario for variable speed limit control based on highway information; S02: Considering the randomness of traffic demand and driver compliance, the dissipated state parameters in the SPECIALIST theory are used as decision variables. The dissipated state parameters include the flow and density of states 4 and 5 in the SPECIALIST theory. The total travel delay on the main line of the highway and the number of traffic accidents on the main line of the highway are used as dual objective evaluation indicators to construct an offline optimization model for the variable speed limit control of the main line of the highway. S03: Obtain historical traffic data of the highway and input it into the simulation experiment scenario of variable speed limit control. During the simulation, the LLESC-VSL variable speed limit control method is used to carry out lane-specific variable speed limit control and collect simulation evaluation data. The process is repeated until the congestion wave dissipates or the upper limit of the detection cycle is reached. S04: Based on all collected simulation evaluation data, solve the offline optimization model of variable speed limit control for highway mainline to obtain the optimal dissipation state parameters, and apply the obtained optimal dissipation state parameters to step S7 of the LLESC-VSL variable speed limit control method.

2. The variable speed limit control method for highway lane division according to claim 1, characterized in that, The process of determining whether congestion waves occur in each road segment in step S3 includes: The average density and average speed of each road segment are compared with the set thresholds. If the average density is higher than the corresponding threshold and the average speed is lower than the corresponding threshold, then the road segment is considered to have a congestion wave. The threshold is determined as follows: historical data is used to draw a flow-density map of each road segment within the variable speed limit control area, and the average, maximum, or minimum value of the critical density and critical speed of each lane is taken as the corresponding threshold.

3. The variable speed limit control method for highway lane division according to claim 1, characterized in that, In step S7, a flow-density map is drawn by calculating flow and density by gradually increasing traffic demand.

4. The variable speed limit control method for highway lane division according to claim 1, characterized in that, When drawing a traffic flow-density map, the traffic flow data is converted into an equivalent number of cars based on the different vehicle types in different lanes.

5. The variable speed limit control method for highway lane division according to claim 1, characterized in that, The online speed limit optimization model is represented as follows: in, As decision variables, , indicating the first Speed ​​limit values ​​for each lane and road section in each detection cycle; Indicates the first The first detection cycle The first section of the road Speed ​​limits for each lane; Indicates the first The first detection cycle The first section of the road Ideal speed limit for each lane; and These are the weighting coefficients; Indicates the range of change in the speed limit; Indicates the first The first detection cycle The first section of the road The detected speed values ​​for each lane; This represents the safe speed limit calculated based on the current road alignment information. This represents the safe speed limit calculated based on current weather information. J Indicates the total number of lanes; N This indicates the total number of road segments.

6. The variable speed limit control method for highway lane division according to claim 1, characterized in that, The offline optimization model for variable speed limit control on the main line of the expressway is represented as follows: in, As decision variables, , , These represent the density and flux of state 4 in SPECIALIST theory, respectively. , These represent the density and flux of state 5 in SPECIALIST theory, respectively. This refers to the relative indicator of total travel delay. This refers to the relative indicator of the number of traffic accidents. and These represent the traffic demand randomness disturbance coefficient and the driver compliance disturbance coefficient, respectively. Representing decision variables Value space; Indicates the expected value; Indicates simulation evaluation, Indicates the output; This indicates a variable speed limiting control scheme obtained using the LLESC-VSL variable speed limiting control method; This represents the collection of ideal speed limit values ​​under traffic conditions for each detection cycle; This represents the collection of speed limit values ​​under traffic conditions for each detection cycle. This represents the set of variable speed limit control ranges for each detection cycle; This represents the collection of variable speed limit control times for each detection cycle; Indicates the free-flow velocity; , These represent the density and flux of state 1 in SPECIALIST theory, respectively.

7. An electronic device, characterized in that, include: A memory that stores computer programs; The processor, when calling and executing the computer program, implements the highway lane-specific variable speed limit control method as described in any one of claims 1 to 6.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the highway lane-specific variable speed limit control method as described in any one of claims 1 to 6.

Images