CN117804490B - Comprehensive planning method and device for vehicle running route - Google Patents
Comprehensive planning method and device for vehicle running route Download PDFInfo
- Publication number
- CN117804490B CN117804490B CN202410222721.0A CN202410222721A CN117804490B CN 117804490 B CN117804490 B CN 117804490B CN 202410222721 A CN202410222721 A CN 202410222721A CN 117804490 B CN117804490 B CN 117804490B
- Authority
- CN
- China
- Prior art keywords
- information
- route
- processing
- traffic
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 50
- 238000012545 processing Methods 0.000 claims abstract description 161
- 230000006399 behavior Effects 0.000 claims abstract description 99
- 238000005457 optimization Methods 0.000 claims abstract description 78
- 238000004458 analytical method Methods 0.000 claims abstract description 30
- 238000004088 simulation Methods 0.000 claims abstract description 29
- 230000007613 environmental effect Effects 0.000 claims abstract description 18
- 238000010276 construction Methods 0.000 claims abstract description 15
- 238000011156 evaluation Methods 0.000 claims abstract description 11
- 238000004422 calculation algorithm Methods 0.000 claims description 44
- 238000009826 distribution Methods 0.000 claims description 29
- 230000008569 process Effects 0.000 claims description 21
- 230000006870 function Effects 0.000 claims description 16
- 239000013598 vector Substances 0.000 claims description 15
- 230000003993 interaction Effects 0.000 claims description 13
- 238000005206 flow analysis Methods 0.000 claims description 12
- 230000000694 effects Effects 0.000 claims description 10
- 238000000547 structure data Methods 0.000 claims description 10
- 238000013178 mathematical model Methods 0.000 claims description 9
- 230000035945 sensitivity Effects 0.000 claims description 9
- 238000002922 simulated annealing Methods 0.000 claims description 9
- 230000002068 genetic effect Effects 0.000 claims description 8
- 238000013507 mapping Methods 0.000 claims description 8
- 239000000446 fuel Substances 0.000 claims description 7
- 230000008859 change Effects 0.000 claims description 6
- 238000005520 cutting process Methods 0.000 claims description 6
- 238000010586 diagram Methods 0.000 claims description 6
- 238000000513 principal component analysis Methods 0.000 claims description 6
- 230000002787 reinforcement Effects 0.000 claims description 6
- 238000012300 Sequence Analysis Methods 0.000 claims description 3
- 238000005728 strengthening Methods 0.000 claims 1
- 230000003542 behavioural effect Effects 0.000 description 7
- 230000008901 benefit Effects 0.000 description 4
- 238000010801 machine learning Methods 0.000 description 4
- 238000007726 management method Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 238000007621 cluster analysis Methods 0.000 description 2
- 230000001934 delay Effects 0.000 description 2
- 230000008451 emotion Effects 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000003909 pattern recognition Methods 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 238000012731 temporal analysis Methods 0.000 description 2
- 238000000700 time series analysis Methods 0.000 description 2
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000000137 annealing Methods 0.000 description 1
- 230000003466 anti-cipated effect Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 239000003795 chemical substances by application Substances 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000008846 dynamic interplay Effects 0.000 description 1
- 230000002996 emotional effect Effects 0.000 description 1
- 238000003912 environmental pollution Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 238000007789 sealing Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Landscapes
- Traffic Control Systems (AREA)
Abstract
The invention provides a comprehensive planning method and device for a vehicle running route, and relates to the technical field of path planning, wherein the method comprises the steps of acquiring first information and second information; performing behavior pattern learning processing according to the first information to obtain third information, wherein the third information is a behavior pattern model for predicting the behavior preference of the driver under different conditions; performing network construction processing according to the second information and the third information, and obtaining fourth information through key node importance evaluation and traffic influence analysis; according to the fourth information, carrying out driver decision prediction processing and collective route optimization processing to obtain fifth information; and carrying out path optimization processing on all the running routes in the fifth information, and carrying out comprehensive simulation processing on the optimized routes to obtain sixth information. According to the invention, by comprehensively considering the physical characteristics of the road, the environmental factors and the behavior mode of the driver, the route planning is optimized, the traffic jam and the vehicle emission are effectively reduced, and the efficiency of the traffic system is improved.
Description
Technical Field
The invention relates to the technical field of path planning, in particular to a comprehensive planning method and device for a vehicle running route.
Background
With the global enhancement of sustainable development and environmental protection consciousness, green travel becomes an important issue in the field of modern traffic planning. Green travel aims to mitigate the impact on the environment by optimizing traffic routes, reducing congestion, and reducing vehicle emissions. However, current vehicle travel route planning techniques face challenges, particularly in adapting to complex urban environments and achieving green travel goals. Existing vehicle travel route planning methods rely on underlying map services and traffic flow data, focusing mainly on the shortest or fastest route, with less consideration on environmental impact, physical characteristics of the road, and driver behavior preferences. These methods generally employ static planning methods, and lack adaptability to dynamic changes in road conditions and traffic conditions, so that they have limited effects in supporting green travel and effectively coping with peak traffic congestion.
In order to overcome the defects of the prior art, a comprehensive planning method and device for a vehicle running route are needed.
Disclosure of Invention
The invention aims to provide a comprehensive planning method and device for a vehicle running route so as to solve the problems. In order to achieve the above purpose, the technical scheme adopted by the invention is as follows:
in a first aspect, the present application provides a method for comprehensive planning of a vehicle running route, including: acquiring first information and second information, wherein the first information comprises behavior data of at least two drivers in an area to be planned, the second information comprises real-time traffic network state data, environment data and road network structure data, and the behavior data comprises a departure point, a destination, travel time and historical route selection of each driver;
performing behavior pattern learning processing according to the first information to obtain third information, wherein the third information is a behavior pattern model for predicting the behavior preference of the driver under different conditions;
Performing network construction processing according to the second information and the third information, and obtaining fourth information through key node importance evaluation and traffic influence analysis, wherein the fourth information is a network flow analysis result containing road network efficiency and robustness;
carrying out driver decision prediction processing and collective route optimization processing according to the fourth information to obtain fifth information, wherein the fifth information is a vehicle running route scheme which comprises a running route of each driver;
And carrying out path optimization processing on all the running routes in the fifth information, and carrying out comprehensive simulation processing on the optimized routes to obtain sixth information, wherein the sixth information is a final vehicle running route scheme.
In a second aspect, the present application also provides an integrated planning apparatus for a vehicle running route, including: the system comprises a first acquisition unit, a second acquisition unit and a control unit, wherein the first acquisition unit is used for acquiring first information and second information, the first information comprises behavior data of at least two drivers in an area to be planned, the second information comprises real-time traffic network state data, environment data and road network structure data, and the behavior data comprises a departure point, a destination, travel time and historical route selection of each driver;
The first processing unit is used for performing behavior pattern learning processing according to the first information to obtain third information, wherein the third information is a behavior pattern model for predicting the behavior preference of the driver under different conditions;
The second processing unit is used for carrying out network construction processing according to the second information and the third information, and obtaining fourth information through key node importance evaluation and traffic influence analysis, wherein the fourth information is a network flow analysis result comprising road network efficiency and robustness;
The third processing unit is used for carrying out driver decision prediction processing and collective route optimization processing according to the fourth information to obtain fifth information, wherein the fifth information is a vehicle running route scheme which comprises a running route of each driver;
And the fourth processing unit is used for carrying out path optimization processing on all the running routes in the fifth information, and carrying out comprehensive simulation processing on the optimized routes to obtain sixth information, wherein the sixth information is a final vehicle running route scheme.
The beneficial effects of the invention are as follows:
According to the invention, by comprehensively considering the physical characteristics of the road, the environmental factors and the behavior mode of the driver, the route planning is optimized, the traffic jam and the vehicle emission are effectively reduced, the efficiency of the traffic system is improved, the environmental pollution is reduced, and the sustainable development target is supported. According to the invention, by considering the influence of the physical characteristics of the road and the environmental conditions on the traffic flow, the potential dangerous road sections can be avoided, and the driving safety is increased.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be apparent from the description, or may be learned by practice of the embodiments of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims thereof as well as the appended drawings.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a comprehensive planning method for a vehicle running route according to an embodiment of the invention;
fig. 2 is a schematic structural diagram of an integrated planning apparatus for a vehicle running route according to an embodiment of the present invention.
