Multi-dimensional map-based global management and control and instant command method and system
Through the multi-dimensional map-based whole-region control and real-time command methods, the dispatch of patrol vehicles is dynamically optimized, and the problem of congestion encountered by patrol vehicles on the way is solved, the efficiency and accuracy of road dredging are improved, and the management level and emergency response capabilities of urban traffic flow are improved.
Patent Information
- Application Number
- CN202510054855.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-14
- Publication Date
- 2025-04-29
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In urban traffic, when a traffic accident or congestion occurs, relying solely on dispatching patrol vehicles nearby may cause the vehicle to encounter additional congestion on the way, unable to arrive at the accident site in time, and delay the implementation of dredging work.
The multi-dimensional map-based whole-region control and real-time command method are adopted. By obtaining historical traffic monitoring data, the historical congestion characteristics of traffic sections are extracted, the congestion pattern collection and congestion characteristic map are generated, and the patrol vehicle scheduling is dynamically optimized to select the optimal path.
It improves the efficiency and accuracy of road dredging, helps patrol vehicles to quickly reach their destinations, avoid delays in the implementation of dredging work, and improves the management level and emergency response capabilities of urban traffic flows.
Smart Images
Figure CN119904989A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of emergency resource dispatching management, and in particular to a global control and instant command method and system based on a multi-dimensional map. Background Art
[0002] Traffic congestion is a common and increasingly serious problem in modern urban traffic management, especially during peak hours and when accidents occur, traffic flow increases sharply, causing serious road congestion. Some traffic management methods usually rely on road patrols, traffic signal control, and manual dispatching to alleviate congestion.
[0003] When a traffic accident or congestion occurs, the traffic command center dispatches the patrol vehicles closest to the accident site to the scene to perform the unblocking task based on real-time monitoring data. However, due to the complexity of the urban traffic network, especially during peak hours, certain sections of the road are often congested. If patrol vehicles are simply dispatched nearby after an accident, and if the congestion trends and changes in the entire area are not understood in real time, the patrol vehicles may encounter additional congestion on the way due to improper route selection and fail to arrive at the accident site in time, thus delaying the implementation of unblocking work. Summary of the invention
[0004] In order to solve the above technical problems, the present invention proposes a global control and real-time command method and system based on a multi-dimensional map, which improves the efficiency and accuracy of road clearing by dynamically optimizing the scheduling of patrol vehicles.
[0005] To achieve the above object, the present invention provides the following technical solutions:
[0006] A first aspect of the present invention provides a global control and real-time command method based on a multi-dimensional map, comprising:
[0007] Acquire historical traffic monitoring data of the target area, extract historical congestion characteristics of each traffic section in the target area based on the historical traffic monitoring data, and cluster congestion patterns of multiple traffic sections according to the historical congestion characteristics to generate multiple congestion pattern sets;
[0008] Extracting the congestion pattern features of each congestion pattern set, determining the congestion pattern set to which each traffic section belongs, optimizing the historical congestion features of each traffic section according to the congestion pattern features of the congestion pattern set, and constructing a congestion feature map of the target area, wherein the congestion feature map includes congestion period data of each traffic section;
[0009] After receiving the congestion clearing instruction, obtaining the position data of multiple patrol vehicles, determining a first vehicle candidate list for executing the congestion clearing instruction according to the position data of each patrol vehicle, and determining multiple candidate execution paths for each patrol vehicle in the first vehicle candidate list;
[0010] The estimated arrival time of each candidate execution path is analyzed according to the congestion feature map of the target area, the target execution path of each patrol vehicle is determined according to the estimated arrival time, the target vehicle is determined according to the estimated arrival time corresponding to the target execution path of the patrol vehicle, and the target vehicle is commanded to execute the congestion clearing instruction.
[0011] Preferably, the historical congestion characteristics of each traffic section are optimized according to the congestion pattern characteristics of the congestion pattern set to construct a congestion characteristic map of the target area, including:
[0012] For the congestion pattern characteristics of the congestion pattern set, average congestion parameters corresponding to multiple traffic sections in the congestion pattern set in multiple unit time periods are calculated, and a global congestion time series of the congestion pattern set is generated as the congestion pattern characteristics of the congestion pattern set;
[0013] According to the historical congestion characteristics of the traffic section, the local congestion parameters in multiple unit time periods are extracted to construct the local congestion time series of each traffic section. According to the global congestion time series and the local congestion time series, the recursive congestion distance matrix of each traffic section is constructed. For the recursive congestion distance matrix:
[0014] D ij =d(i,j)+min(d(i-1,j),d(i,j-1),d(i-1,j-1))
[0015] Where D ij represents the recursive congestion distance corresponding to the point with coordinates (i, j) in the recursive congestion distance matrix, d(i, j), d(i-1, j), d(i, j-1), and d(i-1, j-1) represent the distances between the corresponding element points in the global congestion time series and the local congestion time series respectively;
[0016] The optimal alignment path corresponding to the minimum recursive congestion distance is determined according to the recursive congestion distance matrix, multiple local congestion periods in the traffic section are determined according to the local congestion time series, and multiple global congestion periods of the congestion mode set are determined according to the global congestion time series. The multiple local congestion periods in the traffic section are optimized according to the optimal alignment path and the multiple global congestion periods to obtain the corrected congestion period corresponding to each local congestion period of the traffic section, and a congestion feature map of the target area is constructed according to the multiple corrected congestion periods of each traffic section.
