Variable speed limit management method for expressway

Abstract

The invention discloses a variable speed limit management method for an expressway, and the method comprises the steps: obtaining the multivariate traffic scene data of a road segment sequence in a jurisdiction area of the expressway in a preset time window, and the multivariate traffic scene data comprises the traffic flow dynamic feature, the road segment static feature and the scene feature of each road segment in the road segment sequence; based on the multivariate traffic scene data, utilizing a preset road section speed limit matching mechanism to obtain an initial speed limit sequence of the current road section sequence; and based on a preset adjacent road section transmission model, generating a speed-limiting transmission instruction of each road section, and synchronizing the speed-limiting transmission instruction to the road section. Therefore, the differentiated speed limit management effect and safety of the expressway are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of highway control, and in particular to a variable speed limit management method for highways. Background Art

[0002] With the continuous increase in the number of vehicles, traditional static speed limit management can no longer meet modern traffic needs. Variable speed limit management can flexibly adjust according to real-time traffic conditions, providing drivers with a safer and more efficient driving environment. It is an indispensable part of future smart city construction.

[0003] Existing methods for managing variable speed limits on highways primarily utilize static and dynamic speed limit methods. The static speed limit method pre-sets a fixed speed limit based on factors such as road design standards, historical traffic data, and weather conditions. This method is simple but lacks flexibility and cannot adapt to real-time traffic conditions. The dynamic speed limit method uses algorithms to calculate the optimal speed limit based on real-time collected traffic flow parameters (such as traffic volume, vehicle speed, density), as well as road and weather data. This is then displayed using devices such as variable speed limit signs.

[0004] However, the key challenges facing the dynamic speed limit method lie in the accuracy of traffic flow predictions and the real-time computational performance of the control method. Due to the high complexity and uncertainty of traffic flow, accurately predicting its changing trends is a major challenge. Furthermore, to achieve effective real-time speed limit control, efficient algorithms are required to process massive amounts of data and quickly calculate the optimal speed limit. Therefore, while the dynamic speed limit method has improved highway efficiency and safety to a certain extent, further research and improvement are still needed to overcome these technical difficulties. Summary of the Invention

[0005] This application provides a variable speed limit management method for highways to improve the effectiveness and safety of differentiated speed limit management on highways.

[0006] This application provides a method for managing variable speed limits on a highway, comprising:

[0007] S101, obtaining multivariate traffic scene data of a road section sequence within a predetermined time window within a jurisdictional area of ​​an expressway, wherein the multivariate traffic scene data includes dynamic traffic flow characteristics, static road section characteristics, and scene characteristics of each road section in the road section sequence;

[0008] S102, based on the multi-dimensional traffic scene data, using a preset road section speed limit matching mechanism, obtaining an initial speed limit sequence for the current road section sequence;

[0009] S103: Based on a preset adjacent road section transfer model, a speed limit transfer instruction is generated for each road section and synchronized to the road section.

[0010] Preferably, the road segment sequence is composed of a number of road segments that are sequentially divided according to road design standards within the jurisdiction of the expressway; the multivariate traffic scene data of each road segment is obtained by:

[0011] A1. Obtain traffic flow dynamic characteristics, which are set as follows: the mean and variance values ​​of vehicle headway within different lanes within a road section, the mean and variance values ​​of vehicle speed, the mean speed difference between adjacent lanes, and the vehicle crossing rate within a preset time window;

[0012] A2. Obtain static characteristics of the road section to reflect the inherent properties of the road, which are set as: the road design standard of the corresponding road section;

[0013] A3. Obtain scene features to reflect environmental information within the time window, which are set to: rainfall and visibility of the corresponding road section.

[0014] Preferably, the vehicle crossing rate between adjacent lanes is calculated according to the following formula:

[0015]

[0016] Among them, V is the vehicle crossing rate between adjacent lanes, N is the number of lanes in the road section, CR i is the vehicle crossing rate between the i-th lane and the adjacent lane; where the vehicle crossing rate between the i-th lane and the adjacent lane is expressed as:

[0017]

[0018] Among them, CR i is the vehicle crossing rate between the i-th lane and the adjacent lane, k is the number of adjacent lanes of the i-th lane, T is the size of the preset time window, L i is the set of vehicles in the i-th lane, L j is the set of vehicles in the jth adjacent lane, T a,b It is a condition indicator function. When the preset condition is met, the value is assigned to the duration of the preset condition, otherwise it is 0. The preset condition is set to |x a (t)-x b (t)|≤L a ,|x a (t)-x b (t)| represents the distance between the front ends of vehicles a and b in the direction of travel at time t, L a is the body length of vehicle a.

[0019] Preferably, the preset road section speed limit matching mechanism includes:

[0020] B1. Input the multi-dimensional traffic scenario data into a pre-built historical speed limit solution database. The pre-built historical speed limit solution database includes a number of historical multi-dimensional traffic scenario data within the jurisdiction of the expressway and their corresponding actual speed limits and accident rates.

[0021] B2. Filter out historical multivariate traffic scene data with an accident rate less than a preset accident threshold as the object to be matched;

[0022] B3. Obtain the similarity between the input multi-dimensional traffic scene data and the objects to be matched, remove the objects to be matched whose similarity values ​​are less than the similarity threshold, update the objects to be matched, and calculate the initial speed limit value of the road section according to the following formula:

[0023]

[0024] Among them, v q is the initial speed limit value of the qth road section, Sim m is the similarity value between the mth object to be matched and the input multivariate traffic scene data, v m is the actual speed limit value of the mth object to be matched on the qth road section, and M is the total number of objects to be matched;

[0025] B4. Arrange the initial speed limit values ​​obtained for all road sections according to the road section numbers to obtain the initial speed limit sequence of the current road section sequence.