The marks in the figure: 701. a first acquisition unit; 702. a first processing unit; 703. a second processing unit; 704. a third processing unit; 705. a fourth processing unit; 7021. a first analysis subunit; 7022. a first clustering subunit; 7023. a first mapping subunit; 7024. a second analysis subunit; 7025. a first building subunit; 7031. a third analysis subunit; 7032. a first analog subunit; 7033. a first processing subunit; 7034. a first optimization subunit; 7041. a second optimization subunit; 7042. a second processing subunit; 7043. a third optimization subunit; 7044. a fourth optimization subunit; 7051. a fifth optimization subunit; 7052. a sixth optimization subunit; 7053. a second analog subunit; 7054. a first adjustment subunit; 70511. a first modeling subunit; 70512. a seventh optimization subunit; 70513. a fourth analysis subunit; 70514. and an eighth optimization subunit.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. The components of the embodiments of the present invention generally described and illustrated in the figures herein can be arranged and designed in a wide variety of configurations each. Thus, the following detailed description of the embodiments of the invention, as presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures. Meanwhile, in the description of the present invention, the terms "first", "second", and the like are used only to distinguish the description, and are not to be construed as indicating or implying relative importance.
Example 1
As shown in fig. 1, the present embodiment provides a comprehensive planning method for a vehicle running route, which includes step S1, step S2, step S3, step S4 and step S5.
Step S1, acquiring first information and second information, wherein the first information comprises behavior data of at least two drivers in an area to be planned, the second information comprises real-time traffic network state data, environment data and road network structure data, and the behavior data comprises departure points, destinations, travel time and historical route selection of each driver.
It will be appreciated that driver behavior data includes departure points, destinations, travel times and historical route selections, which are key to understanding the individual needs and travel habits of the driver. Specifically, from the departure point and destination information, a common route of the driver can be revealed; travel time information helps to identify travel preferences of drivers during peak or off-peak hours; historical route selection provides insight into driver preference routes. In addition, real-time traffic network status data, such as road conditions, traffic flow, accident or road construction information, and environmental data (weather conditions, visibility, etc.) and road network structure data (road layout, traffic light positions, time sequences, etc.), enables the system to accurately understand current and future traffic conditions and consider the influence of environmental factors on traffic flow. Integrating the data is not only beneficial to creating personalized route planning and improving the compliance rate of drivers to recommended routes, but also can optimize the overall traffic flow distribution, reduce congestion, support green travel and enhance road safety. By the comprehensive data driving method, the accuracy and the adaptability of route planning can be effectively improved, more flexible and reliable route selection can be provided particularly when complex traffic environments and emergencies are handled, and the integral sustainable development of the urban traffic system is promoted.
And S2, performing behavior pattern learning processing according to the first information to obtain third information, wherein the third information is a behavior pattern model for predicting the behavior preference of the driver under different conditions.
It will be appreciated that in this step, the system first analyzes the data of the departure point, destination, travel time, and historical route selection of the driver, reflecting the individual travel habits and preferences of the driver. Further, by applying data analysis and machine learning techniques, such as cluster analysis, pattern recognition and predictive modeling, the system is able to recognize behavior patterns of different drivers and build models to predict the driver's behavioral trends under certain conditions.
And step S3, carrying out network construction processing according to the second information and the third information, and obtaining fourth information through key node importance evaluation and traffic influence analysis, wherein the fourth information is a network flow analysis result containing road network efficiency and robustness.
It can be understood that the second information is utilized to construct a detailed road network model, and the model not only reflects real-time traffic flow and road conditions, but also integrates environmental factors and structural characteristics of the road network, such as traffic signal lamp layout, road connectivity and the like. In combination with the behavior pattern model in the third information, the system can simulate behavior trends of different drivers in the network model, so as to evaluate the importance of each node (such as an intersection and a main road section) and the influence of each node on the whole traffic flow. Through this analysis, the system is able to identify key traffic bottleneck points and high efficiency road segments, providing valuable insight into traffic management and route planning. Such network flow analysis results not only take into account the efficiency of the road network, but also focus on its robustness in the face of emergencies (e.g., traffic accidents, bad weather), i.e., the ability of the network to remain functional and efficient in the face of challenges.
And S4, carrying out driver decision prediction processing and collective route optimization processing according to the fourth information to obtain fifth information, wherein the fifth information is a vehicle running route scheme, and the vehicle running route scheme comprises a running route of each driver.
It will be appreciated that this step uses the results of the network flow analysis to predict driver decisions under specific road network conditions. This involves analyzing the behavior trends of different drivers in the face of traffic flow changes, road network bottlenecks, etc., which are based on previous acquisitions from behavior pattern models. The predictive process takes into account the individual needs and preferences of the driver and the complexity of the real-time road network conditions. These predictions are then integrated by a collective route optimization process to formulate an overall route plan that is both efficient and sustainable. The optimization process involves balancing individual driver demands with overall road network efficiency in an attempt to reduce congestion, reduce emissions, while ensuring that each driver's journey is as efficient as possible.
And S5, carrying out path optimization processing on all the running routes in the fifth information, and carrying out comprehensive simulation processing on the optimized routes to obtain sixth information, wherein the sixth information is a final vehicle running route scheme.
It will be appreciated that this step is a path optimization processing stage that includes consideration of factors such as the actual feasibility of the route, anticipated traffic conditions, possible delays, and environmental impact. Preferably, a variety of optimization algorithms, such as genetic or simulated annealing algorithms, may be employed in the process to find optimal routes that are balanced in terms of efficiency, safety, and environmental friendliness. Further, the comprehensive simulation process involves using advanced simulation techniques, such as traffic flow simulation and multi-agent system simulation, to test the performance of the optimized route in an actual traffic environment. The goal of this stage is to ensure the effectiveness and robustness of the proposed route in the face of real world complexity, while assessing its impact on the overall traffic flow. The method not only ensures the practicability and effectiveness of route planning, but also enhances the adaptability and reliability through simulation verification, and provides powerful support for realizing efficient and environment-friendly traffic operation.
In one embodiment of the present disclosure, step S2 includes step S21, step S22, step S23, step S24, and step S25.
And S21, performing principal component analysis and weight distribution processing according to the first information, and obtaining a feature vector set by integrating travel basic data of a driver and sensitivity data of specific route features.
It will be appreciated that principal component analysis helps to refine the most critical information by identifying the principal variables and patterns in the data, thereby reducing the complexity of the data set. In the context of vehicle travel route planning, the most important features, such as common routes, travel time preferences, etc., can be extracted from a large amount of driver behavior data. Next, the system will weight the principal features extracted from the principal component analysis. The purpose of weight allocation is to emphasize the importance of certain features in route planning, such as sensitivity to congested routes or preference for specific road segments. Doing so may help the model better understand each driver's unique behavior patterns and preferences. Finally, by integrating the travel base data of the driver (such as departure point, destination, common route) and the sensitivity data of specific route characteristics (such as preference of peak time or specific type of road), the system forms a characteristic vector set. This set represents the comprehensive features of the driver's behavior, providing the basis for subsequent cluster analysis and behavior pattern recognition.
And S22, processing by using a density-based clustering algorithm according to the feature vector set, and obtaining a group division result by combining discussion data of drivers on traffic and routes on social media.
It will be appreciated that the feature vector set contains travel base data for the driver (e.g., start point, destination) and sensitivity data for specific route features (e.g., preferences for specific road types or traffic conditions). These feature vector sets are used as input data in a density-based clustering algorithm to identify patterns of behavior in a population of drivers. In the context of vehicle travel route planning, the setting of the neighborhood radius (ε) and minimum points (MinPts) is determined based on the nature of the driver behavior data. Specifically, if the feature vector set shows that the travel density of the driver in a certain specific area is high, epsilon can be set to a small value to identify a fine pattern in this dense area. The setting of MinPts is then dependent on how densely clusters are desired to be identified, and larger MinPts values can help identify more pronounced patterns of behavior. The density-based clustering algorithm traverses each feature vector (representing a driver's behavior pattern) and checks whether the number of other vectors in its neighborhood meets the requirements of MinPts. The vector meeting the conditions is considered the core point and represents one particular type of driver behavior pattern. The core points and their density of reachable points form a cluster representing a group of drivers with similar behavior patterns. While those points within the vicinity of the core point (edge points) that do not meet the conditions may represent drivers whose behavior patterns are more complex or between patterns. In this embodiment, fully isolated points (noise) represent drivers with extreme or unusual behavior patterns, not considered. Next, in conjunction with the driver's discussion data on traffic and routes on social media, this requires a prior textual analysis of the social media content to extract keywords and emotional trends related to traffic, routes. These data are processed by natural language processing techniques (preferably such as emotion analysis and keyword extraction) and then combined with the clustering results of density-based clustering algorithms to further refine and adjust the clustering groupings. For example, if a group of drivers frequently discuss congestion of a route on social media and exhibit negative emotions, this may indicate that they are highly sensitive to the route, tending to find alternative routes. Finally, the group classification result obtained in this step classifies drivers into different categories according to their behavioral characteristics and social media feedback. Such grouping helps the system more accurately understand the behavior patterns and preferences of different drivers, providing more personalized route planning suggestions in subsequent steps. The calculation formula involved is as follows:
wherein/> A data point representing behavioral data of a particular driver; /(I)A behavior data point representing another driver in the data set; /(I)Representing a radius of the field; /(I)Representing a dot/>Is a set of points within the domain of (a).