[0017] Preferably, multiple local congestion periods in the traffic section are optimized according to the optimal alignment path and multiple global congestion periods to obtain a modified congestion period corresponding to each local congestion period of the traffic section, including:
[0018] According to the optimal alignment path, multiple local congestion periods are mapped to the global congestion time series, a target mapping period for each local congestion period, and a time deviation parameter and a congestion intensity deviation parameter for each target mapping period are determined, and multiple local congestion periods of the traffic section are optimized according to the time deviation parameter and the congestion intensity deviation parameter to obtain a corrected congestion period corresponding to each local congestion period of the traffic section;
[0019] Among them, for the time deviation parameter and the congestion intensity deviation parameter, the absolute value of the time difference between the local congestion period and the target mapping period is calculated to obtain the time deviation parameter, the average value of the local congestion parameters under multiple unit time periods in the local congestion period is calculated to obtain the first congestion parameter, and the average value of the average congestion parameters of multiple traffic sections under multiple unit time periods in the global congestion period is calculated to obtain the second congestion parameter, and the absolute value of the difference between the first congestion parameter and the second congestion parameter is used as the congestion intensity deviation parameter;
[0020] The time deviation parameter is corrected according to the congestion intensity deviation parameter to generate a target time parameter. If the target time parameter is less than the preset time deviation threshold, the local congestion period is evenly diffused based on the target mapping period to obtain a corrected congestion period. Otherwise, the local congestion period is offset based on the target mapping period to obtain a corrected congestion period.
[0021] Preferably, the process of directing the target vehicle to execute the congestion clearing instruction also includes:
[0022] Acquire multiple derived reference paths about the target execution path in real time when the target vehicle is on its way to the destination for executing the congestion relief instruction through the target execution path, and add the local congestion time series associated with each derived reference path to the congestion feature map;
[0023] The current congestion parameter of each derived reference path is calculated in real time, and a target derived path is selected according to the local congestion time series associated with the derived reference path in the congestion feature map, including selecting a derived reference path with the shortest driving distance to the destination for executing the congestion relief instruction as the target derived path after detecting that the current congestion parameter of at least one derived reference path is less than the congestion parameter at the current moment in the local congestion time series, and immediately directing the target vehicle to change its path according to the target derived path.
[0024] Preferably, mapping multiple local congestion periods into the global congestion time series according to the optimal alignment path, and determining a target mapping period for each local congestion period, comprises:
[0025] The correspondence between the local congestion time series and multiple unit time periods in the global congestion time series is determined according to the optimal alignment path. For any local congestion time period, the unit time period corresponding to each unit time period after being mapped in the global congestion time series is determined according to the multiple unit time periods involved in the local congestion time series, and the mapping reference time period after the local congestion time period is mapped to the global congestion time series is obtained. The global congestion time period in the global congestion time series that is closest to the mapping reference time period is recorded as the target mapping time period of the local congestion time period.
[0026] Preferably, the unit time period in which the average congestion parameter in the global congestion time series is greater than a preset congestion parameter threshold is recorded as the global congestion time period, and the unit time period in which the local congestion parameter in the local congestion time series is greater than a preset congestion parameter threshold is recorded as the local congestion time period.
[0027] A second aspect of the present invention provides a global control and real-time command system based on a multi-dimensional map, which is used to implement the above-mentioned global control and real-time command method based on a multi-dimensional map, including:
[0028] A historical data analysis module is used to obtain historical traffic monitoring data of a target area, extract historical congestion characteristics of each traffic section in the target area based on the historical traffic monitoring data, and cluster congestion patterns of multiple traffic sections according to the historical congestion characteristics to generate multiple congestion pattern sets;
[0029] A congestion feature map generation module is used to extract the congestion mode features of each congestion mode set, determine the congestion mode set to which each traffic section belongs, optimize the historical congestion features of each traffic section according to the congestion mode features of the congestion mode set, and construct a congestion feature map of the target area, wherein the congestion feature map includes the congestion period data of each traffic section;
[0030] A congestion clearing instruction execution analysis module is used to obtain the position data of multiple patrol vehicles after receiving the congestion clearing instruction, determine a first vehicle candidate list for executing the congestion clearing instruction according to the position data of each patrol vehicle, and determine multiple candidate execution paths for each patrol vehicle in the first vehicle candidate list;
[0031] The congestion relief instruction execution management module is used to analyze the estimated arrival time of each candidate execution path according to the congestion feature map of the target area, determine the target execution path of each patrol vehicle according to the estimated arrival time, determine the target vehicle according to the estimated arrival time corresponding to the target execution path of the patrol vehicle, and command the target vehicle to execute the congestion relief instruction.
[0032] Preferably, the congestion feature map generation module optimizes the historical congestion features of each traffic section according to the congestion mode features of the congestion mode set to construct a congestion feature map of the target area, including:
[0033] For the congestion pattern characteristics of the congestion pattern set, average congestion parameters corresponding to multiple traffic sections in the congestion pattern set in multiple unit time periods are calculated, and a global congestion time series of the congestion pattern set is generated as the congestion pattern characteristics of the congestion pattern set;
[0034] According to the historical congestion characteristics of the traffic section, the local congestion parameters in multiple unit time periods are extracted to construct the local congestion time series of each traffic section. According to the global congestion time series and the local congestion time series, the recursive congestion distance matrix of each traffic section is constructed. For the recursive congestion distance matrix:
[0035] D ij =d(i,j)+min(d(i-1,j),d(i,j-1),d(i-1,j-1))
[0036] Where D ij represents the recursive congestion distance corresponding to the point with coordinates (i, j) in the recursive congestion distance matrix, d(i, j), d(i-1, j), d(i, j-1), and d(i-1, j-1) represent the distances between the corresponding element points in the global congestion time series and the local congestion time series respectively;
[0037] The optimal alignment path corresponding to the minimum recursive congestion distance is determined according to the recursive congestion distance matrix, multiple local congestion periods in the traffic section are determined according to the local congestion time series, and multiple global congestion periods of the congestion mode set are determined according to the global congestion time series. The multiple local congestion periods in the traffic section are optimized according to the optimal alignment path and the multiple global congestion periods to obtain the corrected congestion period corresponding to each local congestion period of the traffic section, and a congestion feature map of the target area is constructed according to the multiple corrected congestion periods of each traffic section.