[0026] Preferably, the similarity value between the mth object to be matched and the input multivariate traffic scene data is calculated according to the following formula:

[0027]

[0028] Among them, Sim m is the similarity value between the mth object to be matched and the input multivariate traffic scene data, w i is the weight value of the i-th feature in the multivariate traffic scene data, d(r,r m ) is the Euclidean distance between the input multivariate traffic scene data and the mth object to be matched.

[0029] Preferably, the preset adjacent road segment transfer model specifically includes:

[0030] S201, based on each adjacent road segment, obtaining the length of the transition zone of the adjacent road segment according to the initial speed limit values ​​determined for the two road segments;

[0031] S202: Based on the initial speed limit values ​​of the adjacent road sections, a speed limit gradient point sequence is set, specifically:

[0032]

[0033] Among them, v x is the speed limit value of the xth gradient point in the speed limit gradient point sequence, v q is the initial speed limit value of the qth road section, v q+1 is the initial speed limit value of the q+1th road section, X is the number of gradient points, which is determined according to the length of the transition zone and the preset sign spacing;

[0034] S203: Using the speed limit gradient point sequence as a speed limit transfer instruction, the speed limit transfer instruction is set at the first position in the current road segment that is the same distance from the starting point of the next road segment to the transition zone.

[0035] Preferably, the length of the transition zone between adjacent road sections is calculated according to the following formula:

[0036]

[0037] Among them, L trans is the length of the transition zone between adjacent road sections, v q is the initial speed limit value of the qth road section, v q+1 is the initial speed limit value of the q+1th road section, a max is the maximum safe deceleration, which is set according to the actual scene of the road section and international road safety regulations. react It is the preset standard driver reaction time to speed limit changes, set based on expert experience.

[0038] Preferably, in S203, after generating the speed limit transmission instruction, the method further includes:

[0039] S301, based on each driver, obtaining a driving profile feature to reflect the driver's safety awareness level, obtained based on a reaction factor and an aggressiveness coefficient;

[0040] S302: Determine an adjustment coefficient value for the driver based on the driving profile characteristics and a preset profile type-adjustment coefficient strategy. The preset profile type-adjustment coefficient strategy includes: profile types including low safety awareness, moderate safety awareness, and high safety awareness, which are denoted as type 1, type 2, and type 3, respectively. The adjustment coefficient for type 1 is set to 1.2, the adjustment coefficient for type 2 is set to 1, and the adjustment coefficient for type 3 is set to 0.8.

[0041] Determining the adjustment coefficient value of the driver includes:

[0042] The driving profile features are compared with the preset feature threshold interval [a1, a2]. If the driving profile features are greater than the upper limit of the interval, it is determined to be the third type; if the driving profile features are less than the lower limit of the interval, it is determined to be the first type; otherwise, it is determined to be the second type; among them, 0<a1<a2, a1<a2<1, and the specific values ​​are set according to actual needs and expert experience.

[0043] S303, multiplying the original transition zone length by the adjustment coefficient to obtain the transition zone length corresponding to the driver, repeating steps S202 to S203, updating the speed limit transmission instruction of the driver, and synchronizing it to the navigation APP when the driver passes the first position.

[0044] Preferably, the driving profile feature is calculated according to the following formula:

[0045]

[0046] e2=tanh(0.1×N hard )

[0047] e=e -(0.6×e1+0.4×e2)

[0048] Among them, e1 is the reaction factor, t hist is the driver’s historical reaction time to speed limit changes, t react is the preset standard reaction time, e2 is the radical coefficient, N hard is the number of sudden brakes of the driver in the historical time window, and e is the driving profile feature of the driver.

[0049] Preferably, before S103, the method further includes:

[0050] S401, generating a trajectory heat map of the road section within a preset time window based on the roadside sensors of each road section;

[0051] S402: Input the trajectory heat map and multivariate traffic scenario data of the preset time window into a pre-trained accident potential probability prediction model, and output the current accident probability value of the road section;

[0052] S403: Based on the accident probability value of each road section, the initial speed limit is dynamically modified to obtain a target speed limit value to replace the original initial speed limit value.

[0053] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0054] Comprehensively collect real-time traffic information from all sections within the jurisdiction of the expressway to provide a data basis for subsequent speed limit management. By obtaining dynamic characteristics of traffic flow, static characteristics of sections and scene characteristics, it is possible to characterize the traffic conditions and environmental factors of the sections from multiple dimensions; Dynamic characteristics of traffic flow: The average and variance values ​​of vehicle distance, the average and variance values ​​of vehicle speed, etc. can reflect the congestion level and driving stability of traffic flow, and the mean speed difference between adjacent lanes and the vehicle crossing rate can reflect the traffic coordination between different lanes; Static characteristics of sections: Road design standards reflect the inherent properties of roads, such as road type, number of lanes, etc. These factors will affect the driving characteristics and speed limit requirements of vehicles; Scene characteristics: Environmental information such as rainfall and visibility has an important impact on traffic safety.