And S23, performing supervised learning processing according to the group division result, and constructing a discrimination model capable of distinguishing the route selection preference of a driver in the peak and off-peak time periods through kernel function mapping processing by combining the route selection frequency characteristics.
It will be appreciated that this step involves analysing the driver's route selection frequency characteristics, in particular preferences for different time periods. The supervised learning algorithm employed, preferably a Support Vector Machine (SVM), uses known travel patterns as training data, mapping these data by kernel functions to a higher dimensional space where travel preferences for different time periods are more easily distinguishable. Such processing not only improves the classification capabilities of the model, but also enables the model to more accurately identify and predict driver behavior differences during rush hour and off-rush hour. The finally obtained discrimination model provides deep insight for the system, is beneficial to making more accurate and personalized route planning for drivers, thereby optimizing traffic flow distribution, relieving traffic pressure in peak time and improving efficiency and fluency of the whole traffic system.
And S24, carrying out time sequence analysis processing according to the time sequence data of the discrimination model and the historical route, and obtaining a trend model for predicting the behavior of the driver in a specific future time period by adopting a mixed effect model to adapt to individual differences of different drivers.
It will be appreciated that this step utilizes a mixed effect model that uniquely combines fixed effects (such as time periodicity, weather, etc. prevalent factors) and random effects (specific behavioral characteristics for individual drivers) to accommodate individual differences among individual drivers. By analyzing trends and patterns in the historical data, such as the driver's travel habits over a particular period of time or preferences for a particular route, the system is able to construct a trend model. The model captures the regularity and periodicity of behaviors and also considers individual differences, so that the prediction is more accurate and personalized. Such time series analysis method enables the model to predict possible behavior trends of the driver in a specific time period in the future, and provides powerful data support for real-time and dynamic route planning. The adaptability and the accuracy of route planning are improved, personalized route recommendation is optimized, and the travel experience of a driver is enhanced.
And S25, according to the trend model and the real-time road condition data, a Bayesian network processing is applied to construct a behavior mode model.
It will be appreciated that this process involves first fusing predictions of historical behavioral trends with current traffic information, such as traffic flow and incidents, to accommodate real-time changing traffic conditions. The bayesian network is used as a probabilistic graph model to effectively cope with uncertainty, and in this scenario, it is used to simulate the influence of different factors (such as time, weather, road conditions) on the behavior of the driver and predict behavior patterns under different conditions. The behavior model constructed by the method not only can reflect the behavior trend of a driver under specific conditions, but also can dynamically adjust prediction according to real-time data, so that the flexibility and adaptability of the model are improved. This enables the model to provide more accurate, more personalized guidance for route planning, particularly when dealing with rapidly changing traffic environments and emergency situations. The step S25 includes steps S251, S252, and S253.
Step S251, performing data integration processing according to the trend model and the real-time road condition data to obtain a comprehensive data set, and performing Bayesian network structure construction processing according to the comprehensive data set to obtain a network structure reflecting interaction of different factors, wherein nodes of the network represent different influencing factors (such as time, weather and current road condition), and edges represent probability relations among the factors.
It can be appreciated that in this step, the combination of the historical behavior analysis provided by the trend model and the real-time road condition data creates a dynamic and comprehensive data environment. The fusion not only considers the long-term behavior mode of the driver, but also considers the instant influence of the current road network condition. Such data fusion provides a more comprehensive view of predictions, particularly in rapidly changing traffic environments, which can provide more accurate behavioral predictions. The construction of bayesian networks is optimized specifically for the complexity of traffic and driver behavior. Nodes in the network not only comprise standard factors such as time and weather, but also consider factors which have great influence on traffic fluidity, such as specific events (accidents and large activities) or specific conditions of road sections (such as construction and temporary road sealing). Such customized networks can more accurately reflect changes in driver behavior under certain circumstances.
And step 252, according to the constructed network structure, obtaining the behavior mode prediction results under different conditions through the probabilistic reasoning process.
It can be understood that this step combines multivariate data in complex traffic networks, such as time periods, weather conditions and real-time road conditions, and uses bayesian networks for probabilistic reasoning to predict the driver's behavior patterns under different conditions. The method can dynamically adapt to real-time traffic changes, and particularly, as new data is input, such as the latest traffic flow or accident report, the Bayesian network correspondingly adjusts the probability relation among nodes of the Bayesian network, and the behavior prediction is updated in real time. In addition, the reasoning process not only considers the general influencing factors, but also can conduct personalized prediction according to the historical behavior data of individual drivers, and improves the accuracy of route planning. For example, for commuter drivers and occasionally driving drivers, the model can predict different behavior patterns based on their historical behavior and current situation. Therefore, the method not only improves the adaptability and accuracy of the traffic prediction model, but also provides more personalized and real-time route advice for the driver, and effectively improves the travel experience of the driver.
Step 253, performing adaptive adjustment processing according to the behavior mode prediction result to obtain a final behavior mode model.
It can be appreciated that in this step, not only is the adjustment performed according to the preliminary behavior pattern prediction result generated by the bayesian network, but also real-time traffic data and constantly changing environmental factors are taken into consideration. The dynamic adjustment mechanism enables the model to reflect the latest changes of traffic conditions, such as the updating of emergencies or traffic policies, in real time, so that the timeliness and the accuracy of prediction are ensured.
In one embodiment of the present disclosure, step S3 includes step S31, step S32, step S33, and step S34.
And S31, analyzing and processing the shortest path of the road network according to the real-time traffic network state data in the second information and a preset path optimization mathematical model to obtain an optimal path set among the nodes.
It will be appreciated that this step first collects and analyzes real-time traffic data, such as current traffic flow, road surface conditions, accident or road construction information, etc., which provide critical information about current road network conditions. Path optimization is then performed using a shortest path algorithm such as the dijkstra algorithm or the a-algorithm. These algorithms are adjusted in this embodiment to accommodate real-time data, for example, they may take into account traffic congestion situations, thereby providing the driver with the best route during peak hours.
And S32, performing simulation processing by using a preset flow distribution model according to the obtained optimal path set and the environment data, and performing flow distribution and congestion prediction processing by integrating the influence of weather conditions on traffic flow to obtain a traffic flow distribution map.
It is appreciated that the traffic distribution model can analyze the predicted traffic flow direction based on the optimal path set. This model not only takes into account the physical characteristics and capacity limitations of the road, but also incorporates the driver's routing behavior. Thus, this simulation process reflects not only the characteristics of the road itself, but also the driver's preferences and behavior patterns. In the simulation process, the system also integrates environmental factors, especially the influence of weather conditions on traffic flow. For example, bad weather may cause traffic flow to decrease in certain road segments or increase traffic flow in other routes. This consideration makes flow distribution and congestion prediction more accurate, and can reflect variability of actual traffic conditions. In combination with the analysis of the optimal path and environmental factors, the system is able to generate an exhaustive traffic flow profile. The distribution map not only displays the expected traffic volume of each road section, but also predicts the potential congestion point, and provides important decision support for traffic management and planning.
And step S33, carrying out node influence strength treatment by combining the geometric structural features of the road network and the dynamic features of the traffic flow according to the traffic flow distribution diagram and the road network structural data to obtain influence scores of all nodes on the traffic flow.
It can be appreciated that this step provides insight into the role of different nodes in vehicle routing by analyzing the geometric characteristics of road networks (e.g., intersections, road length, and width) and the dynamic changes in traffic flow (e.g., flow distribution over a particular period of time). Based on these analyses, the system assigns an impact score to each node (e.g., intersection, main road segment) in the road network. This score reflects the importance of the node in affecting the routing of the vehicle, e.g., an intersection connecting multiple major roads may have a higher score. By identifying these key nodes and their impact, a more accurate planning of the vehicle's travel route is facilitated. For example, for nodes with high impact scores, the system may prioritize or avoid to optimize the vehicle's run time and reduce congestion. Further, this step also enables the route planning to be adapted to different traffic conditions. For example, during peak hours or severe weather conditions, nodes with high impact may become points of congestion, from which the system may adjust or provide alternative routes.