[0038] The present invention has the following beneficial effects:
[0039] The present invention analyzes the congestion variation rules of the target area in different time periods through historical traffic monitoring data, identifies the congestion modes corresponding to different traffic sections, extracts the congestion period information of each traffic section and constructs a congestion feature map, which is used to assist in analyzing and predicting the congestion trends of different traffic sections in the target area. It can make early warnings and optimize path selection in advance when patrol vehicles need to go to different locations to perform unblocking tasks, and assist patrol vehicles to quickly reach the destination to avoid delays in the implementation of unblocking work as much as possible; at the same time, when the patrol vehicle is driving on the way to the destination, the collected real-time traffic data is combined with the congestion feature map, and on the basis of the path planned by considering the congestion prediction situation, the pre-set driving path is locally optimized according to the real-time data, and the target vehicle is helped to avoid congestion through precise path selection and real-time adjustment, thereby improving road traffic efficiency to enhance emergency response capabilities, and improving the management level and emergency handling capabilities of urban traffic flow. BRIEF DESCRIPTION OF THE DRAWINGS
[0040] Figure 1 A flowchart of a global control and real-time command method based on a multi-dimensional map provided in an embodiment of the present invention.
[0041] Figure 2 A schematic diagram of the structure of a global control and real-time command system based on a multi-dimensional map provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0042] In order to make the technical scheme in the present invention better understood by the persons skilled in the art, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.
[0043] See also Figure 1 The embodiment of the present invention provides a global control and real-time command method based on a multi-dimensional map, which specifically includes the following steps:
[0044] Step S1, obtaining historical traffic monitoring data of a target area, extracting historical congestion characteristics of each traffic section in the target area based on the historical traffic monitoring data, and clustering congestion patterns of multiple traffic sections according to the historical congestion characteristics to generate multiple congestion pattern sets.
[0045] Specifically, the historical traffic monitoring data can be the long-term monitoring data of all traffic sections in the target area, such as one month or three months, to ensure that the data covers a sufficiently long time range (to capture a variety of different types of congestion patterns, and to perform preprocessing such as outlier removal, data smoothing, and time synchronization on these data to ensure data quality and accuracy. Then, for each traffic section in the target area, the corresponding historical congestion features are extracted, including but not limited to traffic flow, average vehicle speed, road utilization, and the proportion of traffic congestion in different time periods. Based on the extracted historical congestion features, a traffic congestion model for each traffic section is constructed. The historical congestion feature vectors of the sections are obtained, and the congestion patterns of multiple traffic sections are clustered based on the historical congestion feature vectors. For example, the K-means algorithm is used to cluster all traffic sections for the purpose of congestion pattern analysis and identification. Each cluster obtained by clustering represents a specific congestion pattern, so that sections with similar congestion characteristics are classified into one category, and then multiple congestion pattern sets are constructed based on the clustering results, and each set represents a different type of congestion scenario. For example, some sections may often be congested during certain periods of time, other sections may be temporarily congested due to accidents or construction, and some sections are smooth all day long.
[0046] Step S2: extract the congestion pattern features of each congestion pattern set, determine the congestion pattern set to which each traffic section belongs, optimize the historical congestion features of each traffic section according to the congestion pattern features of the congestion pattern set, and construct a congestion feature map of the target area.
[0047] Specifically, the historical data of multiple traffic sections included in each congestion pattern set are analyzed in detail to extract the representative features of the pattern. In this embodiment, the data in multiple unit time periods are analyzed specifically to extract the congestion-related parameters that characterize the overall level of the set in each unit time period to obtain the congestion pattern features of each congestion pattern set, and for multiple traffic sections in the congestion pattern set, the historical congestion features of each traffic section are optimized based on the congestion pattern features of the congestion pattern set to reduce the impact of noise data on each traffic section. In this process, the congestion period data of each traffic section is optimized to obtain congestion period data with less noise and in line with group characteristics, and a congestion feature map is constructed based on these data, which at least includes information such as the congestion period and traffic flow of each section in the target area, which can be used to analyze the historical congestion patterns of certain sections, and as a decision support tool when receiving new real-time congestion dredging instructions.
[0048] Step S3, after receiving the congestion clearing instruction, obtain the position data of multiple patrol vehicles, determine a first vehicle candidate list for executing the congestion clearing instruction according to the position data of each patrol vehicle, and determine multiple candidate execution paths for each patrol vehicle in the first vehicle candidate list.
[0049] Specifically, when a traffic accident occurs on a certain road section or other factors cause congestion and the road needs to be cleared, the location information of the road section that needs to be cleared is determined according to the congestion clearing instruction, and then multiple patrol vehicles in the target area are obtained. For example, these data can be obtained through GPS, vehicle-mounted positioning systems, etc., and multiple patrol vehicles that are closer to the road section that needs to be cleared are determined to construct a first vehicle candidate list. For the multiple patrol vehicles in the first vehicle candidate list, multiple paths are determined by which each patrol vehicle can quickly reach the road section that needs to be cleared, and multiple candidate execution paths for each patrol vehicle in the first vehicle candidate list are obtained, wherein the candidate execution paths can be roughly determined based on the actual traffic conditions of each current road section.
[0050] Step S4, analyzing the estimated arrival time of each candidate execution path according to the congestion feature map of the target area, determining the target execution path of each patrol vehicle according to the estimated arrival time, determining the target vehicle according to the estimated arrival time corresponding to the target execution path of the patrol vehicle, and commanding the target vehicle to execute the congestion clearing instruction.
[0051] Specifically, for multiple candidate execution paths for each patrol vehicle, the estimated arrival time of each candidate execution path is determined based on the congestion period information of each traffic section contained in the congestion feature map of the target area, that is, the time periods that may be congested obtained based on historical data analysis are considered, and the travel time based on road congestion is estimated based on historical rules. Therefore, according to the estimated arrival time of the candidate execution path, the target execution path of each patrol vehicle in the first vehicle candidate list is determined, that is, the path that each patrol vehicle can estimate to reach the section that needs to be unblocked the fastest. Finally, the vehicle with the shortest time is selected from multiple patrol vehicles to be dispatched to execute the congestion unblocking instruction, so as to better deal with the road congestion that may occur in the process of going to the section that needs to be unblocked, and select a suitable route to facilitate the patrol vehicle to quickly reach the location where congestion unblocking is required, avoid delays in the implementation of unblocking work, and improve the efficiency of road unblocking.
[0052] In a possible implementation, for step S2, the historical congestion characteristics of each traffic section are optimized according to the congestion pattern characteristics of the congestion pattern set to construct a congestion characteristic map of the target area, specifically including:
[0053] For the congestion pattern characteristics of the congestion pattern set, the average congestion parameters corresponding to multiple traffic sections in the congestion pattern set in multiple unit time periods are calculated, and the global congestion time series of the congestion pattern set is generated as the congestion pattern characteristics of the congestion pattern set.
[0054] Specifically, for the congestion pattern characteristics of each congestion pattern set, the average congestion parameters corresponding to multiple traffic sections in the congestion pattern set in multiple unit time periods are calculated, and the global congestion time series of the congestion pattern set is constructed according to the average congestion parameters in each unit time period, which is recorded as the congestion pattern characteristics of the congestion pattern set. Among them, for the congestion parameters, the congestion parameters of each traffic section in the unit time period are calculated. In this embodiment, the ratio of the average flow rate to the average vehicle speed of the traffic section in the unit time period is recorded as the congestion parameter. The unit time period is specifically a reference time period for analyzing data, which can be 10 minutes, 15 minutes, etc. Every day is divided into multiple unit time periods, and the congestion-related characteristics of each traffic section in each unit time period are analyzed. The average of the congestion parameters corresponding to multiple traffic sections in each unit time period is obtained to obtain the average congestion parameters in the congestion pattern set for each unit time period, thereby constructing the global congestion time series of the congestion pattern set.
[0055] Similarly, according to the historical congestion characteristics of the traffic section, the local congestion parameters in multiple unit time periods are extracted to construct the local congestion time series of each traffic section. That is, according to the historical congestion characteristics of the traffic section, the mean of the congestion parameters of each traffic section in each unit time period is calculated to obtain the local congestion parameters of the traffic section in each unit time period, and thus the local congestion time series of each traffic section is constructed based on multiple local congestion parameters.
[0056] Then, the local congestion time series of each traffic section is analyzed according to the global congestion time series, and the recursive congestion distance matrix of each traffic section is constructed. For the recursive congestion distance matrix:
[0057] D ij =d(i,j)+min(d(i-1,j),d(i,j-1),d(i-1,j-1))
[0058] Where D ijIt represents the recursive congestion distance corresponding to the point with coordinates (i, j) in the recursive congestion distance matrix. d(i, j), d(i-1, j), d(i, j-1), and d(i-1, j-1) respectively represent the distances between the corresponding element points in the global congestion time series and the local congestion time series. Specifically, the horizontal and vertical coordinates of the recursive congestion distance matrix are used to represent the congestion characteristic values of the corresponding element points in the global congestion time series and the local congestion time series, that is, the average congestion parameter and the local congestion parameter. The value under each coordinate represents the difference between the corresponding element points in the global congestion time series and the local congestion time series. d(i-1, j), d(i, j-1), and d(i-1, j-1) respectively represent the difference between the congestion characteristic values of the corresponding element points in the global congestion time series and the local congestion time series at the three points to the left, above, and upper left of the point with coordinates (i, j).
[0059] The recursive congestion distance of each element point in the recursive congestion distance matrix takes into account the local shortest path. The optimal alignment path corresponding to the minimum recursive congestion distance can be determined according to the recursive congestion distance matrix. Specifically, starting from the lower right corner of the recursive congestion distance matrix, determine which path has the shortest cumulative distance to the left, above and upper right of the coordinate point, so as to determine the second path point in the local shortest path except the starting point. The three path points all take into account the cumulative distance relationship with the surrounding coordinate points to achieve the concept of local optimality. In the process of generating the optimal alignment path, the selection of each path point except the starting point adopts the concept of local optimality, so as to obtain the optimal alignment path that characterizes the local optimal relationship, so as to reflect the corresponding relationship between different time periods in the global congestion time series and the local congestion time series.
[0060] After determining multiple local congestion periods in a traffic section according to a local congestion time series and determining multiple global congestion periods of a congestion pattern set according to a global congestion time series, the multiple local congestion periods in the traffic section can be optimized according to the optimal alignment path and the multiple global congestion periods to obtain a corrected congestion period corresponding to each local congestion period of the traffic section, and finally a congestion feature map of the target area is constructed according to the multiple corrected congestion periods of each traffic section.
[0061] Specifically, for the determination of congestion time periods, they can be divided based on preset congestion parameter thresholds. If the congestion characteristic value of a unit time period is greater than the preset congestion parameter threshold, the unit time period is recorded as a congestion time period. For multiple local congestion time periods in each traffic section, since the optimal alignment path of the traffic section contains the correspondence between each unit time period and multiple unit sections of the congestion mode set, it may specifically involve corresponding change relationships such as translation and scaling of the time period. Therefore, each local congestion time period of the traffic section can be processed accordingly according to this correspondence to obtain multiple corrected congestion time periods that are more in line with group characteristics, so as to remove some interference information in the local characteristics of each traffic section based on group characteristics, while ensuring compliance with group characteristics and maintaining the personalized characteristics of each section. Finally, a congestion feature map of the target area is constructed based on the multiple corrected congestion time periods of each traffic section obtained by optimization, so as to facilitate the display of the congestion distribution of different traffic sections in the target area based on historical data analysis, and accurately reflect the congestion-related characteristics of each traffic section in the historical data.
[0062] In a possible implementation, multiple local congestion periods in a traffic section are optimized according to the optimal alignment path and multiple global congestion periods to obtain a modified congestion period corresponding to each local congestion period of the traffic section, specifically including:
[0063] According to the optimal alignment path, multiple local congestion periods are mapped into the global congestion time series, and the target mapping period of each local congestion period, as well as the time deviation parameter and the congestion intensity deviation parameter of each target mapping period are determined.
[0064] Specifically, the optimal alignment path includes the correspondence between multiple unit time periods in different local congestion time series and the global congestion time series. For example, assuming that a path point in the optimal alignment path moves upward, it means that the horizontal coordinate corresponds to two vertical coordinates, indicating that the unit time period corresponding to the horizontal coordinate is related to the unit time period corresponding to the two vertical coordinates. In this way, the correspondence between multiple unit time periods in the local congestion time series and the global congestion time series can be determined. Therefore, for any local congestion period, according to the multiple unit time periods involved in the local congestion period, the unit time period corresponding to each unit time period after being mapped in the global congestion time series is determined, and finally the multiple unit time periods obtained by mapping are combined to obtain the time period corresponding to the local congestion period in the global congestion time series, which is recorded as the mapping reference time period, which represents the time period actually corresponding to the local congestion period in the global congestion time series after operations such as scaling and translation, and then the global congestion period in the global congestion time series that is closest to the mapping reference time period is determined, which is recorded as the target mapping time period of the local congestion period, that is, the association between the local characteristics of the congestion period in the traffic section and the global characteristics of the congestion period in the congestion pattern set after considering the translation and scaling of the time period.
[0065] For the target mapping period of the local congestion period, the time deviation parameter and congestion intensity deviation parameter of the target mapping period are further analyzed and calculated. Among them, for the time deviation parameter, the absolute value of the time difference between the local congestion period and the target mapping period is calculated to obtain the time deviation parameter. The time deviation parameter is specifically the length of the period that is not covered at the same time between the local congestion period and the target mapping period, which is used to quantify the time difference between the local congestion period and the target mapping period, and characterize the deviation between the two in terms of time period coverage.
[0066] For the congestion intensity deviation parameter, the first congestion parameter is obtained by calculating the average of the local congestion parameters in multiple unit time periods in the local congestion period, and the second congestion parameter is obtained by calculating the average of the average congestion parameters in multiple unit time periods in the global congestion period. The absolute value of the difference between the first congestion parameter and the second congestion parameter is used as the congestion intensity deviation parameter, which is used to quantify the difference in congestion intensity between the local congestion period and the global congestion period, and to characterize the deviation between the two in the congestion degree of the section.
[0067] After calculating the time deviation parameter and the congestion intensity deviation parameter of the target mapping period, multiple local congestion periods of the traffic section are optimized according to the time deviation parameter and the congestion intensity deviation parameter to obtain a corrected congestion period corresponding to each local congestion period of the traffic section.
[0068] Specifically, in the process of optimizing the local congestion period, the time deviation parameter is first corrected according to the congestion intensity deviation parameter to generate the target time parameter. The generation of the target time parameter starts from the difference in congestion degree, and the difference in congestion degree is incorporated into the deviation between the local congestion period and the global congestion period in the period coverage. Specifically, the congestion intensity deviation parameter is used as the weight to weight the time deviation parameter to obtain the target time parameter, and then based on the generated target time parameter, it is decided whether to evenly diffuse or offset the local congestion period.
[0069] Specifically, if the target time parameter is less than the preset time deviation threshold, the local congestion period is uniformly diffused based on the target mapping period to obtain a corrected congestion period, otherwise the local congestion period is offset based on the target mapping period to obtain a corrected congestion period. Among them, uniform diffusion of the local congestion period specifically includes, with the target mapping period as the target, diffusing the local congestion period to the middle period of the corresponding target mapping period. It can be understood that the local congestion period and the target mapping period respectively involve a part of the period. After determining the corresponding period boundary of the two, the average of the left and right boundaries of the two is used as the boundary of the corrected congestion period to generate the corrected congestion period obtained by uniform diffusion of the local congestion period, so as to achieve uniform expansion or compression of the local congestion period according to the corresponding congestion mode of the target mapping section. The offset processing of the local congestion period to obtain the corrected congestion period includes taking the target mapping period as the offset target of the local congestion period, and performing a translation operation on the local congestion period so that the overlap between the local congestion period and the target mapping period is maximized, that is, the center point of the local congestion period is translated to coincide with the center of the target mapping period, and the corrected congestion period after the offset processing of the local congestion period is obtained. Through the above method, each local congestion period of the traffic section can better meet the congestion situation of the global congestion period while retaining its own personalized characteristics, and more accurately reflect the changing law of traffic flow in the target area.
[0070] In a possible implementation, for step S4, in the process of directing the target vehicle to execute the congestion clearing instruction, the process further includes:
[0071] Multiple derived reference paths about the target execution path are obtained in real time when the target vehicle is on its way to the destination for executing the congestion relief instruction through the target execution path, and the local congestion time series associated with each derived reference path is added to the congestion feature map.
[0072] Specifically, in the process of the target vehicle executing the congestion relief instruction, the target execution path of the target vehicle from the current location to the destination is data generated by analyzing the congestion trends of different roads through historical data. Some sections may be congested with a high probability in the future, so they are avoided in advance. However, considering that the predicted congestion may not actually occur, some of the nearest derived reference paths are obtained in real time during the vehicle's advancement. The derived reference path can specifically be a traffic section at a certain distance adjacent to it, such as a traffic section to which the vehicle can be transferred after leaving the current traffic section. The selection of the derived reference path can be based on real-time road congestion related information. For example, when it is preliminarily detected that a certain section is not congested, and the target vehicle can travel to the destination faster by choosing to travel from this section than by taking the target execution path to the destination, a derived reference path can be determined based on this section. In this way, multiple derived reference paths can be determined. It is worth noting that the existing technology is to roughly analyze whether a road section is congested and perform route planning based on the real-time traffic information of the road. This implementation does not specifically limit it. The present invention mainly considers that under the background of the existing real-time analysis of the congestion situation of the road section, the path optimization is combined with the congestion prediction situation, so as to more efficiently command and dispatch patrol vehicles. After determining multiple derived reference paths, the local congestion time series associated with each derived reference path is added to the congestion feature map.
[0073] Then, the current congestion parameters of each derived reference path are calculated in real time, and the target derived path is selected according to the local congestion time series associated with the derived reference path in the congestion feature map.
[0074] Specifically, for multiple derived reference paths determined in real time, the current congestion parameters of each derived reference path are obtained, and the congestion parameters of the derived reference path at the current moment in the local congestion time series, that is, the parameters obtained through historical data analysis, are compared with the congestion parameters obtained through current actual monitoring and analysis. After detecting that the current congestion parameters of at least one derived reference path are smaller than the congestion parameters at the current moment in the local congestion time series, the derived reference path with the shortest driving distance to the destination where the congestion relief instruction is to be executed is selected as the target derived path. It is worth noting that the congestion parameter corresponding to each period of the local congestion time series integrates the global performance of the traffic road under normal unobstructed conditions and congested conditions. The calculated congestion parameter is higher than the unobstructed condition and lower than the congested condition. Taking this parameter as the benchmark, if the current congestion parameter of the traffic section exceeds the congestion parameter determined based on historical data, it means that the current congestion situation of the traffic road is higher than the performance under unobstructed conditions, and it is likely to enter a congested condition. Otherwise, it is believed that it has not yet developed an obvious trend towards congestion. In this case, for the derived reference path whose current congestion parameter is less than the congestion parameter at the current moment in the local congestion time series, it means that there will be no congestion in the short term. In order to facilitate the target vehicle to reach the destination faster, the derived reference path that meets the corresponding conditions can be selected as the new driving path. In this process, the derived reference path with the shortest driving distance between the destinations to execute the congestion relief command is selected as the target derived path. The original target execution path is locally optimized through the target derived path, that is, the local path optimization is performed in a relatively close range. The command system can perform instant command on the target vehicle to change the path according to the target derived path to assist the target vehicle to reach the destination faster.
[0075] See also Figure 2 The embodiment of the present invention further provides a global control and real-time command system based on a multi-dimensional map, which is used to implement the above-mentioned global control and real-time command method based on a multi-dimensional map, specifically including:
[0076] A historical data analysis module is used to obtain historical traffic monitoring data of a target area, extract historical congestion characteristics of each traffic section in the target area based on the historical traffic monitoring data, and cluster congestion patterns of multiple traffic sections according to the historical congestion characteristics to generate multiple congestion pattern sets;
[0077] A congestion feature map generation module is used to extract the congestion mode features of each congestion mode set, determine the congestion mode set to which each traffic section belongs, optimize the historical congestion features of each traffic section according to the congestion mode features of the congestion mode set, and construct a congestion feature map of the target area, wherein the congestion feature map includes the congestion period data of each traffic section;
[0078] Among them, the historical congestion characteristics of each traffic section are optimized according to the congestion pattern characteristics of the congestion pattern set, and a congestion characteristic map of the target area is constructed, including:
[0079] For the congestion pattern characteristics of the congestion pattern set, average congestion parameters corresponding to multiple traffic sections in the congestion pattern set in multiple unit time periods are calculated, and a global congestion time series of the congestion pattern set is generated as the congestion pattern characteristics of the congestion pattern set;
[0080] According to the historical congestion characteristics of the traffic section, the local congestion parameters in multiple unit time periods are extracted to construct the local congestion time series of each traffic section. According to the global congestion time series and the local congestion time series, the recursive congestion distance matrix of each traffic section is constructed. For the recursive congestion distance matrix:
[0081] D ij =d(i,j)+min(d(i-1,j),d(i,j-1),d(i-1,j-1))
[0082] Where D ij represents the recursive congestion distance corresponding to the point with coordinates (i, j) in the recursive congestion distance matrix, d(i, j), d(i-1, j), d(i, j-1), and d(i-1, j-1) represent the distances between the corresponding element points in the global congestion time series and the local congestion time series respectively;
[0083] According to the optimal alignment path and the multiple global congestion periods, multiple local congestion periods in the traffic section are optimized to obtain a modified congestion period corresponding to each local congestion period of the traffic section, including:
[0084] According to the optimal alignment path, multiple local congestion periods are mapped to the global congestion time series, a target mapping period for each local congestion period, and a time deviation parameter and a congestion intensity deviation parameter for each target mapping period are determined, and multiple local congestion periods of the traffic section are optimized according to the time deviation parameter and the congestion intensity deviation parameter to obtain a corrected congestion period corresponding to each local congestion period of the traffic section;
[0085] Among them, for the time deviation parameter and the congestion intensity deviation parameter, the absolute value of the time difference between the local congestion period and the target mapping period is calculated to obtain the time deviation parameter, the average value of the local congestion parameters under multiple unit time periods in the local congestion period is calculated to obtain the first congestion parameter, and the average value of the average congestion parameters of multiple traffic sections under multiple unit time periods in the global congestion period is calculated to obtain the second congestion parameter, and the absolute value of the difference between the first congestion parameter and the second congestion parameter is used as the congestion intensity deviation parameter;
[0086] The time deviation parameter is corrected according to the congestion intensity deviation parameter to generate a target time parameter. If the target time parameter is less than the preset time deviation threshold, the local congestion period is evenly diffused based on the target mapping period to obtain a corrected congestion period. Otherwise, the local congestion period is offset based on the target mapping period to obtain a corrected congestion period.
[0087] A congestion clearing instruction execution analysis module is used to obtain the position data of multiple patrol vehicles after receiving the congestion clearing instruction, determine a first vehicle candidate list for executing the congestion clearing instruction according to the position data of each patrol vehicle, and determine multiple candidate execution paths for each patrol vehicle in the first vehicle candidate list;
[0088] The congestion relief instruction execution management module is used to analyze the estimated arrival time of each candidate execution path according to the congestion feature map of the target area, determine the target execution path of each patrol vehicle according to the estimated arrival time, determine the target vehicle according to the estimated arrival time corresponding to the target execution path of the patrol vehicle, and command the target vehicle to execute the congestion relief instruction.
[0089] The optimal alignment path corresponding to the minimum recursive congestion distance is determined according to the recursive congestion distance matrix, multiple local congestion periods in the traffic section are determined according to the local congestion time series, and multiple global congestion periods of the congestion mode set are determined according to the global congestion time series. The multiple local congestion periods in the traffic section are optimized according to the optimal alignment path and the multiple global congestion periods to obtain the corrected congestion period corresponding to each local congestion period of the traffic section, and a congestion feature map of the target area is constructed according to the multiple corrected congestion periods of each traffic section.
[0090] The above is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. Parts not described in detail in this specification belong to the prior art known to those skilled in the art.
Claims
1. A global control and real-time command method based on a multi-dimensional map, characterized in that: include: Acquire historical traffic monitoring data of the target area, extract historical congestion characteristics of each traffic section in the target area based on the historical traffic monitoring data, and cluster congestion patterns of multiple traffic sections according to the historical congestion characteristics to generate multiple congestion pattern sets; Extracting the congestion pattern features of each congestion pattern set, determining the congestion pattern set to which each traffic section belongs, optimizing the historical congestion features of each traffic section according to the congestion pattern features of the congestion pattern set, and constructing a congestion feature map of the target area, wherein the congestion feature map includes congestion period data of each traffic section; After receiving the congestion clearing instruction, obtaining the position data of multiple patrol vehicles, determining a first vehicle candidate list for executing the congestion clearing instruction according to the position data of each patrol vehicle, and determining multiple candidate execution paths for each patrol vehicle in the first vehicle candidate list; The estimated arrival time of each candidate execution path is analyzed according to the congestion feature map of the target area, the target execution path of each patrol vehicle is determined according to the estimated arrival time, the target vehicle is determined according to the estimated arrival time corresponding to the target execution path of the patrol vehicle, and the target vehicle is commanded to execute the congestion clearing instruction.
2. According to claim 1, a global control and real-time command method based on a multi-dimensional map is characterized in that: The historical congestion characteristics of each traffic section are optimized according to the congestion pattern characteristics of the congestion pattern set, and a congestion characteristic map of the target area is constructed, including: For the congestion pattern characteristics of the congestion pattern set, average congestion parameters corresponding to multiple traffic sections in the congestion pattern set in multiple unit time periods are calculated, and a global congestion time series of the congestion pattern set is generated as the congestion pattern characteristics of the congestion pattern set; According to the historical congestion characteristics of the traffic section, the local congestion parameters in multiple unit time periods are extracted to construct the local congestion time series of each traffic section. According to the global congestion time series and the local congestion time series, the recursive congestion distance matrix of each traffic section is constructed. For the recursive congestion distance matrix: D ij =d(i,j)+min(d(i-1,j),d(i,j-1),d(i-1,j-1)) Where D ij represents the recursive congestion distance corresponding to the point with coordinates (i, j) in the recursive congestion distance matrix, d(i, j), d(i-1, j), d(i, j-1), and d(i-1, j-1) represent the distances between the corresponding element points in the global congestion time series and the local congestion time series respectively; The optimal alignment path corresponding to the minimum recursive congestion distance is determined according to the recursive congestion distance matrix, multiple local congestion periods in the traffic section are determined according to the local congestion time series, and multiple global congestion periods of the congestion mode set are determined according to the global congestion time series. The multiple local congestion periods in the traffic section are optimized according to the optimal alignment path and the multiple global congestion periods to obtain the corrected congestion period corresponding to each local congestion period of the traffic section, and a congestion feature map of the target area is constructed according to the multiple corrected congestion periods of each traffic section.
3. The method for global control and real-time command based on a multi-dimensional map according to claim 2 is characterized in that: According to the optimal alignment path and the multiple global congestion periods, multiple local congestion periods in the traffic section are optimized to obtain a modified congestion period corresponding to each local congestion period of the traffic section, including: According to the optimal alignment path, multiple local congestion periods are mapped to the global congestion time series, a target mapping period for each local congestion period, and a time deviation parameter and a congestion intensity deviation parameter for each target mapping period are determined, and multiple local congestion periods of the traffic section are optimized according to the time deviation parameter and the congestion intensity deviation parameter to obtain a corrected congestion period corresponding to each local congestion period of the traffic section; Among them, for the time deviation parameter and the congestion intensity deviation parameter, the absolute value of the time difference between the local congestion period and the target mapping period is calculated to obtain the time deviation parameter, the average value of the local congestion parameters under multiple unit time periods in the local congestion period is calculated to obtain the first congestion parameter, and the average value of the average congestion parameters of multiple traffic sections under multiple unit time periods in the global congestion period is calculated to obtain the second congestion parameter, and the absolute value of the difference between the first congestion parameter and the second congestion parameter is used as the congestion intensity deviation parameter; The time deviation parameter is corrected according to the congestion intensity deviation parameter to generate a target time parameter. If the target time parameter is less than the preset time deviation threshold, the local congestion period is evenly diffused based on the target mapping period to obtain a corrected congestion period. Otherwise, the local congestion period is offset based on the target mapping period to obtain a corrected congestion period.
4. The method for global control and real-time command based on a multi-dimensional map according to claim 3 is characterized in that: The process of directing the target vehicle to execute the congestion clearing instruction also includes: Acquire multiple derived reference paths about the target execution path in real time when the target vehicle is on its way to the destination for executing the congestion relief instruction through the target execution path, and add the local congestion time series associated with each derived reference path to the congestion feature map; The current congestion parameter of each derived reference path is calculated in real time, and a target derived path is selected according to the local congestion time series associated with the derived reference path in the congestion feature map, including selecting a derived reference path with the shortest driving distance to the destination for executing the congestion relief instruction as the target derived path after detecting that the current congestion parameter of at least one derived reference path is less than the congestion parameter at the current moment in the local congestion time series, and immediately directing the target vehicle to change its path according to the target derived path.
5. The method for global control and real-time command based on a multi-dimensional map according to claim 4 is characterized in that: Mapping multiple local congestion periods into the global congestion time series according to the optimal alignment path, and determining the target mapping period for each local congestion period, including: The correspondence between the local congestion time series and multiple unit time periods in the global congestion time series is determined according to the optimal alignment path. For any local congestion time period, the corresponding unit time period after each unit time period is mapped in the global congestion time series is determined according to the multiple unit time periods involved in the local congestion time series, and the mapping reference time period after the local congestion time period is mapped to the global congestion time series is obtained. The global congestion time period in the global congestion time series that is closest to the mapping reference time period is recorded as the target mapping time period of the local congestion time period.
6. The method for global control and real-time command based on a multi-dimensional map according to claim 2, characterized in that: The unit time period in which the average congestion parameter in the global congestion time series is greater than the preset congestion parameter threshold is recorded as the global congestion period, and the unit time period in which the local congestion parameter in the local congestion time series is greater than the preset congestion parameter threshold is recorded as the local congestion period.
7. A global control and real-time command system based on a multi-dimensional map, characterized in that: The system is used to implement a global control and real-time command method based on a multi-dimensional map as described in any one of claims 1 to 6, including: A historical data analysis module is used to obtain historical traffic monitoring data of a target area, extract historical congestion characteristics of each traffic section in the target area based on the historical traffic monitoring data, and cluster congestion patterns of multiple traffic sections according to the historical congestion characteristics to generate multiple congestion pattern sets; A congestion feature map generation module is used to extract the congestion mode features of each congestion mode set, determine the congestion mode set to which each traffic section belongs, optimize the historical congestion features of each traffic section according to the congestion mode features of the congestion mode set, and construct a congestion feature map of the target area, wherein the congestion feature map includes the congestion period data of each traffic section; A congestion clearing instruction execution analysis module is used to obtain the position data of multiple patrol vehicles after receiving the congestion clearing instruction, determine a first vehicle candidate list for executing the congestion clearing instruction according to the position data of each patrol vehicle, and determine multiple candidate execution paths for each patrol vehicle in the first vehicle candidate list; The congestion relief instruction execution management module is used to analyze the estimated arrival time of each candidate execution path according to the congestion feature map of the target area, determine the target execution path of each patrol vehicle according to the estimated arrival time, determine the target vehicle according to the estimated arrival time corresponding to the target execution path of the patrol vehicle, and command the target vehicle to execute the congestion relief instruction.
8. The global control and real-time command system based on multi-dimensional maps according to claim 7 is characterized in that: For the congestion feature map generation module, the historical congestion features of each traffic section are optimized according to the congestion mode features of the congestion mode set, and a congestion feature map of the target area is constructed, including: For the congestion pattern characteristics of the congestion pattern set, average congestion parameters corresponding to multiple traffic sections in the congestion pattern set in multiple unit time periods are calculated, and a global congestion time series of the congestion pattern set is generated as the congestion pattern characteristics of the congestion pattern set; According to the historical congestion characteristics of the traffic section, the local congestion parameters in multiple unit time periods are extracted to construct the local congestion time series of each traffic section. According to the global congestion time series and the local congestion time series, the recursive congestion distance matrix of each traffic section is constructed. For the recursive congestion distance matrix: D ij =d(i,j)+min(d(i-1,j),d(i,j-1),d(i-1,j-1)) Where D ij represents the recursive congestion distance corresponding to the point with coordinates (i, j) in the recursive congestion distance matrix, d(i, j), d(i-1, j), d(i, j-1), and d(i-1, j-1) represent the distances between the corresponding element points in the global congestion time series and the local congestion time series respectively; The optimal alignment path corresponding to the minimum recursive congestion distance is determined according to the recursive congestion distance matrix, multiple local congestion periods in the traffic section are determined according to the local congestion time series, and multiple global congestion periods of the congestion mode set are determined according to the global congestion time series. The multiple local congestion periods in the traffic section are optimized according to the optimal alignment path and the multiple global congestion periods to obtain the corrected congestion period corresponding to each local congestion period of the traffic section, and a congestion feature map of the target area is constructed according to the multiple corrected congestion periods of each traffic section.