[0055] By using historical data and similarity matching methods, a reasonable initial speed limit value is determined for the current road section sequence. By selecting historical data with low accident rates as matching objects, the reliability and safety of the speed limit value can be improved. The historical speed limit scheme database contains rich historical multi-traffic scenario data and its corresponding actual speed limits and accident rates, providing a reference basis for speed limit matching. By selecting historical data with accident rates below a preset accident threshold, speed limit schemes that may cause accidents can be eliminated, thereby improving the accuracy of the initial speed limit value. Weights are assigned to dynamic characteristics of traffic flow, static characteristics of road sections, and scene characteristics, respectively, reflecting the degree of influence of different characteristics on the speed limit value. By calculating the similarity value between the input multi-traffic scenario data and the matching objects, and eliminating matching objects with similarity values ​​below the similarity threshold, the historical data most similar to the traffic conditions of the current road section can be found, thereby determining a reasonable initial speed limit value. The initial speed limit values ​​obtained for all sections are arranged according to the section number, forming an initial speed limit sequence for the current road section sequence, providing a basis for subsequent speed limit transmission. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 2 is a flow chart of a variable speed limit management method for expressways according to an embodiment of the present invention. DETAILED DESCRIPTION

[0057] To facilitate understanding of the present invention, the present application will be described more comprehensively below with reference to the relevant drawings; the drawings show preferred embodiments of the present invention, but the present invention can be implemented in many different forms and is not limited to the embodiments described herein; on the contrary, the purpose of providing these embodiments is to enable a more thorough and comprehensive understanding of the disclosed content of the present invention.

[0058] It should be noted that the terms “vertical”, “horizontal”, “up”, “down”, “left”, “right” and similar expressions used in this document are for illustrative purposes only and do not represent the only implementation method.

[0059] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this invention pertains; the terms used herein in the specification of the present invention are for the purpose of describing specific embodiments only and are not intended to limit the present invention; the term "and / or" used herein includes any and all combinations of one or more of the associated listed items.

[0060] Example 1: Figure 1 It is a flowchart of a variable speed limit management method for expressways according to an embodiment of the present invention.

[0061] like Figure 1 As shown, a variable speed limit management method for a highway includes the following steps:

[0062] S101, obtaining multivariate traffic scene data of a road section sequence within a preset time window within the jurisdiction of the expressway, wherein the road section sequence is composed of a number of road sections that are sequentially divided according to road design standards within the jurisdiction of the expressway, and the multivariate traffic scene data includes traffic flow dynamic characteristics, road section static characteristics, and scene characteristics of each road section in the road section sequence.

[0063] Specifically, road design standards include but are not limited to road type, number of lanes, lane width, slope and curvature. The highway jurisdiction area is divided into several sections based on the road design standards. It can also be understood that within the jurisdiction area of ​​a one-way highway, the road is divided in sequence, and each division represents a possible change in the speed limit value.

[0064] Specifically, the multiple traffic scenes of each road section can be obtained based on sensor monitoring, such as radar monitoring, camera monitoring, weather station monitoring, etc., which will not be elaborated in the present invention.

[0065] The method for obtaining the multivariate traffic scene data of each road section is as follows:

[0066] A1. Obtain dynamic traffic flow characteristics to reflect the real-time traffic status within a time window. These characteristics are: the mean and variance of vehicle headway within different lanes within a road section within a preset time window; the mean and variance of vehicle speed (the mean and fluctuation of the distance between front and rear vehicles within the same lane, and the mean and dispersion of vehicle speed within the same lane); the mean speed difference between adjacent lanes; and the vehicle crossing rate (the mean absolute speed difference between adjacent lanes and the temporal and spatial proportion of longitudinal overlap of vehicles in adjacent lanes).

[0067] It should be noted that the vehicle crossing rate between adjacent lanes is calculated according to the following formula:

[0068]

[0069] Among them, V is the vehicle crossing rate between adjacent lanes, N is the number of lanes in the road section, CR i is the vehicle crossing rate between the i-th lane and the adjacent lane.

[0070]

[0071] Among them, CR i is the vehicle crossing rate between the i-th lane and the adjacent lane, k is the number of adjacent lanes of the i-th lane (k = 1, 2), T is the size of the preset time window, L i is the set of vehicles in the i-th lane, L j is the set of vehicles in the jth adjacent lane, T a,b It is a condition indicator function. When the preset condition is met, the value is assigned to the duration of the preset condition, otherwise it is 0. The preset condition is set to |x a (t)-x b (t)|≤L a ,|x a (t)-x b (t)| represents the distance between the front ends of vehicles a and b in the direction of travel at time t, L a is the body length of vehicle a.

[0072] A2. Obtain static characteristics of the road section to reflect inherent properties of the road, and set them as: road design standards for the corresponding road section.

[0073] A3. Obtain scene features to reflect environmental information within the time window, which are set to: rainfall and visibility of the corresponding road section.

[0074] S102 : Based on the multi-dimensional traffic scene data, an initial speed limit sequence of the current road section sequence is obtained using a preset road section speed limit matching mechanism.

[0075] It should be noted that each road segment in the road segment sequence may include more than one lane. For multi-lane roads, the speed limit values ​​are assigned in sequence according to the horizontal order. Therefore, the initial speed limit sequence may also be represented in k×Q dimensions, where k represents the number of lanes in the road segment and Q represents the total number of road segments.

[0076] Specifically, the multivariate traffic scene data of each road section is quantified and expressed as r q =[r1,r2,r3] q , q is the number of the road segment in the road segment sequence, r q is the quantized multivariate traffic scenario data for the qth road segment, r1 is the quantized dynamic characteristics of the traffic flow, r2 is the quantized static characteristics of the road segment, and r3 is the quantized scene characteristics. Quantization refers to encoding non-numeric content. For example, each road type corresponds to a unique code value for identification.

[0077] In some embodiments, the preset road section speed limit matching mechanism includes:

[0078] B1. Input the multi-variate traffic scenario data into a pre-built historical speed limit solution database. The pre-built historical speed limit solution database includes a number of historical multi-variate traffic scenario data within the jurisdiction of the expressway and their corresponding actual speed limit values ​​and accident rates.