And step S34, carrying out road network efficiency and robustness balance by using a multi-objective optimization algorithm according to the influence score and the behavior mode model to obtain a network flow analysis result.
It will be appreciated that this step uses a multi-objective optimization algorithm to balance the two generally conflicting objectives of road network efficiency and robustness. The algorithm considers the influence of each node on the whole running of the road network and the influence of different driver behavior modes on route selection, and aims to find an optimized route planning scheme so as to improve the traffic efficiency of the whole road network and reduce congestion. By combining the influence scoring of the nodes and the behavior pattern model, traffic flow in the road network can be predicted and managed more accurately. This means that route planning takes into account not only the influence of the physical structure of the road network, but also the driver's behavioral trends, such as routing preferences under certain conditions. This step allows for simultaneous consideration of the speed (efficiency) of reaching the destination and the ability to cope with the emergency (robustness) when route planning is performed. For example, during peak hours or when an emergency is encountered, the algorithm can provide both quick and reliable routing, reducing possible delays.
In one embodiment of the present disclosure, step S4 includes step S41, step S42, and step S43.
And S41, performing driver interaction strategy optimization processing according to fourth information and a preset Stankleber model to obtain strategy balance points of each driver in the current road network state.
It is appreciated that the Stankleberger model is used in economics to analyze strategic interactions between leaders and followers. In this embodiment, the model is applied to the traffic domain to simulate and optimize the interaction strategy between drivers. In this model, the routing of some drivers (leaders) may affect the selection of other drivers (followers), reflecting dynamic interactions in real traffic. By applying the stoneley model, the system is able to analyze and predict how different driver groups adjust their travel routes to achieve strategic equality under a given road network situation. This equalization is a strategic state where each driver's routing takes into account the behavior of the other drivers and the overall road network conditions. Through the optimization treatment of the step, a driver can be guided to make more effective route selection, congestion is reduced, and the traffic efficiency of the whole road network is improved. This optimization takes into account the individual driver's route preferences and the overall road network traffic distribution, with the aim of balancing personal and overall benefits.
And step S42, carrying out route coordination and conflict resolution processing according to the strategy balance points to obtain a cooperative route decision frame for reducing the overall travel time and avoiding the traffic bottleneck.
It will be appreciated that the collaborative route decision framework aims to coordinate the routing of different drivers to maximize overall road network efficiency. This collaborative planning takes into account the interplay between drivers and the flow distribution throughout the road network to optimize the routing for each driver. Through this coordination, the system can effectively reduce overall travel time and avoid traffic bottlenecks. For example, if an area is expected to be congested, the collaborative route decision framework may direct some drivers to select alternative routes, thereby relieving traffic pressure for a particular road segment. This step also includes a conflict resolution mechanism aimed at resolving conflicts that may arise due to the routing of different drivers. By analyzing the route preferences and current road network conditions of multiple drivers, the system is able to identify and resolve these potential conflicts, ensuring the effectiveness and rationality of route selection. The generated collaborative route decision framework not only provides optimized route selection for a single driver, but also considers the efficiency of the whole traffic flow, so that the whole road network can run more smoothly and efficiently.
And S43, optimizing the collective route by adopting a genetic algorithm according to the collaborative route decision frame and the real-time road condition data, evaluating the route efficiency by adopting a fitness function, and obtaining an optimized route scheme on the group level through iterative evolutionary processing.
It will be appreciated that genetic algorithm is an optimization technique that mimics natural selection and genetic principles, and is used in this step to find the optimal solution from among numerous possible routes. To evaluate the efficiency of the different routing schemes, an fitness function is defined. This function takes into account a number of factors such as travel time, distance and possible congestion levels of the route, thereby ensuring that the selected route is optimized. Genetic algorithms continually iterate and optimize routing schemes by simulating the process of natural evolution (e.g., selection, crossover, mutation). In each iteration, the algorithm evaluates the route schemes in the current population and selects the scheme with the highest fitness for the generation of the next generation. Further, this step focuses more on route optimization at the group level, meaning that route selection for the entire driver group is considered, not just a single driver. This approach helps to achieve more efficient traffic distribution and congestion management throughout the road network. The fitness function in this step is defined as follows:
wherein, Representing a fitness function; /(I)、/>、/>Weight coefficients respectively representing time efficiency, distance efficiency and congestion avoidance; /(I)Representing a time efficiency function; /(I)Representing a distance efficiency function; /(I)Representing a congestion avoidance function; /(I)Is an example of a routing scheme; /(I)Representation route/>Is set, the estimated travel time of (a); /(I)Representing a theoretical shortest travel time; /(I)A parameter indicative of a sensitivity of the adjusted time efficiency; /(I)Representation route/>Is set in the vehicle, the actual travel distance of (a) is set; /(I)Representing the shortest possible travel distance; /(I)A parameter indicative of sensitivity to adjustment of the distance efficiency; /(I)Representation route/>Average congestion level on; indicating the highest congestion level.
And S44, performing global search optimization on the combination by adopting a simulated annealing algorithm according to the optimized route scheme, and obtaining the vehicle running route scheme under different traffic flow changing conditions through energy state evaluation and probability acceptance criterion processing.
It is understood that in a real traffic environment, traffic flow conditions may change continuously. This step employs a simulated annealing algorithm to address this situation to ensure that the vehicle's path of travel is still effective under different traffic flow changing conditions. The simulated annealing algorithm is a global search and optimization algorithm that finds the optimal solution by simulating the molecular arrangement during solid annealing. Specifically, this step searches for possible route combinations through multiple iterations and random perturbations to find the best vehicle travel route. In this process, the key concepts are "energy state assessment" and "probability acceptance criteria". The energy state assessment is used to measure the quality of each route combination, which takes into account a number of factors, such as total travel time, range and traffic congestion conditions of the route. Our goal is to minimize these energy states to find the best route. The probability acceptance criteria is the core of the simulated annealing algorithm, which decides whether to accept a new combination of routes. When the new route combination has a lower energy state (better quality), it is almost always accepted. However, when the new route combination quality is poor, there is an opportunity to accept it with probability P. This is to prevent trapping in local minima, making the algorithm have the ability to jump out of the local optimal solution, looking for a global optimal solution. In summary, the present step uses a simulated annealing algorithm to perform global search optimization on the vehicle operating route under different traffic flow conditions, so as to ensure that the optimal route scheme can operate efficiently under various traffic conditions.
In one embodiment of the present disclosure, step S5 includes step S51, step S52, step S53, and step S54.
And S51, carrying out path optimization processing by adopting a maximum flow minimum cutting theorem in a network flow theory according to all running routes in the fifth information, and obtaining a primary route scheme for maximizing traffic fluidity.
It will be appreciated that this process uses the basic principles of network flow theory to ensure efficient flow of vehicles in a road network by finding the best way for the path allocation of the vehicle flow. The maximum flow minimum theorem states that the maximum flow amount is equal to the minimum flow-blocking path, and thus the goal of this step is to maximize the mobility of the vehicle while reducing the path that impedes the flow of the vehicle. In this way, a preliminary route solution can be obtained, which has been optimized by mathematical principles to increase the traffic efficiency of the road network, thus reducing the occurrence of congestion and traffic disturbances. This preliminary routing scheme will provide a basis for subsequent steps to further tune and optimize to meet different traffic demands and conditions.
And step S52, performing multi-objective optimization processing according to the primary route scheme, and obtaining an alternative route scheme by combining three traffic optimization indexes of operation time, safety and fuel efficiency.
It will be appreciated that the run time concerns the time required for a vehicle to travel on a road. Safety concerns the safety and risk of routes. Fuel efficiency is an indicator that takes into account environmental and cost factors. By considering these three metrics in combination, a set of alternative route schemes can be obtained that have balanced performance in several respects. At the same time, these alternative routes will provide more options for vehicle operation and allow appropriate decisions to be made based on specific traffic demands and conditions.
And step S53, carrying out traffic flow and driver behavior simulation processing according to the alternative route scheme, and obtaining prediction data by simulating the interaction condition of individual drivers in the whole traffic flow, wherein the prediction data comprise predicted traffic conditions and route performance data.
It will be appreciated that this step simulates the effect of an individual driver on the overall traffic flow when selecting different alternative routes. This includes consideration of their departure time, destination and route selection, as well as their speed of travel on the road and interaction with each other. Through this simulation process we can predict traffic conditions under different alternative routes, including data on congestion level, vehicle density, etc. Second, there is a need to simulate the driver's performance data on a selected route, including their speed of travel, attitude, possible parking situations, etc. These data may help us better understand the behavior of the driver under different routes, as well as the actual performance of the route. Through this simulation process, predictive data on alternative routes can be obtained, which data has important reference values for decision-making and route planning.