[0079] B2. Filter out historical multivariate traffic scene data with an accident rate less than a preset accident threshold (set according to actual conditions and expert experience, for example, set to 0.2) as objects to be matched.

[0080] B3. Obtain the similarity between the input multi-dimensional traffic scene data and the objects to be matched, remove the objects to be matched whose similarity values ​​are less than the similarity threshold, update the objects to be matched, and calculate the initial speed limit value of the road section according to the following formula:

[0081]

[0082] Among them, v q is the initial speed limit value of the qth road section, Sim m is the similarity value between the mth object to be matched and the input multivariate traffic scene data, v m is the actual speed limit value of the mth object to be matched corresponding to the qth road section, and M is the total number of objects to be matched.

[0083] It should be noted that when calculating the similarity value between the input multivariate traffic scene data and the object to be matched, corresponding weight values ​​are assigned to the dynamic characteristics of traffic flow, static characteristics of road sections and scene characteristics in the multivariate traffic scene data, which are respectively used to indicate the degree of influence of the dynamic characteristics of traffic flow, static characteristics of road sections and scene characteristics on the similarity. For example, static characteristics of road sections and scene characteristics are more important than dynamic characteristics of traffic flow, because similar road section characteristics and environment are extremely important for the determination of speed limit values. Therefore, the weight values ​​of dynamic characteristics of traffic flow, static characteristics of road sections and scene characteristics are set to 0.2, 0.4 and 0.4 respectively.

[0084] For example, the similarity value is calculated as follows:

[0085]

[0086] Among them, Sim m is the similarity value between the mth object to be matched and the input multivariate traffic scene data, w i is the weight value of the i-th feature in the multivariate traffic scene data, d(r,r m ) is the Euclidean distance between the input multivariate traffic scene data and the mth object to be matched.

[0087] B4. Arrange the initial speed limit values ​​obtained for all road sections according to the road section numbers to obtain the initial speed limit sequence of the current road section sequence.

[0088] S103: Based on a preset adjacent road section transfer model, a speed limit transfer instruction is generated for each road section and synchronized to the road section (which can be issued through a variable electronic sign or a navigation APP).

[0089] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages:

[0090] Comprehensively collect real-time traffic information from all sections within the jurisdiction of the expressway to provide a data basis for subsequent speed limit management. By obtaining dynamic characteristics of traffic flow, static characteristics of sections and scene characteristics, it is possible to characterize the traffic conditions and environmental factors of the sections from multiple dimensions; Dynamic characteristics of traffic flow: The average and variance values ​​of vehicle distance, the average and variance values ​​of vehicle speed, etc. can reflect the congestion level and driving stability of traffic flow, and the mean speed difference between adjacent lanes and the vehicle crossing rate can reflect the traffic coordination between different lanes; Static characteristics of sections: Road design standards reflect the inherent properties of roads, such as road type, number of lanes, etc. These factors will affect the driving characteristics and speed limit requirements of vehicles; Scene characteristics: Environmental information such as rainfall and visibility has an important impact on traffic safety.

[0091] By using historical data and similarity matching methods, a reasonable initial speed limit value is determined for the current road section sequence. By selecting historical data with low accident rates as matching objects, the reliability and safety of the speed limit value can be improved. The historical speed limit scheme database contains rich historical multi-traffic scenario data and its corresponding actual speed limits and accident rates, providing a reference basis for speed limit matching. By selecting historical data with accident rates below a preset accident threshold, speed limit schemes that may cause accidents can be eliminated, thereby improving the accuracy of the initial speed limit value. Weights are assigned to dynamic characteristics of traffic flow, static characteristics of road sections, and scene characteristics, respectively, reflecting the degree of influence of different characteristics on the speed limit value. By calculating the similarity value between the input multi-traffic scenario data and the matching objects, and eliminating matching objects with similarity values ​​below the similarity threshold, the historical data most similar to the traffic conditions of the current road section can be found, thereby determining a reasonable initial speed limit value. The initial speed limit values ​​obtained for all sections are arranged according to the section number, forming an initial speed limit sequence for the current road section sequence, providing a basis for subsequent speed limit transmission.

[0092] Embodiment 2: Further limit the adjacent road segment transfer model in embodiment 1.

[0093] Therefore, the embodiments of the present application are optimized based on the above embodiments.

[0094] In some embodiments, the preset adjacent road segment transfer model specifically includes:

[0095] S201: Based on each adjacent road segment, the length of the transition zone between the adjacent road segments is obtained according to the initial speed limit values ​​determined for the two road segments, specifically:

[0096]

[0097] Among them, L trans is the length of the transition zone between adjacent road sections, v q is the initial speed limit value of the qth road section, v q+1 is the initial speed limit value of the q+1th road section, a max The maximum safe deceleration is set according to the actual scene of the road section (scene characteristics, rainfall, etc.) and international road safety regulations. For example, the standard maximum deceleration can be set to -3m / s 2 ;t react It is a preset standard reaction time of the driver to the change of the speed limit value, which can be set based on historical experience and expert experience. Indicates that from v q Change to v q+1 Minimum distance required.

[0098] S202: Based on the initial speed limit values ​​of the adjacent road sections, a speed limit gradient point sequence is set, specifically:

[0099]

[0100] Among them, v x is the speed limit value of the xth gradient point in the speed limit gradient point sequence, v q is the initial speed limit value of the qth road section, v q+1 is the initial speed limit value of the q+1th road section, X is the number of gradient points, which is determined according to the length of the transition zone and the preset sign spacing. It can be set to the rounded-up value of the ratio of the transition zone length to the sign spacing.

[0101] For example, the number of gradient points is determined according to the following formula: X is the number of gradient points, L trans is the length of the transition zone, L min It is the preset mark spacing, which is set according to human experience and actual conditions, and is used to indicate the speed from v q Quickly switch to v q+1 where ρ is the traffic flow density (number of vehicles) of the qth road segment, and α is a preset density coefficient, which is set based on actual conditions and experience. For example, it is set to 0.05. [] indicates rounding up.

[0102] S203: Use the speed limit gradient point sequence as a speed limit transfer instruction, and set the speed limit transfer instruction at the first position of the transition zone length between the starting point of the next section in the current section. The speed limit transfer instruction can also be synchronized to the navigation APP when the vehicle passes this position.

[0103] Specifically, the distribution of each gradient point in the speed-limiting gradient point sequence over the transition zone length is as follows: a nonlinear distribution model is used to achieve a distribution of gradient points that is dense at the front and sparse at the back: Among them, L x The distance value from the first position in the transition zone length interval where the x-th gradient point is located, L trans is the transition zone length, and β is a preset distribution coefficient, for example, set to 1.8.

[0104] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages:

[0105] Accurately calculating the length of the transition zone between adjacent road sections is key to achieving a smooth transition in speed limits. This length takes into account the difference in initial speed limits between adjacent sections, the maximum safe deceleration, and the driver's standard reaction time to speed limit changes, ensuring that the driver has sufficient time and distance to adapt to speed limit changes. By incorporating the difference in initial speed limits between adjacent sections, the maximum safe deceleration, and the driver's standard reaction time into the calculation, the determination of the transition zone length is more scientific and reasonable. The maximum safe deceleration is set based on the actual road scenario (such as rainfall) and international road safety regulations, fully considering the impact of different environmental conditions on vehicle braking performance. In severe weather conditions such as rainfall, the maximum safe deceleration will be appropriately reduced, thereby increasing the length of the transition zone to ensure driving safety.

[0106] The speed limit gradient point sequence provides drivers with clear guidance on speed limit changes, helping them gradually adapt to speed limit changes. By properly determining the number of gradient points and speed limit values, speed limit transitions can be smoother, reducing operational difficulty and discomfort for the driver. The number of gradient points is determined based on the length of the transition zone and the preset sign spacing, while also taking into account the traffic flow density of the section and the preset density coefficient. This dynamic determination method allows the number of gradient points to be flexibly adjusted according to actual conditions, ensuring a reasonable distribution of speed limit signs within the transition zone and improving the effectiveness of speed limit communication. For example, when traffic flow density is high, increasing the number of gradient points can more carefully guide drivers in adjusting their speed, avoiding accidents such as rear-end collisions caused by excessively rapid speed limit changes. The speed limit gradient point sequence allows drivers to smoothly adapt to speed limit changes by gradually adjusting the speed limit value. The speed limit value changes gradually from the initial speed limit value to the initial speed limit value of the next section, reducing the risk of driver errors caused by sudden speed limit changes.

[0107] Clarifying the location and method for publishing speed limit information ensures drivers can receive speed limit information in a timely and accurate manner. By placing a speed limit information instruction at the first location on the current road section that is the length of the transition zone from the start point of the next road section, or synchronizing the instruction to the navigation app when the vehicle passes through this location, the efficiency and coverage of speed limit information transmission are improved. Placing a speed limit information instruction at the first location on the current road section that is the length of the transition zone from the start point of the next road section allows drivers to understand speed limit changes in advance before entering the transition zone, giving them ample time to prepare. This precise publishing location can increase drivers' attention to and compliance with speed limit information. In addition to placing speed limit signs on road sections, synchronizing speed limit information instructions to the navigation app provides drivers with more channels for obtaining information. Drivers using navigation devices, in particular, can obtain real-time speed limit change information, further improving driving safety and traffic efficiency.

[0108] A nonlinear distribution model is used to achieve a denser distribution of gradient points at the front and sparser at the back, which better aligns with drivers' perception and adaptation to speed limit changes. At the front of the transition zone, drivers need more frequent speed limit information to quickly adjust their speed. At the back of the transition zone, drivers have gradually adapted to the speed limit change and can appropriately reduce the frequency of speed limit reminders. This denser distribution at the front and sparser at the back aligns with drivers' perception and adaptation to speed limit changes, improving their acceptance and understanding of speed limit information. At the front of the transition zone, drivers receive denser speed limit reminders and can quickly adjust their speed. At the back of the transition zone, speed limit reminders are reduced, minimizing driver intrusion. By rationally distributing gradient points along the length of the transition zone, unnecessary speed limit signs can be reduced, while ensuring effective speed limit information is conveyed, lowering construction and maintenance costs. This distribution also enhances the driver's driving experience and reduces visual fatigue caused by excessive speed limit information.

[0109] Example 3: Examples 1 and 2 primarily adopt a unified speed limit transmission strategy based on the overall perspective of the road section, without considering individual differences between drivers. In real-world traffic, drivers vary in their ability to react to speed limit changes and their safety awareness. A unified speed limit transmission strategy may not meet the needs of all drivers and may even cause some drivers to react negatively, compromising the effectiveness of speed limit management.

[0110] Therefore, the embodiments of the present application are optimized based on the above embodiments.

[0111] In some embodiments, after generating the speed limit transmission instruction in step S203, the following steps are further included:

[0112] S301: Based on each driver, a driving profile feature is obtained to reflect the driver's safety awareness level. The following is obtained based on the reaction factor and the aggressiveness coefficient:

[0113]

[0114] e2=tanh(0.1×N hard )

[0115] e=e -(0.6×e1+0.4×e2)

[0116] Among them, e1 is the reaction factor, t hist is the average of the driver's historical reaction time to speed limit changes (the average of the driver's reaction time to speed limit changes on the historical sections of the road he has traveled can also be obtained by retrieving the driver's historical driving records on the highway, and it can be determined based on actual conditions), t react is the preset standard reaction time, e2 is the radical coefficient, N hard is the number of sudden braking times of the driver in the historical time window (the sudden braking is defined as the acceleration change amplitude being greater than the preset change threshold), and e is the driving profile feature of the driver.

[0117] It should be noted that within the monitoring range of the current road section (ie, the qth road section), the driving vehicles entering the road section at the entrance are monitored, and step S301 is executed to obtain the driving portrait characteristics of the drivers entering the road section.

[0118] S302: Determine the adjustment coefficient value of the driver according to the driving profile characteristics and the preset profile type-adjustment coefficient strategy.

[0119] Specifically, the preset profile type-adjustment coefficient strategies include:

[0120] The profile types include low security awareness, moderate security awareness, and high security awareness, which are recorded as type 1, type 2, and type 3 respectively;

[0121] The adjustment coefficient of the first type is set to 1.2, the adjustment coefficient of the second type is set to 1, and the adjustment coefficient of the third type is set to 0.8.

[0122] Determining the adjustment coefficient value of the driver specifically includes:

[0123] The driving profile features are compared with the preset feature threshold interval [a1, a2]. If the driving profile features are greater than the upper limit of the interval, it is determined to be the third type and the corresponding adjustment coefficient is obtained; if the driving profile features are less than the lower limit of the interval, it is determined to be the first type and the corresponding adjustment coefficient is obtained; otherwise, it is determined to be the second type and the corresponding adjustment coefficient is obtained.

[0124] For example, 0<a1<a2, a1<a2<1, and specific values ​​are set according to actual needs and expert experience. For example, a1 is 0.4 and a2 is 0.6.

[0125] S303, multiplying the original transition zone length by the adjustment coefficient to obtain the transition zone length corresponding to the driver, repeating steps S202 to S203, updating the speed limit transmission instruction of the driver, and synchronizing it to the navigation APP when the driver passes the first position.

[0126] Therefore, the safety awareness levels of different drivers are differentiated, and the length of the transition interval set under standard conditions is adaptively adjusted. If the safety awareness is high, the length of the transition interval can be appropriately shortened to avoid the driver's rebellious psychology towards the premature appearance of speed limit instructions; otherwise, the length of the transition interval needs to be appropriately increased to improve traffic safety.

[0127] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages:

[0128] By collecting data such as the driver's historical reaction time to speed limit changes and the number of sudden brakes, the reaction factor and aggressiveness coefficient are calculated, and then the driving profile characteristics are obtained to quantitatively reflect the driver's level of safety awareness. This step provides a basic basis for subsequently formulating personalized speed limit delivery strategies for drivers with different levels of safety awareness. The driver's safety awareness level is converted into a specific numerical value (driving profile characteristics) to make the driver assessment more objective and accurate. For example, the shorter the reaction time and the fewer the number of sudden brakes, the lower the driving profile characteristic value may be, indicating a higher level of safety awareness. This provides the prerequisite for subsequently determining the adjustment coefficient value based on the driving profile characteristics, which helps to adopt different speed limit delivery strategies for drivers with different levels of safety awareness.

[0129] Through the preset portrait type-adjustment coefficient strategy, the driving portrait characteristics are compared with the preset feature threshold range to determine the driver's portrait type and obtain the corresponding adjustment coefficient value. This step realizes the classification of drivers with different levels of safety awareness and provides key parameters for personalized adjustment of the transition zone length.

[0130] For drivers with high safety awareness, the transition zone length is appropriately reduced to prevent them from prematurely reacting to speed limit instructions without compromising their safe driving. For drivers with low safety awareness, the transition zone length is appropriately increased to give them more time and distance to adapt to speed limit changes, thereby improving traffic safety. By adjusting the transition zone length differently, drivers can be better guided to comply with speed limit regulations, reducing violations such as speeding, thereby improving the safety and efficiency of the entire transportation system.

[0131] Embodiment 4: Embodiments 1, 2 and 3 mainly focus on the transmission of speed limits between adjacent road sections and speed limit management based on individual driver differences, without considering that the actual current traffic conditions of the road sections may reflect potential future accident risks.

[0132] Therefore, the embodiments of the present application are optimized based on the above embodiments.

[0133] In some embodiments, before step S103, the method further includes:

[0134] S401 : Based on the roadside sensors of each road section, a trajectory heat map of the road section within a preset time window is generated.

[0135] Specifically include:

[0136] C1. Use roadside sensors (such as cameras, radars, etc.) to collect vehicle trajectory data for each road section within a preset time window (for example, 5 minutes). The trajectory data includes information such as the vehicle's location and timestamp.

[0137] C2. Divide the road section into several small grid units. For example, the size of each grid unit is 5m×5m.

[0138] C3. For each grid cell, count the number of vehicles passing through the grid cell within a preset time window and the acceleration values ​​of the vehicles passing through the grid cell.

[0139] C4. Color each grid cell according to the number of vehicles. The more vehicles there are, the darker the color is, thus generating a trajectory heat map. The trajectory heat map can also be represented as a matrix H, where H ij represents an array consisting of the vehicle count statistics and the average acceleration values ​​of the vehicles in the grid cell of the i-th row and j-th column, i = 1, 2, ..., h, j = 1, 2, ..., g, h and g are the grid cell numbers in the row and column directions of the trajectory heat map respectively, H ij =[n ij ,a ij ],n ij is the vehicle quantity statistics of the grid cell in row i and column j, a ij is the average acceleration of the vehicle in the grid cell in row i and column j.

[0140] S402: The trajectory heat map of the preset time window and the multivariate traffic scene data are input as input features into a pre-trained accident potential probability prediction model, and the current accident probability value of the road section is output.

[0141] Specifically, the pre-trained accident potential probability prediction model is used to:

[0142] D1. Collect a large number of historically recorded traffic accident trajectory heat maps within a preset time window before the occurrence of traffic accidents, and accident precursor features composed of multi-faceted traffic scene data;

[0143] D2. Traverse all historical accident precursor features and calculate the similarity between them and the input features (the trajectory heat map and multi-dimensional traffic scene data are assigned corresponding weights, and the similarities are weighted summed. The weights are set according to the actual judgment situation); the similarity calculation method refers to the relevant existing technology and is not described in detail in this invention;

[0144] D3. The maximum similarity value is used as the current accident probability value of the road section and is output.

[0145] S403: Based on the accident probability value of each road section, the initial speed limit is dynamically modified to obtain a target speed limit value to replace the original initial speed limit value.

[0146] Specifically, if the accident probability value is greater than a preset probability threshold (for example, set to 0.6, based on expert experience and actual conditions), the initial speed limit corresponding to the road section will be corrected based on the difference between the accident probability value and the probability threshold. The larger the difference, the greater the probability of an accident. Therefore, the ratio of the difference to the probability threshold is used as the correction factor for the initial speed limit. The initial speed limit is multiplied by the correction factor to obtain a new initial speed limit, which replaces the original initial speed limit for the road section. It can be expressed as: v q ' is the new initial speed limit value after correction for the qth road section, v q is the original initial speed limit value of the qth road section, p is the accident probability value of the road section, and p0 is the probability threshold.

[0147] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages:

[0148] Roadside sensors (such as cameras and radar) are used to collect vehicle trajectory data for each road section within a preset time window (e.g., 5 minutes). This data includes information such as vehicle location and timestamps, providing basic data for subsequent analysis of traffic conditions on the road section. The road section is divided into several small grid cells, and the number of vehicles passing through each grid cell within the preset time window and the acceleration values ​​of vehicles passing through the grid cell are counted. This allows for a more detailed understanding of traffic flow and vehicle driving conditions at different locations on the road section. Trajectory heat maps clearly show which areas within the road section are densely populated and which are sparsely populated, providing traffic management departments with an intuitive display of traffic conditions and helping to promptly identify traffic congestion points or abnormal areas. The trajectory heat map is represented as a matrix H, which contains an array of vehicle statistics and average vehicle acceleration values ​​for each grid cell, providing detailed data support for subsequent calculations of potential accident probabilities.

[0149] The accident potential probability prediction model can predict the potential probability of an accident based on the traffic conditions of the current road section. The model takes into account trajectory heat maps and multi-dimensional traffic scenario data when calculating similarity and assigns different weight values. This can more comprehensively reflect the various factors that affect the occurrence of accidents and improve the accuracy of predictions.

[0150] Based on the accident probability value for each road section, the accident risk of that section is determined. If the accident probability value exceeds a preset probability threshold, the section is considered to have a high accident risk. Based on the difference between the accident probability value and the probability threshold, the initial speed limit corresponding to that section is adjusted. The larger the difference, the greater the probability of an accident, the smaller the correction factor, and the lower the new initial speed limit. This lowering of the speed limit reduces the likelihood of accidents. By dynamically adjusting the initial speed limit, the speed limit can be adjusted in real time based on the actual accident risk of the road section, reducing vehicle speeds, reducing traffic accidents, and improving traffic safety. This dynamic speed limit adjustment method allows for flexible response to changing traffic conditions. When traffic volume increases and the accident risk increases, the speed limit can be lowered promptly. When traffic conditions improve, the speed limit can be increased appropriately, improving road traffic efficiency.

[0151] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Various modifications and variations are readily apparent to those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A variable speed limit management method for highways, characterized in that: The method comprises: S101, obtaining multivariate traffic scene data of a road section sequence within a predetermined time window within a jurisdictional area of ​​an expressway, wherein the multivariate traffic scene data includes dynamic traffic flow characteristics, static road section characteristics, and scene characteristics of each road section in the road section sequence; S102, based on the multi-dimensional traffic scene data, using a preset road section speed limit matching mechanism, obtaining an initial speed limit sequence for the current road section sequence; S103: Based on a preset adjacent road section transfer model, a speed limit transfer instruction is generated for each road section and synchronized to the road section.

2. The variable speed limit management method for highways according to claim 1, characterized in that: The road segment sequence is composed of several road segments that are sequentially divided according to road design standards within the jurisdiction of the expressway; the multivariate traffic scene data of each road segment is obtained as follows: A1. Obtain traffic flow dynamic characteristics, which are set as follows: the mean and variance values ​​of vehicle headway within different lanes within a road section, the mean and variance values ​​of vehicle speed, the mean speed difference between adjacent lanes, and the vehicle crossing rate within a preset time window; A2. Obtain static characteristics of the road section to reflect the inherent properties of the road, which are set as: the road design standard of the corresponding road section; A3. Obtain scene features to reflect environmental information within the time window, which are set to: rainfall and visibility of the corresponding road section.

3. The variable speed limit management method for highways according to claim 2, characterized in that: The vehicle crossing rate between adjacent lanes is calculated according to the following formula: Among them, V is the vehicle crossing rate between adjacent lanes, N is the number of lanes in the road section, CR i is the vehicle crossing rate between the i-th lane and the adjacent lane; where the vehicle crossing rate between the i-th lane and the adjacent lane is expressed as: Among them, CR i is the vehicle crossing rate between the i-th lane and the adjacent lane, k is the number of adjacent lanes of the i-th lane, T is the size of the preset time window, L i is the set of vehicles in the i-th lane, L j is the set of vehicles in the jth adjacent lane, T a,b It is a condition indicator function. When the preset condition is met, the value is assigned to the duration of the preset condition, otherwise it is 0. The preset condition is set to |x a (t)-x b (t)|≤L a ,|x a (t)-x b (t)| represents the distance between the front ends of vehicles a and b in the direction of travel at time t, L a is the body length of vehicle a.

4. The variable speed limit management method for highways according to claim 3, characterized in that: The preset road section speed limit matching mechanism includes: B1. Input the multi-dimensional traffic scenario data into a pre-built historical speed limit solution database. The pre-built historical speed limit solution database includes a number of historical multi-dimensional traffic scenario data within the jurisdiction of the expressway and their corresponding actual speed limits and accident rates. B2. Filter out historical multivariate traffic scene data with an accident rate less than a preset accident threshold as the object to be matched; B3. Obtain the similarity between the input multi-dimensional traffic scene data and the objects to be matched, remove the objects to be matched whose similarity values ​​are less than the similarity threshold, update the objects to be matched, and calculate the initial speed limit value of the road section according to the following formula: Among them, v q is the initial speed limit value of the qth road section, Sim m is the similarity value between the mth object to be matched and the input multivariate traffic scene data, v m is the actual speed limit value of the mth object to be matched on the qth road section, and M is the total number of objects to be matched; B4. Arrange the initial speed limit values ​​obtained for all road sections according to the road section numbers to obtain the initial speed limit sequence of the current road section sequence.

5. The variable speed limit management method for highways according to claim 4, characterized in that: The similarity value between the mth object to be matched and the input multivariate traffic scene data is calculated according to the following formula: Among them, Sim m is the similarity value between the mth object to be matched and the input multivariate traffic scene data, w i is the weight value of the i-th feature in the multivariate traffic scene data, d(r,r m ) is the Euclidean distance between the input multivariate traffic scene data and the mth object to be matched.

6. The variable speed limit management method for expressways according to claim 1, wherein: The preset adjacent road segment transfer model specifically includes: S201, based on each adjacent road segment, obtaining the length of the transition zone of the adjacent road segment according to the initial speed limit values ​​determined for the two road segments; S202: Based on the initial speed limit values ​​of the adjacent road sections, a speed limit gradient point sequence is set, specifically: Among them, v x is the speed limit value of the xth gradient point in the speed limit gradient point sequence, v q is the initial speed limit value of the qth road section, v q+1 is the initial speed limit value of the q+1th road section, X is the number of gradient points, which is determined according to the length of the transition zone and the preset sign spacing; S203: Using the speed limit gradient point sequence as a speed limit transfer instruction, the speed limit transfer instruction is set at the first position in the current road segment that is the same distance from the starting point of the next road segment to the transition zone.

7. The variable speed limit management method for highways according to claim 6, characterized in that: The length of the transition zone between adjacent road sections is calculated according to the following formula: Among them, L trans is the length of the transition zone between adjacent road sections, v q is the initial speed limit value of the qth road section, v q+1 is the initial speed limit value of the q+1th road section, a max is the maximum safe deceleration, which is set according to the actual scene of the road section and international road safety regulations. react It is the preset standard driver reaction time to speed limit changes, set based on expert experience.

8. The variable speed limit management method for highways according to claim 6, characterized in that: In S203, after the speed limit transmission instruction is generated, the following steps are further included: S301, based on each driver, obtaining a driving profile feature to reflect the driver's safety awareness level, obtained based on a reaction factor and an aggressiveness coefficient; S302: Determine an adjustment coefficient value for the driver based on the driving profile characteristics and a preset profile type-adjustment coefficient strategy. The preset profile type-adjustment coefficient strategy includes: profile types including low safety awareness, moderate safety awareness, and high safety awareness, which are denoted as type 1, type 2, and type 3, respectively. The adjustment coefficient for type 1 is set to 1.2, the adjustment coefficient for type 2 is set to 1, and the adjustment coefficient for type 3 is set to 0.

8. Determining the adjustment coefficient value of the driver includes: The driving profile features are compared with the preset feature threshold interval [a1, a2]. If the driving profile features are greater than the upper limit of the interval, it is determined to be the third type; if the driving profile features are less than the lower limit of the interval, it is determined to be the first type; otherwise, it is determined to be the second type; among them, 0<a1<a2, a1<a2<1, and the specific values ​​are set according to actual needs and expert experience. S303, multiplying the original transition zone length by the adjustment coefficient to obtain the transition zone length corresponding to the driver, repeating steps S202 to S203, updating the speed limit transmission instruction of the driver, and synchronizing it to the navigation APP when the driver passes the first position.

9. The variable speed limit management method for expressways according to claim 8, characterized in that: The driving profile feature is calculated according to the following formula: e2=tanh(0.1×N hard ) and=and -(0.6×e1+0.4×e2) Among them, e1 is the reaction factor, t hist is the driver’s historical reaction time to speed limit changes, t react is the preset standard reaction time, e2 is the radical coefficient, N hard is the number of sudden brakes of the driver in the historical time window, and e is the driving profile feature of the driver.

10. The variable speed limit management method for expressways according to claim 1, wherein: Before S103, the method further includes: S401, generating a trajectory heat map of the road section within a preset time window based on the roadside sensors of each road section; S402: Input the trajectory heat map and multivariate traffic scenario data of the preset time window into a pre-trained accident potential probability prediction model, and output the current accident probability value of the road section; S403: Based on the accident probability value of each road section, the initial speed limit is dynamically modified to obtain a target speed limit value to replace the original initial speed limit value.

Images