And S54, carrying out dynamic system adjustment processing on the predicted data based on a preset reinforcement learning mathematical model, and obtaining a final vehicle running route scheme of self-adaptive environment change through real-time feedback and adjustment.
It will be appreciated that the reinforcement learning mathematical model is designed to learn and optimize the vehicle travel path. The model takes into account a number of factors including traffic conditions, route efficiency, safety, etc. to make an optimal decision strategy. The model is fed back and adjusted in real time, and the running route scheme of the vehicle is continuously updated according to the current traffic condition and environmental change. This means that the system can dynamically make decisions during actual travel to accommodate changes in traffic conditions, including traffic congestion, accidents, and other incidents. By means of the reinforcement learning model, self-adaptive route planning can be achieved, so that vehicles can run in an optimal mode under different environmental conditions, and efficiency and safety of routes are improved. The method can better cope with complex traffic conditions, and provides a more intelligent and flexible vehicle running route scheme.
In one embodiment of the present disclosure, step S51 includes step S511, step S512, step S513, and step S514.
And step S511, obtaining a physical characteristic network diagram through road network graph modeling processing according to all the running routes, the geographic information system data and the road physical attributes in the fifth information.
It will be appreciated that the main purpose of this step is to model the physical characteristics of the actual road graphically. First, various data about the road including information on the geographical position, length, width, road condition, etc. of the road are collected. A road network graph is then created using the data, wherein the roads are represented as nodes in the graph and road segments connecting the roads are represented as edges in the graph. By taking the physical characteristics of the roads into consideration, the driving conditions of the vehicle on different roads can be more accurately simulated, and the running route of the vehicle is further optimized, so that the efficiency and the safety of the route are improved. This also helps to better understand the structure and characteristics of the road network and thus better plan the travel route of the vehicle.
And step S512, according to the physical characteristic network diagram, combining the influence of the gradient and the curvature of each road on the movement and the flow of the vehicle, and obtaining network optimization data through network flow optimization processing, wherein the network optimization data comprises the maximum flow and the key bottleneck path of each intersection and the road.
It will be appreciated that factors such as grade and curvature may affect the speed and fuel efficiency of the vehicle and it is therefore important to consider them in route planning. The step uses a network flow optimization method in combination with a physical characteristic network map to determine the maximum flow of each intersection and road, i.e. the upper limit of the number of vehicles they can accommodate. This helps to avoid traffic jams and road overload. In addition, critical bottleneck paths are identified, which are bottlenecks in traffic flow, resulting in route inefficiency. By knowing these critical bottleneck paths, measures can be taken to improve traffic flow to ensure a smoother and more efficient vehicle route.
Step S513, analyzing and processing the congested road sections by using a machine learning algorithm according to the network optimization data, and obtaining an identification result by combining the influence of the road sections with large gradient and sharp curvature on the traffic flow and bottleneck identification processing, wherein the identification result is a key road section which is easy to generate congestion.
It will be appreciated that this step utilizes network optimization data, including maximum traffic per intersection and road and critical bottleneck paths as input data. Machine learning algorithms are then used to analyze the data, particularly for those road segments with steep gradients and sharp curvatures. These road segments are more prone to traffic congestion under certain conditions, e.g. steep hills may lead to slowing of the vehicle and sharp curvatures may increase driving difficulties. And obtaining an identification result through analysis processing, namely a key road section which is easy to generate congestion. These road segments are prone to traffic congestion in certain situations and therefore require special attention in route planning and traffic management. By identifying these critical road segments, preventative measures may be taken to reduce the occurrence of traffic congestion and improve the efficiency and mobility of the route. This helps to improve the travel experience of the driver and reduces the inconvenience of traffic congestion.
And step S514, according to the identification result and the adaptability of different vehicles to the gradient and the curvature, adopting a multi-objective optimization algorithm to conduct personalized path planning, and obtaining a preliminary path scheme through the path optimization processing assisted by the maximum flow minimum cutting theorem.
It will be appreciated that this step allows for the identification of congested road segments, but not just stops at the sign of congestion, but rather analyzes characteristics of the congested road segments, including grade and curvature. This is because these factors may have a significant impact on congestion during vehicle travel, particularly in mountainous areas or on curved roads. By taking these specific factors into account, we can more accurately predict possible traffic congestion, thereby providing a more accurate basis for route planning. Secondly, we introduce consideration of the adaptability of different vehicle types to grade and curvature. This is to better meet the needs of different types of vehicles. For example, a truck may be more suitable for handling smaller grades and curvatures, while an off-road vehicle may be more suitable for driving on steep hills. Such personalized considerations make the generated route more compatible with the characteristics of the vehicle, thereby improving the safety and efficiency of travel. Finally, a multi-objective optimization algorithm is employed to balance a number of factors of the route, including travel time, safety, and fuel efficiency. The comprehensive consideration not only helps to improve the overall performance of the vehicle, but also can be better adapted to different driving conditions.
Example 2
As shown in fig. 2, the present embodiment provides an integrated planning apparatus for a vehicle running route, which includes a first acquisition unit 701, a first processing unit 702, a second processing unit 703, a third processing unit 704, and a fourth processing unit 705.
The first obtaining unit 701 is configured to obtain first information and second information, where the first information includes behavior data of at least two drivers in an area to be planned, the second information includes real-time traffic network status data, environment data, and road network structure data, and the behavior data includes a departure point, a destination, a trip time, and a historical route selection of each driver.
The first processing unit 702 is configured to perform behavior pattern learning processing according to the first information to obtain third information, where the third information is a behavior pattern model for predicting behavior preferences of the driver under different conditions.
The second processing unit 703 is configured to perform network construction processing according to the second information and the third information, and obtain fourth information through key node importance assessment and traffic impact analysis, where the fourth information is a network flow analysis result including road network efficiency and robustness.
And a third processing unit 704, configured to perform a driver decision prediction process and a collective route optimization process according to the fourth information to obtain fifth information, where the fifth information is a vehicle running route scheme, and the vehicle running route scheme includes a running route of each driver.
The fourth processing unit 705 is configured to perform path optimization processing on all the running routes in the fifth information, and perform comprehensive simulation processing on the optimized routes to obtain sixth information, where the sixth information is a final vehicle running route scheme.
In one embodiment of the present disclosure, the first processing unit 702 includes:
The first analysis subunit 7021 is configured to perform principal component analysis and weight distribution processing according to the first information, and obtain a feature vector set by integrating travel basic data of a driver and sensitivity data to a feature of a specific route.
The first clustering subunit 7022 is configured to obtain a group classification result by combining the discussion data of the driver on the social media about traffic and routes according to the feature vector set by using a density-based clustering algorithm.
The first mapping subunit 7023 is configured to perform supervised learning processing according to the group division result, and construct a discrimination model capable of distinguishing the route selection preference of the driver in the peak period and the off-peak period through kernel function mapping processing by combining the route selection frequency characteristics.
The second analysis subunit 7024 is configured to perform time-series analysis processing according to the discrimination model and the time-series data of the historical route, and obtain a trend model for predicting the behavior of the driver in a specific time period in the future by adapting to individual differences of different drivers by using the mixed effect model.
The first construction subunit 7025 is configured to apply bayesian network processing to construct a behavior pattern model according to the trend model and the real-time road condition data.
In one embodiment of the present disclosure, the second processing unit 703 includes:
And a third analysis subunit 7031, configured to perform a road network shortest path analysis process according to the real-time traffic network status data in the second information and the preset path optimization mathematical model, so as to obtain an optimal path set between the nodes.
The first simulation subunit 7032 is configured to perform simulation processing using a preset traffic distribution model according to the obtained optimal path set and the environmental data, and perform traffic distribution and congestion prediction processing by integrating the influence of the weather condition on the traffic flow, so as to obtain a traffic flow distribution map.
The first processing subunit 7033 is configured to obtain an impact score of each node on the traffic flow by performing node impact strength processing according to the traffic flow distribution map and the road network structure data and by combining the geometric structure feature of the road network and the dynamic feature of the traffic flow.
The first optimizing subunit 7034 is configured to perform road network efficiency and robustness tradeoff by using a multi-objective optimizing algorithm according to the influence score and the behavior pattern model, so as to obtain a network flow analysis result.
In one embodiment of the present disclosure, the third processing unit 704 includes:
the second optimizing subunit 7041 is configured to perform optimization processing on the driver interaction strategy according to the fourth information and a preset stonelberg model, so as to obtain a strategy balance point of each driver in the current road network state.
The second processing subunit 7042 is configured to perform route coordination and conflict resolution processing according to the policy balancing point, so as to obtain a collaborative route decision framework for reducing overall travel time and avoiding traffic bottlenecks.
And a third optimizing subunit 7043, configured to optimize the collective route by using a genetic algorithm according to the collaborative route decision frame and the real-time road condition data, evaluate the route efficiency by using the fitness function, and obtain an optimized route scheme on the group level by using iterative evolutionary processing.
And a fourth optimizing subunit 7044, configured to perform global search optimization on the combination by using a simulated annealing algorithm according to the optimized route scheme, and obtain the vehicle running route scheme under different traffic flow variation conditions through energy state evaluation and probability acceptance criterion processing.
In one embodiment of the present disclosure, the fourth processing unit 705 includes:
And a fifth optimizing subunit 7051, configured to perform path optimization processing according to all the running routes in the fifth information by using the maximum flow minimum cutting theorem in the network flow theory, so as to obtain a preliminary route scheme that maximizes traffic fluidity.
A sixth optimization subunit 7052 is configured to perform multi-objective optimization processing according to the preliminary route scheme, and obtain an alternative route scheme by combining traffic optimization indexes including running time, safety and fuel efficiency.
The second simulation subunit 7053 is configured to perform a traffic flow and driver behavior simulation process according to the alternative route scheme, and obtain prediction data by simulating an interaction condition of an individual driver in the overall traffic flow, where the prediction data includes predicted traffic conditions and route performance data.
The first adjustment subunit 7054 performs dynamic system adjustment processing on the prediction data based on a preset reinforcement learning mathematical model, and obtains a final vehicle running route scheme of self-adaptive environment change through real-time feedback and adjustment.
In one embodiment of the present disclosure, the fifth optimization subunit 7051 comprises:
The first modeling subunit 70511 is configured to obtain a physical characteristic network map through road network graph modeling processing according to all the running routes, the geographic information system data and the road physical attribute in the fifth information.
The seventh optimization subunit 70512 is configured to obtain, according to the physical characteristic network diagram, network optimization data by performing network flow optimization processing according to the influence of the gradient and the curvature of each road on the movement and the flow of the vehicle, where the network optimization data includes the maximum flow and the critical bottleneck path of each intersection and each road.
And the fourth analysis subunit 70513 is configured to perform analysis processing on the congested road segment by using a machine learning algorithm according to the network optimization data, and obtain an identification result through bottleneck identification processing by combining the influence of the road segment with a large gradient and a sharp curvature on the traffic flow, where the identification result is a key road segment that is easy to generate congestion.
And an eighth optimizing subunit 70514, configured to perform personalized path planning by using a multi-objective optimizing algorithm according to the recognition result and the adaptability of different vehicles to the gradient and the curvature, and obtain a preliminary path scheme through the route optimizing process assisted by the maximum flow minimum cutting theorem.
The above is only a preferred embodiment of the present invention, and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
The foregoing is merely illustrative embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily think about variations or substitutions within the technical scope of the present invention, and the invention should be covered. Therefore, the protection scope of the invention is subject to the protection scope of the claims.
Claims (4)
1. A method for comprehensive planning of a vehicle travel route, comprising:
Acquiring first information and second information, wherein the first information comprises behavior data of at least two drivers in an area to be planned, the second information comprises real-time traffic network state data, environment data and road network structure data, and the behavior data comprises a departure point, a destination, travel time and historical route selection of each driver;
performing behavior pattern learning processing on the first information to obtain third information, wherein the third information is a behavior pattern model for predicting the behavior preference of the driver under different conditions;
Performing network construction processing according to the second information and the third information, and obtaining fourth information through key node importance evaluation and traffic influence analysis, wherein the fourth information is a network flow analysis result containing road network efficiency and robustness;
carrying out driver decision prediction processing and collective route optimization processing according to the fourth information to obtain fifth information, wherein the fifth information is a vehicle running route scheme which comprises a running route of each driver;
Path optimization processing is carried out on all running routes in the fifth information, comprehensive simulation processing is carried out on the optimized routes to obtain sixth information, and the sixth information is a final vehicle running route scheme;
The behavior pattern learning processing is performed on the first information to obtain third information, which comprises the following steps:
performing principal component analysis and weight distribution processing according to the first information, and obtaining a feature vector set by integrating travel basic data of a driver and sensitivity data of specific route features;
Processing according to the feature vector set by using a density-based clustering algorithm, and obtaining a group division result by combining discussion data of drivers on traffic and routes on social media;
Performing supervised learning processing according to the group division result, and constructing a discrimination model capable of distinguishing the route selection preference of a driver in peak and off-peak periods through kernel function mapping processing by combining route selection frequency characteristics;
Performing time sequence analysis processing according to the time sequence data of the discrimination model and the historical route, and obtaining a trend model for predicting the behavior of the driver in a specific future time period by adopting a mixed effect model to adapt to individual differences of different drivers;
according to the trend model and the real-time road condition data, a Bayesian network is applied to process and construct to obtain a behavior mode model;
the network construction processing is performed according to the second information and the third information, and fourth information is obtained through key node importance evaluation and traffic influence analysis, and the method comprises the following steps:
carrying out path network shortest path analysis processing according to the real-time traffic network state data in the second information and a preset path optimization mathematical model to obtain an optimal path set among all nodes;
according to the obtained optimal path set and environmental data, performing simulation processing by using a preset flow distribution model, and performing flow distribution and congestion prediction processing by integrating the influence of weather conditions on traffic flow to obtain a traffic flow distribution map;
according to the traffic flow distribution diagram and road network structure data, carrying out node influence power processing by combining the geometric structure characteristics of the road network and the dynamic characteristics of the traffic flow to obtain influence scores of all nodes on the traffic flow;
according to the influence score and the behavior mode model, carrying out road network efficiency and robustness balance by using a multi-objective optimization algorithm to obtain a network flow analysis result;
And carrying out driver decision prediction processing and collective route optimization processing according to the fourth information to obtain fifth information, wherein the method comprises the following steps of:
Performing driver interaction strategy optimization processing according to the fourth information and a preset Stankleber model to obtain strategy balance points of each driver in the current road network state;
carrying out route coordination and conflict resolution processing according to the strategy balance points to obtain a cooperative route decision frame for reducing the overall travel time and avoiding traffic bottlenecks;
Optimizing a collective route by adopting a genetic algorithm according to the collaborative route decision frame and the real-time road condition data, evaluating route efficiency by a fitness function, and obtaining an optimized route scheme on a group level by iterative evolutionary processing;
And according to the optimized route scheme, performing global search optimization on the combination by adopting a simulated annealing algorithm, and obtaining the vehicle running route scheme under different traffic flow changing conditions through energy state evaluation and probability acceptance criterion processing.
2. The comprehensive planning method of a vehicle running route according to claim 1, wherein performing path optimization processing on all running routes in the fifth information, and performing comprehensive simulation processing on the optimized routes to obtain sixth information, includes:
according to all the running routes in the fifth information, adopting a maximum flow minimum cutting theorem in a network flow theory to perform path optimization treatment, and obtaining a primary route scheme for maximizing traffic fluidity;
performing multi-objective optimization processing according to the preliminary route scheme, and obtaining an alternative route scheme by combining three traffic optimization indexes of running time, safety and fuel efficiency;
According to the alternative route scheme, traffic flow and driver behavior simulation processing is carried out, prediction data are obtained by simulating interaction conditions of individual drivers in the whole traffic flow, and the prediction data comprise predicted traffic conditions and route performance data;
and carrying out dynamic system adjustment processing on the predicted data based on a preset reinforcement learning mathematical model, and obtaining a final vehicle running route scheme of self-adaptive environment change through real-time feedback and adjustment.
3. An integrated planning apparatus for a vehicle travel route, comprising:
The system comprises a first acquisition unit, a second acquisition unit and a control unit, wherein the first acquisition unit is used for acquiring first information and second information, the first information comprises behavior data of at least two drivers in an area to be planned, the second information comprises real-time traffic network state data, environment data and road network structure data, and the behavior data comprises a departure point, a destination, travel time and historical route selection of each driver;
The first processing unit is used for performing behavior pattern learning processing on the first information to obtain third information, wherein the third information is a behavior pattern model for predicting the behavior preference of the driver under different conditions;
The second processing unit is used for carrying out network construction processing according to the second information and the third information, and obtaining fourth information through key node importance evaluation and traffic influence analysis, wherein the fourth information is a network flow analysis result comprising road network efficiency and robustness;
The third processing unit is used for carrying out driver decision prediction processing and collective route optimization processing according to the fourth information to obtain fifth information, wherein the fifth information is a vehicle running route scheme which comprises a running route of each driver;
The fourth processing unit is used for carrying out path optimization processing on all the running routes in the fifth information, and carrying out comprehensive simulation processing on the optimized routes to obtain sixth information, wherein the sixth information is a final vehicle running route scheme;
Wherein the first processing unit includes:
The first analysis subunit is used for carrying out principal component analysis and weight distribution processing according to the first information, and obtaining a feature vector set by integrating travel basic data of a driver and sensitivity data of specific route features;
the first clustering subunit is used for processing by using a density-based clustering algorithm according to the feature vector set, and obtaining a group division result by combining discussion data of drivers on traffic and routes on social media;
The first mapping subunit is used for performing supervised learning processing according to the group division result, and constructing a discrimination model capable of distinguishing the route selection preference of a driver in the peak and off-peak time period through kernel function mapping processing by combining the route selection frequency characteristics;
The second analysis subunit is used for carrying out time sequence analysis processing according to the time sequence data of the discrimination model and the historical route, and obtaining a trend model for predicting the behavior of the driver in a specific future time period by adopting a mixed effect model to adapt to individual differences of different drivers;
the first construction subunit is used for constructing a behavior pattern model by applying Bayesian network processing according to the trend model and the real-time road condition data;
Wherein the second processing unit includes:
The third analysis subunit is used for carrying out path network shortest path analysis processing according to the real-time traffic network state data in the second information and a preset path optimization mathematical model to obtain an optimal path set among all nodes;
The first simulation subunit is used for performing simulation processing by using a preset flow distribution model according to the obtained optimal path set and environment data, and performing flow distribution and congestion prediction processing by integrating the influence of weather conditions on traffic flow to obtain a traffic flow distribution map;
the first processing subunit is used for obtaining the influence score of each node on the traffic flow by carrying out node influence strengthening processing according to the traffic flow distribution map and the road network structure data and by combining the geometric structure characteristics of the road network and the dynamic characteristics of the traffic flow;
the first optimizing subunit is used for carrying out road network efficiency and robustness balance by using a multi-objective optimizing algorithm according to the influence score and the behavior mode model to obtain a network flow analysis result;
Wherein the third processing unit includes:
the second optimizing subunit is used for carrying out optimization processing on the interaction strategy of the driver according to the fourth information and a preset Stankleberg model to obtain strategy balance points of each driver under the current road network state;
The second processing subunit is used for carrying out route coordination and conflict resolution processing according to the strategy balancing point to obtain a cooperative route decision frame for reducing the whole travel time and avoiding traffic bottlenecks;
The third optimizing subunit is used for optimizing the collective route by adopting a genetic algorithm according to the collaborative route decision frame and the real-time road condition data, evaluating the route efficiency by a fitness function, and obtaining an optimized route scheme on a group level by iterative evolutionary processing;
and the fourth optimizing subunit is used for carrying out global search optimization on the combination by adopting a simulated annealing algorithm according to the optimized route scheme, and obtaining the vehicle running route scheme under different traffic flow changing conditions through energy state evaluation and probability acceptance criterion processing.
4. A comprehensive planning apparatus for a vehicle running route according to claim 3, wherein the fourth processing unit includes:
a fifth optimizing subunit, configured to perform path optimization processing by using a maximum flow minimum cutting theorem in a network flow theory according to all the running routes in the fifth information, so as to obtain a preliminary route scheme that maximizes traffic mobility;
A sixth optimizing subunit, configured to perform multi-objective optimization processing according to the preliminary route scheme, and obtain an alternative route scheme by combining traffic optimization indexes including running time, safety and fuel efficiency;
the second simulation subunit is used for carrying out traffic flow and driver behavior simulation processing according to the alternative route scheme, and obtaining prediction data by simulating the interaction condition of individual drivers in the whole traffic flow, wherein the prediction data comprises predicted traffic conditions and route performance data;
and the first adjusting subunit performs dynamic system adjustment processing on the predicted data based on a preset reinforcement learning mathematical model, and obtains a final vehicle running route scheme of self-adaptive environment change through real-time feedback and adjustment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410222721.0A CN117804490B (en) | 2024-02-28 | 2024-02-28 | Comprehensive planning method and device for vehicle running route |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410222721.0A CN117804490B (en) | 2024-02-28 | 2024-02-28 | Comprehensive planning method and device for vehicle running route |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117804490A CN117804490A (en) | 2024-04-02 |
| CN117804490B true CN117804490B (en) | 2024-05-17 |
Family
ID=90420313
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410222721.0A Active CN117804490B (en) | 2024-02-28 | 2024-02-28 | Comprehensive planning method and device for vehicle running route |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117804490B (en) |
Citations (28)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102005008266A1 (en) * | 2005-02-22 | 2006-08-24 | Daimlerchrysler Ag | Dynamic control of the motion of a road vehicle, using sensors that detect accident causing objects to the side of the vehicles |
| CN101197077A (en) * | 2006-12-07 | 2008-06-11 | 厦门雅迅网络股份有限公司 | Analyzing and statistical method for vehicle running route |
| CN101639871A (en) * | 2009-07-23 | 2010-02-03 | 上海理工大学 | Vehicle-borne dynamic traffic information induction system analog design method facing behavior research |
| CN103730006A (en) * | 2014-01-26 | 2014-04-16 | 吉林大学 | Short-time traffic flow combined forecasting method |
| CN103985267A (en) * | 2014-06-06 | 2014-08-13 | 郝明学 | System and method for synchronously displaying traffic signal lamp state of front intersection |
| CN104331743A (en) * | 2014-10-11 | 2015-02-04 | 清华大学 | Electric vehicle travel planning method based on multi-target optimization |
| CN104574970A (en) * | 2015-01-26 | 2015-04-29 | 中国人民解放军理工大学 | Method for evaluating traffic state in connection position of entrances and exits of urban underground parking lot system |
| CN105825310A (en) * | 2016-04-11 | 2016-08-03 | 湖南科技大学 | Taxi passenger-searching path recommendation method based on information entropy |
| CN106096756A (en) * | 2016-05-31 | 2016-11-09 | 武汉大学 | A kind of urban road network dynamic realtime Multiple Intersections routing resource |
| CN108597218A (en) * | 2018-06-13 | 2018-09-28 | 桂林电子科技大学 | Path planning based on vehicle monitoring and car networking technology optimizes device with car speed |
| CN109612485A (en) * | 2018-12-17 | 2019-04-12 | 许建国 | A kind of intelligent route planning system of new-energy automobile |
| CN109816978A (en) * | 2019-01-28 | 2019-05-28 | 上海海事大学 | Consider the region group system for traffic guiding and method of driver dynamic respondent behavior |
| CN110090012A (en) * | 2019-03-15 | 2019-08-06 | 上海图灵医疗科技有限公司 | A kind of human body diseases detection method and testing product based on machine learning |
| CN110174117A (en) * | 2019-05-29 | 2019-08-27 | 国网浙江电动汽车服务有限公司 | A kind of electric car charging route planning method |
| CN110763225A (en) * | 2019-11-13 | 2020-02-07 | 内蒙古工业大学 | Trolley path navigation method and system and transport vehicle system |
| CN110920552A (en) * | 2019-11-15 | 2020-03-27 | 吉林大学 | Vehicle safety system and method for preventing interlink accident after collision on highway |
| CN111063208A (en) * | 2019-12-26 | 2020-04-24 | 山东科技大学 | Lane-level traffic guidance method and system based on Internet of vehicles |
| CN111854777A (en) * | 2019-04-30 | 2020-10-30 | 长城汽车股份有限公司 | Updating method of navigation route driving time, navigation method, navigation system and vehicle |
| CN112959966A (en) * | 2021-04-26 | 2021-06-15 | 积善云科技(武汉)有限公司 | Vehicle-mounted multimedia regulation and control method, system, equipment and computer storage medium based on internet trip and user habit deep learning |
| CN113114733A (en) * | 2021-03-24 | 2021-07-13 | 重庆邮电大学 | Distributed task unloading and computing resource management method based on energy collection |
| CN113240182A (en) * | 2021-05-19 | 2021-08-10 | 广州广电运通金融电子股份有限公司 | Short-term traffic flow prediction method, storage medium and system under complex road network |
| CN113954858A (en) * | 2020-07-20 | 2022-01-21 | 华为技术有限公司 | Method for planning vehicle driving route and intelligent automobile |
| CN114559959A (en) * | 2022-03-18 | 2022-05-31 | 重庆长安汽车股份有限公司 | Automatic driving behavior decision system and method based on driving preference |
| CN116206463A (en) * | 2023-03-06 | 2023-06-02 | 吉林大学 | Public road operation vehicle dispatch system |
| CN116279512A (en) * | 2021-12-20 | 2023-06-23 | 通用汽车环球科技运作有限责任公司 | Predicting driver state using gaze behavior |
| CN116539004A (en) * | 2023-05-04 | 2023-08-04 | 中国人民解放军国防科技大学 | Communication line engineering investigation design method and system adopting unmanned aerial vehicle mapping |
| CN116804560A (en) * | 2023-08-23 | 2023-09-26 | 四川交通职业技术学院 | Unmanned automobile safety navigation method and device under controlled road section |
| CN116805449A (en) * | 2023-07-05 | 2023-09-26 | 哈尔滨工业大学 | Traffic signal fuzzy control method and system considering individual priori data |
-
2024
- 2024-02-28 CN CN202410222721.0A patent/CN117804490B/en active Active
Patent Citations (28)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102005008266A1 (en) * | 2005-02-22 | 2006-08-24 | Daimlerchrysler Ag | Dynamic control of the motion of a road vehicle, using sensors that detect accident causing objects to the side of the vehicles |
| CN101197077A (en) * | 2006-12-07 | 2008-06-11 | 厦门雅迅网络股份有限公司 | Analyzing and statistical method for vehicle running route |
| CN101639871A (en) * | 2009-07-23 | 2010-02-03 | 上海理工大学 | Vehicle-borne dynamic traffic information induction system analog design method facing behavior research |
| CN103730006A (en) * | 2014-01-26 | 2014-04-16 | 吉林大学 | Short-time traffic flow combined forecasting method |
| CN103985267A (en) * | 2014-06-06 | 2014-08-13 | 郝明学 | System and method for synchronously displaying traffic signal lamp state of front intersection |
| CN104331743A (en) * | 2014-10-11 | 2015-02-04 | 清华大学 | Electric vehicle travel planning method based on multi-target optimization |
| CN104574970A (en) * | 2015-01-26 | 2015-04-29 | 中国人民解放军理工大学 | Method for evaluating traffic state in connection position of entrances and exits of urban underground parking lot system |
| CN105825310A (en) * | 2016-04-11 | 2016-08-03 | 湖南科技大学 | Taxi passenger-searching path recommendation method based on information entropy |
| CN106096756A (en) * | 2016-05-31 | 2016-11-09 | 武汉大学 | A kind of urban road network dynamic realtime Multiple Intersections routing resource |
| CN108597218A (en) * | 2018-06-13 | 2018-09-28 | 桂林电子科技大学 | Path planning based on vehicle monitoring and car networking technology optimizes device with car speed |
| CN109612485A (en) * | 2018-12-17 | 2019-04-12 | 许建国 | A kind of intelligent route planning system of new-energy automobile |
| CN109816978A (en) * | 2019-01-28 | 2019-05-28 | 上海海事大学 | Consider the region group system for traffic guiding and method of driver dynamic respondent behavior |
| CN110090012A (en) * | 2019-03-15 | 2019-08-06 | 上海图灵医疗科技有限公司 | A kind of human body diseases detection method and testing product based on machine learning |
| CN111854777A (en) * | 2019-04-30 | 2020-10-30 | 长城汽车股份有限公司 | Updating method of navigation route driving time, navigation method, navigation system and vehicle |
| CN110174117A (en) * | 2019-05-29 | 2019-08-27 | 国网浙江电动汽车服务有限公司 | A kind of electric car charging route planning method |
| CN110763225A (en) * | 2019-11-13 | 2020-02-07 | 内蒙古工业大学 | Trolley path navigation method and system and transport vehicle system |
| CN110920552A (en) * | 2019-11-15 | 2020-03-27 | 吉林大学 | Vehicle safety system and method for preventing interlink accident after collision on highway |
| CN111063208A (en) * | 2019-12-26 | 2020-04-24 | 山东科技大学 | Lane-level traffic guidance method and system based on Internet of vehicles |
| CN113954858A (en) * | 2020-07-20 | 2022-01-21 | 华为技术有限公司 | Method for planning vehicle driving route and intelligent automobile |
| CN113114733A (en) * | 2021-03-24 | 2021-07-13 | 重庆邮电大学 | Distributed task unloading and computing resource management method based on energy collection |
| CN112959966A (en) * | 2021-04-26 | 2021-06-15 | 积善云科技(武汉)有限公司 | Vehicle-mounted multimedia regulation and control method, system, equipment and computer storage medium based on internet trip and user habit deep learning |
| CN113240182A (en) * | 2021-05-19 | 2021-08-10 | 广州广电运通金融电子股份有限公司 | Short-term traffic flow prediction method, storage medium and system under complex road network |
| CN116279512A (en) * | 2021-12-20 | 2023-06-23 | 通用汽车环球科技运作有限责任公司 | Predicting driver state using gaze behavior |
| CN114559959A (en) * | 2022-03-18 | 2022-05-31 | 重庆长安汽车股份有限公司 | Automatic driving behavior decision system and method based on driving preference |
| CN116206463A (en) * | 2023-03-06 | 2023-06-02 | 吉林大学 | Public road operation vehicle dispatch system |
| CN116539004A (en) * | 2023-05-04 | 2023-08-04 | 中国人民解放军国防科技大学 | Communication line engineering investigation design method and system adopting unmanned aerial vehicle mapping |
| CN116805449A (en) * | 2023-07-05 | 2023-09-26 | 哈尔滨工业大学 | Traffic signal fuzzy control method and system considering individual priori data |
| CN116804560A (en) * | 2023-08-23 | 2023-09-26 | 四川交通职业技术学院 | Unmanned automobile safety navigation method and device under controlled road section |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117804490A (en) | 2024-04-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20220222556A1 (en) | Road traffic jam early warning method and system | |
| CN111160753B (en) | Road network node importance assessment method based on knowledge graph | |
| EP3035314A1 (en) | A traffic data fusion system and the related method for providing a traffic state for a network of roads | |
| Fan et al. | A simple multi-objective optimization algorithm for the urban transit routing problem | |
| Zheng et al. | A stochastic simulation-based optimization method for equitable and efficient network-wide signal timing under uncertainties | |
| KR20230057558A (en) | Short-term traffic flow prediction method and device using deep learning | |
| Shamsi et al. | Reinforcement learning for traffic light control with emphasis on emergency vehicles | |
| Amiri et al. | An ensemble-based machine learning model for forecasting network traffic in VANET | |
| CN117671992A (en) | Intelligent bus dispatching method and system | |
| Ghoul et al. | Real-time safest route identification: Examining the trade-off between safest and fastest routes | |
| Abpeykar et al. | Supervised and unsupervised learning DSS for incident management in intelligent tunnel: A case study in Tehran Niayesh tunnel | |
| CN117804490B (en) | Comprehensive planning method and device for vehicle running route | |
| CN116592903B (en) | Ecological driving path real-time planning method for group preference under vehicle-road cooperative environment | |
| Sarker et al. | A data-driven reinforcement learning based multi-objective route recommendation system | |
| CN115713206A (en) | Bus individual trip decision model | |
| Yu et al. | Optimization of urban bus operation frequency under common route condition with rail transit | |
| JP7403421B2 (en) | Operation plan change support device, judgment model creation device, judgment model creation program, and train operation management system | |
| Habibi et al. | Improvement of Multi-agent Routing Guidance with an Intelligent Traffic Light Scheduling and the Ability to Select Intermediate Destinations | |
| Goparaju et al. | Optimization and Performance Evaluation of Hybrid Deep Learning Models for Traffic Flow Prediction | |
| Li et al. | Using Random Undersampling Boosting Classifier to Estimate Mode Shift Response to Bus Local Network Expansion and Bus Rapid Transit Services | |
| Prathilothamai et al. | Traffic prediction system using IoT cluster based evolutionary under sampling approach | |
| Mejdoubi et al. | Road traffic optimisation based on a learning approach | |
| Reddy et al. | Survey on traffic flow prediction for intelligent transportation system using machine learning | |
| Barua | ADiscrete ROUTE CHOICE MODEL USING SUPPORT VECTOR MACHINE IN CONTEXT OF DHAKA CITY | |
| Yuan | Emergency Transportation and Structural Optimization Issues of E-commerce Logistics Network Parcels |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |