Urban distribution vehicle travel feature extraction method, optimization method and device based on space-time and geographic information system, and medium

Abstract

The invention relates to an urban distribution vehicle travel feature extraction method, optimization method and device based on a space-time and geographic information system, and a medium. The extraction method comprises the following steps: S1, obtaining a track of an urban distribution vehicle; s2, constructing one or more trip chains based on the trajectory of each urban distribution vehicle; and S3, based on the constructed trip chain, respectively extracting corresponding trip chain features. Compared with the prior art, the characteristics are extracted based on the trip chain, so that the breadth and depth of the urban distribution vehicle trip characteristics can be improved, and a more valuable planning basis is provided.

Description

Technical Field

[0001] The present invention relates to, in particular to, a method for extracting travel characteristics of urban distribution vehicles based on time-space and geographic information systems, an optimization method, a device and a medium. Background Art

[0002] The travel of urban distribution vehicles not only has a direct impact on urban transportation and freight logistics, but also affects the formulation of traffic management policies. Therefore, it is particularly important to conduct a systematic analysis of the travel characteristics of urban distribution vehicles. The travel characteristics of urban distribution vehicles mainly refer to the static and dynamic laws reflected in time, space, etc. when urban distribution vehicles carry out freight distribution activities.

[0003] The trip chain theory of urban distribution vehicles is the core of traffic flow analysis. It describes in detail the movement sequence of individuals or vehicles within a specific time, thus providing a theoretical basis for traffic flow allocation, path planning, and the formulation of traffic policies. In the field of freight, the trip chain not only reflects the route selection of vehicles, but also goes deep into the details of the stops at freight business points, the loading and unloading process of goods, and the stay time at each stop, thus providing a comprehensive and in-depth understanding of freight activities.

[0004] The research object of traditional freight travel chain is mainly medium and long-distance freight. Compared with urban freight distribution, there are many differences. The main characteristics of the construction of traditional freight travel chain are:

[0005] 1. Focus on important nodes. Traditional freight travel chain research focuses on identifying large sites such as major warehouses, factories and freight bases as important nodes. These nodes are usually located in the outskirts of cities or industrial areas and are the starting and end points of freight activities.

[0006] 2. Set a higher dwell time threshold. In order to distinguish important freight activities, traditional methods usually set a higher node dwell time threshold, such as 1 hour or longer. Only when the vehicle stays at a specific location for more than this threshold, the location is considered part of the travel chain.

[0007] 3. Exclude all stopover points in the road network. Since traditional methods focus on long-term stops at important nodes, stopover points in the road network are usually regarded as non-freight business points, and behaviors such as short-term parking on the road are usually excluded. However, urban freight distribution is different from traditional freight in many aspects, which leads to the need to consider the following factors in particular when constructing the urban distribution vehicle travel chain:

[0008] (3.1) Diverse locations of freight activities. Urban freight distribution includes not only large warehouses and factories, but also temporary stops at retail stores, restaurants, and other commercial locations, which are more widely distributed and diverse.

[0009] (3.2) Shorter loading and unloading time. Compared with long-distance freight, the loading and unloading time of urban distribution vehicles is relatively short, generally between 15 and 45 minutes, which requires a lower dwell time threshold in the construction of travel chains.

[0010] (3.3) Including on-street parking points. Urban distribution activities include a large number of on-street parking behaviors, and these parking points are crucial to understanding urban freight distribution activities.

[0011] In summary, the travel characteristics in the existing technology cannot reflect the travel patterns of urban distribution vehicles and cannot solve congestion. Summary of the invention

[0012] The purpose of the present invention is to provide a method, an optimization method, a device and a medium for extracting the travel characteristics of urban distribution vehicles based on time, space and geographic information systems, and extracting characteristics based on the travel chain, so as to improve the breadth and depth of the travel characteristics of urban distribution vehicles and provide more valuable planning basis.

[0013] The purpose of the present invention can be achieved by the following technical solutions:

[0014] A method for extracting travel characteristics of urban distribution vehicles based on spatiotemporal and geographic information systems, comprising:

[0015] Step S1: Obtain the trajectory of the city distribution vehicle;

[0016] Step S2: construct one or more travel chains based on the trajectories of the city distribution vehicles;

[0017] Step S3: Based on the constructed travel chains, corresponding travel chain features are extracted respectively.

[0018] The attribute information of each track point in the track of the urban distribution vehicle includes at least time and longitude and latitude.

[0019] The types of the travel chain include closed type and linear type.

[0020] The process of constructing a travel chain based on the trajectory of a single city distribution vehicle in step S2 includes:

[0021] Step S2-1: Perform motion detection on all track points in the track of the city distribution vehicle, extract all stop points, and generate the position, stop start and end time, and stop duration of each stop point;

[0022] Step S2-2: taking the first stop point as the starting point of the first travel chain;

[0023] Step S2-3: Select the next stop point according to the time sequence and calculate the distance with the starting point of the current travel chain;

[0024] Step S2-4: Determine whether the distance between the current stop point and the current travel link is less than a first distance threshold, if yes, execute step S2-5, otherwise, execute step S2-7;

[0025] Step S2-5: taking the current stop point as the end point of the current travel chain, and determining whether there are any stop points that have not been traversed, if yes, executing step S2-6, otherwise, executing step S2-10;

[0026] Step S2-6: taking the current stop point as the starting point of the next travel chain and returning to step S2-3;

[0027] Step S2-7: Determine whether the duration of stay at the current stay point exceeds a pre-configured first set threshold duration, if yes, execute step S2-8, otherwise, execute step S2-9;

[0028] Step S2-8: taking the current stop point as the end point of the current travel chain, and determining whether there are any stop points that have not been traversed, if yes, executing step S2-6, otherwise, executing step S2-11;

[0029] Step S2-9: Determine whether there are any untraversed stop points, if yes, execute step S2-10, otherwise, execute step S2-11;

[0030] Step S2-10: adding the current stop point to the current travel chain, and returning to step S2-3;

[0031] Step S2-11: Output the complete travel chain of all starting points and end points.

[0032] In step S2-5, after the current stop point is taken as the end point of the current travel chain, the type of the current travel chain is marked as closed;

[0033] In the step S2-8, after the current stop point is taken as the end point of the current travel link, the type of the current travel link is marked as a line type.

[0034] The step S2-1 comprises:

[0035] Step S2-1-1: Filter all trajectory points and remove outliers;

[0036] Step S2-1-2: For the remaining trajectory points, calculate the distance between them and the previous trajectory point one by one. If the distance is less than a pre-configured second set threshold distance, mark the trajectory point as a stationary point, otherwise, mark it as a moving point;

[0037] Step S2-1-3: Filter out 1 or 2 continuous moving points, calculate whether the distance between the front and rear stationary points is less than a preconfigured second set threshold distance, and the time interval is less than a preconfigured second time length, if yes, mark the moving point between the front and rear stationary points as a stationary point;

[0038] Step S2-1-4: Aggregate all stationary points, and aggregate each group of continuous stationary points to obtain a preliminary stop point;

[0039] Step S2-1-5: Calculate the average distance between the position of all static points in the preliminary stay point and the center of the stay point. If the average distance is less than the pre-configured third set threshold distance, the preliminary stay point is used as the stay point, and the position, stay start and end time, and stay duration of each stay point are generated. Otherwise, execute step S2-1-6;

[0040] Step S2-1-6: Delete the marking status of all stationary points in the initial stay points, reduce the second set threshold distance, and then return to step S2-1-2.

[0041] The travel chain characteristics include the spatiotemporal dynamic characteristics of urban distribution vehicle travel, the characteristics of urban distribution vehicle stop points, and the travel chain characteristics of urban distribution vehicles;

[0042] The spatiotemporal dynamic characteristics of vehicle travel include travel time dynamic indicators and travel space dynamic indicators. The travel time dynamic indicators include travel intensity in different time periods within a day and travel intensity in different weeks within a week. The travel space dynamic indicators include the turnaround distance of urban distribution vehicles.

[0043] The characteristics of the urban distribution vehicle stop points include regional distribution characteristic indicators, spatial aggregation characteristic indicators and stop point attribute characteristic indicators. The regional distribution characteristic indicators include the regional distribution statistics of the stop points. The spatial aggregation characteristic indicators include the average distance between adjacent stop points and the thermal distribution statistics of the stop points. The stop point attribute characteristic indicators include the spatial autocorrelation of the stop time and the distribution statistics of the POI types around the stop points.

[0044] The travel chain characteristics of urban distribution vehicles include overall characteristic indicators, time characteristic indicators, spatial characteristic indicators, speed characteristic indicators and cross-regional characteristic indicators. The overall characteristic indicators include the average number of chains per vehicle and the average number of nodes per vehicle. The time characteristic indicators include the average time of a single trip, the average stay time at a stop point and the average total time of each travel chain. The spatial characteristic indicators include the average distance of a single trip and the average total mileage of each travel chain. The speed characteristic indicators include the average speed of a single trip and the average speed of each travel chain. The cross-regional characteristic indicators include intra-city freight travel chain characteristics, intra-city-provincial travel chain characteristics, intra-city-domestic travel chain characteristics and non-city travel chain characteristics.

[0045] A road optimization method based on the above travel feature extraction method includes:

[0046] Extracting the characteristics of the trip chain;

[0047] Set parking points based on travel chain characteristics.

[0048] A device comprises a memory, a processor, and a program stored in the memory, wherein the processor implements the above method when executing the program.

[0049] A storage medium stores a program, which implements the above method when executed.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] 1. Extract features based on the travel chain, thereby improving the breadth and depth of urban distribution vehicle travel characteristics and providing more valuable planning basis.

[0052] 2. Extract travel chains based on the stop points, so that closed travel chains and linear travel chains can be extracted, and the extracted travel chains are more detailed.

[0053] 3. Based on the multiple filtering method, the recognition accuracy of the stay point can be improved.

[0054] 4. A multi-layer indicator system for the travel characteristics of urban distribution vehicles is established, which contains three dimensional layers, 10 main indicator layers, and 21 sub-indicator layers. A specific method of obtaining each indicator is given. Combined with the stop points of urban distribution vehicles, a detailed analysis is carried out in the multi-level indicator system of this application, which can intuitively reflect the travel characteristics of urban distribution vehicles, thereby providing a reference basis for urban traffic flow distribution, route planning, and the formulation of traffic policies. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 This is a schematic diagram of the travel intensity distribution of city distribution vehicles in City A at different times of the day in the embodiment;

[0056] Figure 2 Schematic diagram of the travel intensity distribution of city distribution vehicles in City A on different days of a week in the embodiment;

[0057] Figure 3 Schematic diagram of the distribution of turning distances of urban distribution vehicles in City A in the embodiment;

[0058] Figure 4 This is a schematic diagram of the geographical distribution of effective stopover points for urban distribution vehicles in City A in the embodiment;

[0059] Figure 5 This is a statistical diagram of thermal distribution of effective stop points for urban distribution vehicles in City A in the embodiment;

[0060] Figure 6 Schematic diagram of the LISA significance map of the dwelling time of urban distribution vehicles in City A in the embodiment;

[0061] Figure 7 This is a statistical table of POI frequencies near the freight business point in City A in the embodiment;

[0062] Figure 8 The frequency statistics table of POIs near potential temporary stops in City A in the embodiment;

[0063] Fig. 9 Schematic diagram of the distribution of POI categories in City A in the embodiment;

[0064] Fig.10 Schematic diagram of the overall characteristics of the urban distribution vehicle travel chain in City A in the embodiment;

[0065] Fig.11 This is a comparison diagram of the cumulative probability distribution of the two types of travel chains in City A in the embodiment;

[0066] Fig.12 This is a comparison diagram of the cumulative probability distribution of the number of nodes of two types of travel chains in City A in the embodiment;

[0067] Fig.13 The statistical table of the two types of travel chain times in City A in the embodiment;

[0068] Fig.14 It is a single trip time-cumulative probability distribution curve diagram of two types of travel chains in City A in the embodiment;

[0069] Fig.15 It is a graph of the cumulative probability of stay time at two types of travel chain stay points in City A in the embodiment;

[0070] Fig.16 It is a curve diagram of total duration-cumulative probability of each travel chain of two types of travel chains in City A in the embodiment;

[0071] Fig.17 The statistical table of the travel distances of two types of travel chains in City A in the embodiment;

[0072] Fig.18 It is a single trip distance-cumulative probability curve diagram of two types of travel chains in City A in the embodiment;

[0073] Fig.19 It is a distance-cumulative probability curve diagram of each chain of two types of travel chains in City A in the embodiment;

[0074] Fig. 20 The speed statistics table of two types of travel chains in City A in the embodiment;

[0075] Fig.21It is a single trip speed-cumulative probability curve diagram of two types of travel chains in City A in the embodiment;

[0076] Fig. 22 The average speed-cumulative probability curve of each chain of the two types of travel chains in City A in the embodiment

[0077] Fig.23 This is a statistical diagram of regional classification of closed travel chains in City A in the embodiment;

[0078] Fig.24 This is a statistical diagram of regional classification of linear travel chains in City A in the embodiment;

[0079] Fig.25 The figure is a flow chart of the main steps of the method for extracting travel features according to the present invention. DETAILED DESCRIPTION

[0080] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0081] A method for extracting travel characteristics of urban distribution vehicles based on spatiotemporal and geographic information systems, such as Fig.25 As shown, including:

[0082] Step S1: Acquire the trajectory of the urban distribution vehicle, wherein the attribute information of each track point in the trajectory of the urban distribution vehicle includes at least time and longitude and latitude, and generally the data of the Beidou positioning system configured on the truck can be used.

[0083] Step S2: construct one or more travel chains based on the trajectories of each city distribution vehicle. The process of constructing a travel chain based on the trajectory of a single city distribution vehicle includes:

[0084] Step S2-1: Perform motion detection on all track points in the track of the city distribution vehicle, extract all stop points, and generate the position, stop start and end time, and stop duration of each stop point;

[0085] Step S2-2: taking the first stop point as the starting point of the first travel chain;

[0086] Step S2-3: Select the next stop point according to the time sequence and calculate the distance with the starting point of the current travel chain;

[0087] Step S2-4: Determine whether the distance between the current stop point and the current travel chain is less than a first distance threshold. The value can be determined as needed, generally determined according to the parking lot area. For example, in this embodiment, it is set to 1 kilometer. If yes, execute step S2-5, otherwise, execute step S2-7;

[0088] Step S2-5: taking the current stop point as the end point of the current travel chain, and determining whether there are any stop points that have not been traversed, if yes, executing step S2-6, otherwise, executing step S2-10;

[0089] Step S2-6: taking the current stop point as the starting point of the next travel chain and returning to step S2-3;

[0090] Step S2-7: Determine whether the duration of stay at the current stay point exceeds a pre-configured first set threshold duration, which is 12 hours in this embodiment. If yes, execute step S2-8, otherwise, execute step S2-9;

[0091] Step S2-8: taking the current stop point as the end point of the current travel chain, and determining whether there are any stop points that have not been traversed, if yes, executing step S2-6, otherwise, executing step S2-11;

[0092] Step S2-9: Determine whether there are any untraversed stop points, if yes, execute step S2-10, otherwise, execute step S2-11;

[0093] Step S2-10: adding the current stop point to the current travel chain, and returning to step S2-3;

[0094] Step S2-11: Output the complete travel chain of all starting points and end points.

[0095] The types of travel chains include closed type and linear type. In step S2-5, after the current stop point is taken as the end point of the current travel chain, the type of the current travel chain is marked as closed type; in step S2-8, after the current stop point is taken as the end point of the current travel chain, the type of the current travel chain is marked as linear type.

[0096] In particular, in this embodiment, for the extraction process of the stay point, step S2-1 includes:

[0097] Step S2-1-1: Filter all trajectory points and remove outliers;

[0098] Step S2-1-2: For the remaining track points, calculate the distance between them and the previous track point one by one. If the distance is less than the pre-configured second set threshold distance, in this embodiment, it can be set according to the accuracy of the positioning system, for example, it can be set to about 1.5 times the accuracy of the positioning system, then the track point is marked as a stationary point, otherwise, it is marked as a moving point;

[0099] Step S2-1-3: Filter out 1 or 2 continuous moving points, calculate whether the distance between the front and rear stationary points is less than a preconfigured second set threshold distance, and the time interval is less than a preconfigured second time length, if yes, mark the moving point between the front and rear stationary points as a stationary point;

[0100] Step S2-1-4: Aggregate all stationary points, and aggregate each group of continuous stationary points to obtain a preliminary stop point;

[0101] Step S2-1-5: Calculate the average distance between the position of all static points in the preliminary stay point and the center of the stay point. If the average distance is less than the pre-configured third set threshold distance, the preliminary stay point is used as the stay point, and the position, stay start and end time, and stay duration of each stay point are generated. Otherwise, execute step S2-1-6;

[0102] Step S2-1-6: Delete the marking status of all stationary points in the initial stay points, reduce the second set threshold distance, and then return to step S2-1-2.

[0103] Step S3: Based on the constructed travel chains, corresponding travel chain features are extracted respectively.

[0104] The travel chain characteristics include the spatiotemporal dynamic characteristics of urban distribution vehicles, the characteristics of urban distribution vehicles’ stop points, and the travel chain characteristics of urban distribution vehicles;

[0105] The spatiotemporal dynamic characteristics of vehicle travel include travel time dynamic indicators and travel space dynamic indicators. The travel time dynamic indicators include the travel intensity in different time periods of a day and the travel intensity in different weeks of a week. The travel space dynamic indicators include the turnaround distance of urban distribution vehicles.

[0106] The characteristics of the stop points of urban distribution vehicles include regional distribution characteristic indicators, spatial aggregation characteristic indicators and stop point attribute characteristic indicators. The regional distribution characteristic indicators include the regional distribution statistics of the stop points, the spatial aggregation characteristic indicators include the average distance between adjacent stop points and the thermal distribution statistics of the stop points, and the stop point attribute characteristic indicators include the spatial autocorrelation of the stop time and the distribution statistics of the POI types around the stop points.

[0107] The travel chain characteristics of urban distribution vehicles include overall characteristic indicators, time characteristic indicators, spatial characteristic indicators, speed characteristic indicators and cross-regional characteristic indicators. The overall characteristic indicators include the average number of chains per vehicle and the average number of nodes per vehicle. The time characteristic indicators include the average time of a single trip, the average stay time at a stop point and the average total time of each travel chain. The spatial characteristic indicators include the average distance of a single trip and the average total mileage of each travel chain. The speed characteristic indicators include the average speed of a single trip and the average speed of each travel chain. The cross-regional characteristic indicators include the characteristics of the intra-city freight travel chain, the intra-city-province travel chain, the intra-city-domestic travel chain and the non-city travel chain, as shown in Table 1:

[0108] Table 1

[0109]

[0110] In addition, the travel features extracted above can be applied to multiple aspects. For example, the present application also provides a road optimization method based on the travel feature extraction method above, including:

[0111] Extracting the characteristics of the trip chain;

[0112] Set parking points based on travel chain characteristics.

[0113] The premise of this application for the analysis of urban distribution vehicles is that the data information of urban distribution vehicle stop points in a certain city (hereinafter referred to as City A) has been obtained. Combined with the content of the applicant's patent application for urban distribution stop points, it can be seen that City A has approximately 190,000 pieces of data information reflecting its temporary stop points for urban distribution vehicles.

[0114] This application takes the real data of urban distribution vehicles in City A as an example, and combines the three dimensional characteristics, 10 main indicator characteristics and 21 sub-indicator characteristics in Table 1 to provide a specific analysis method for the travel characteristics of urban distribution vehicles in City A.

[0115] 1. Temporal and spatial dynamic characteristics of urban distribution vehicle travel in City A:

[0116] like Figure 1 As shown in the figure, it is a schematic diagram of the travel intensity distribution of urban distribution vehicles in City A at different times of the day. Figure 1 It can be seen that the travel of urban distribution vehicles in City A has a certain regularity, which is mainly concentrated in the daytime, with fewer trips at night. The active period is from 9 am to 7 pm. The average travel distance in the morning from 2 am to 11 am gradually increases, and it declines slightly from 11 pm to 2 pm and then rebounds. The average travel distance continues to decline from 4 pm to midnight.

[0117] Figure 2 As shown in the figure, it is a schematic diagram of the travel intensity distribution of city distribution vehicles in City A on different days of the week. Figure 2It can be seen that the average travel distance of urban distribution vehicles every day in a week is relatively similar, basically maintaining between 100-120km. Only the travel intensity on Sundays is slightly lower, about 20% lower than the average travel distance in a week.

[0118] 2. Dynamic indicators of travel space:

[0119] like Figure 3 As shown in Figure 1, it is a schematic diagram of the distribution of the turning distance of the city distribution vehicles in City A. By substituting the effective stop point data information of City A into Formula 1, the turning distance data information of the city distribution vehicles in City A can be obtained. Figure 3 It can be seen that the average turning distance in City A is 14.46 km, and the turning distance with 80% cumulative probability is 17.40 km. This fully shows that the travel range of each truck is relatively concentrated, and the location of the corresponding freight activities is highly clustered. The turning distance with 95% cumulative probability is 43.27 km, indicating that a small proportion of vehicles have a large travel range and there are cross-regional freight activities.

[0120] II. Characteristics of the parking spots for urban distribution vehicles in City A:

[0121] 3. Geographical distribution characteristic indicators:

[0122] like Figure 4 As shown in the figure, the proportion of each region in the statistical effective stopping points of vehicles is circular. It can be seen from the figure that there are 1,001,515 effective stopping points in City A, accounting for 88.87% of the total number of effective stopping points. It can be concluded that the effective stopping points of urban distribution vehicles are mainly concentrated in City A.

[0123] 4. Spatial aggregation characteristic indicators:

[0124] Will Figure 4 The effective stay point data information of City A is substituted into Formula 2 to obtain the average distance between adjacent stay points in City A, as shown in Table 2, which is the nearest neighbor analysis result of the effective stay points in City A:

[0125] Table 2

[0126] Nearest neighbor analysis results Observed average distance: 39.32154193574 Expected average distance: 1085.30520974526 Nearest Neighbor Index (NNI): 0.03623086076 Points:1126968 Z-value:-1957.31034196908

[0127] As shown in Table 2, the observed data show that the average distance of effective stay points is 39.32 meters, which is significantly shorter than the expected average distance of 1085.31 meters under random distribution. This contrast is reflected by the value of the nearest neighbor index (NNI) of 0.0362. When NNI < 1 indicates a clustered distribution, NNI > 1 indicates a uniform (or discrete) distribution, and NNI = 1 indicates a random distribution; the NNI in Table 2 is much less than 1, which strongly indicates that these points are non-randomly clustered. The statistically significant Z value is -1957.31, which is a significant negative value indicating that the observed points show a significant clustering trend compared to the expectation of random distribution. These quantitative results collectively indicate that the effective stay points in the study area tend to be densely clustered with each other, rather than following a random or uniform spatial distribution pattern.

[0128] like Figure 5 As shown in the figure, the thermal distribution statistics of effective stay points are combined with formula 3, and the effective stay point data are converted into a coordinate system adapted to the QGIS software using QGIS software. The field radius of the observation area is set to 2 km. Figure 5 In the figure, the darker the color, the higher the density, and the lighter the color, the lower the density. It can be seen that the hot spots outline the contour of the urban area of ​​A, but the area with the most concentrated stay points is also concentrated in a small area within the urban area of ​​A. Although there are also stay points in several other areas, they are all dispersed and not as concentrated as the central urban area.

[0129] 5. Stay point attribute characteristic indicators:

[0130] like Figure 6 The figure shows the LISA saliency map of the dwell time of urban distribution vehicles. Combined with Formula 4, the K nearest neighbor method is used to obtain the spatial autocorrelation of the dwell time of urban distribution vehicles for the effective dwelling points in City A. The LISA (Local Indicators of Spatial Association) saliency map is a tool used to reveal the local patterns of spatial data. Different colors represent different levels of saliency. Figure 6 The significance of each level of significant elements is shown in Table 3:

[0131] Table 3

[0132]

[0133] like Figure 7 As shown, this is a statistical table of POI frequencies near freight business points in City A;

[0134] like Figure 8 As shown, this is a statistical table of POI frequency near potential temporary stops in City A;

[0135] like Fig. 9 The following is a schematic diagram showing the distribution of major POI categories in City A.

[0136] Search to get the list of the three nearest POIs near each stop. Figure 7 Freight business points and Figure 8 The POI frequency statistics of the POIs near the temporary stop are obtained. Fig. 9 The ten POI types are shown.

[0137] Depend on Fig. 9 It can be seen that the frequency of corporate POIs is much higher than other types, and higher than shopping and consumption POIs, which have the highest proportion among the major POI categories in City A. Comparing all freight stop points and potential temporary stop points, the proportion of shopping consumption in potential temporary stop points has increased significantly, but is still dominated by corporate POIs. The third most frequent POI type near all freight stop points in City A is parking lots, and after screening, long-term parking places such as parking lots are basically excluded, so they do not appear in the statistics of potential temporary stop points. In addition, the frequency distribution of most POI types in the two types of stop points is relatively similar. However, POIs of the type of transportation facilities-bus stops appear in the statistics of both types of stop points, which may reflect the common phenomenon of suspected illegal and irregular parking of urban distribution vehicles.

[0138] 3. Characteristics of the urban distribution vehicle travel chain:

[0139] The basic characteristics of the urban distribution vehicle travel chain are mainly developed from four aspects: overall characteristics, time characteristics, spatial characteristics, and speed characteristics, and the indicators of the two types of travel chains, closed travel chains and linear travel chains, are compared in detail. After the basic characteristics analysis is completed, a comparative analysis of the travel chain within the city and the cross-regional travel chain is conducted.

[0140] 6. Overall characteristic indicators:

[0141] like Fig.10 As shown in the figure, it is the overall characteristics of the urban distribution vehicle travel chain in City A. Fig.10 It can be seen that the average number of chains in the closed travel chain of each vehicle is 17.15, and the average number of chains in the linear travel chain is 11.68. It can be seen that the travel chain mode of urban distribution vehicles is mainly closed, but the proportion of linear is also not small. The average number of nodes in the closed travel chain is 5.33, and the average number of nodes in the linear travel chain is 8.84. The closed travel chain has fewer nodes and passes fewer business points. Since the journey eventually returns to the starting point, it is difficult to form a long chain from the perspective of the difficulty of closing the loop, and it is easier to form a short chain. The linear travel chain does not need to be closed, the location of the journey end is flexible, and the number of nodes is large.

[0142] like Fig.11 As shown, it is a comparison chart of the cumulative probability distribution of the two types of travel chains;

[0143] like Fig.12As shown, it is a comparison diagram of the cumulative probability distribution of the number of nodes in two types of travel chains;

[0144] Depend on Figure 11-Figure 12 It can be seen that the vehicles with the largest number of linear travel chains are only more than 30, while 20% of the vehicles have more than 26 closed travel chains, which shows that although closed travel chains are mainly short chains, they have a clear advantage in quantity over linear travel chains. Fig.12 ), after reaching a certain threshold, the cumulative probability rises sharply. Even for linear travel chains dominated by long chains, the proportion of those with more than 20 nodes is very limited, indicating that in freight activities, there is also a theoretical threshold for the number of related business points, and the travel chain cannot grow indefinitely.

[0145] 7. Time characteristic indicators:

[0146] like Fig.13 As shown in the figure, there are two types of travel chain time statistics, which reflect the comparison of the average time of a single trip between the closed travel chain and the linear travel chain, the comparison of the average time of the stop point and the average time of each travel chain. Fig.13 It can be seen that the average single trip time of the closed travel chain is 57.15 minutes, the average stay time at the stop point is 97.02 minutes, and the average travel time of each travel chain is 14.47 hours; the average single trip time of the linear travel chain is 57.21 minutes, the average stay time at the stop point is 130.39 minutes, and the average travel time of each chain is 87.48 hours.

[0147] like Fig.14 As shown, the single trip time-cumulative probability curves of two types of travel chains are shown;

[0148] like Fig.15 As shown, the stop time-cumulative probability curves of the two types of travel chain stop points;

[0149] like Fig.16 As shown, it is the total duration-cumulative probability curve of each two types of travel chains;

[0150] Depend on Fig.14 It can be seen that the average time and distribution of a single trip in the closed travel chain and the linear travel chain are very similar, indicating that the travel patterns of urban distribution vehicles between adjacent business points are similar. Fig.15 It can be seen that the dwell time of 80% cumulative probability points of the two types of travel chains is significantly lower than the average value, and the number of dwell time less than 1h of the two types of travel chains exceeds 60% and the distribution is relatively close, indicating that temporary parking is the main parking method for urban distribution vehicles, but there are a few long-term dwelling points, which significantly increase the average value of the dwell time. Fig.16As shown in the figure, in terms of the average travel time of each chain, the closed travel chain is significantly lower than the linear travel chain, but the probability distribution curves of the travel time of the two types of travel chains are not smooth, and there are multiple inflection points, indicating that the total travel time of the urban distribution vehicle travel chain has a certain randomness, and may show great differences depending on the different freight activities.

[0151] 8. Spatial characteristic indicators:

[0152] like Fig.17 As shown in the figure, there are two types of travel chain travel distance statistics, Fig.17 It can be seen that the average travel distance of the closed travel chain is 50.10km, and the average distance of a single trip is 14.70km; the average travel distance of the linear travel chain is 138.80km, and the average distance of a single trip is 20.59km. In terms of the average distance of a single trip, the two are not much different, indicating that the distance between business points is relatively close; but in terms of the average distance of each chain, the two types of travel chains are significantly different. The average travel distance of the linear travel chain is longer, which is related to the medium and long-distance transportation between cities, but there is also the contribution of freight return and search within the city, while the travel distance of the closed travel chain is shorter, which more clearly reflects the situation of freight activities within the city. Combined with the statistics of the number of nodes in the above text, it shows that most of the closed travel chains are short-distance transportation.

[0153] like Fig.18 As shown, the single trip distance-cumulative probability curves of two types of travel chains;

[0154] like Fig.19 As shown, the distance-cumulative probability curve of each chain of the two types of travel chains;

[0155] Depend on Figure 18-19 It can be seen that the probability growth of the two statistics of the closed travel chain and the linear travel chain is relatively smooth, and the upper limit of the single trip distance is relatively close, but in terms of the total trip distance, the upper limit of the linear travel chain reaches more than 600km, which is significantly higher, while the closed travel chain almost reaches the limit at 300km. If a long chain closed loop is needed, the truck needs to continuously search for goods over a long distance and never rest for a long time in a certain place. Such conditions are extremely harsh, so most long chains are linear travel chains.

[0156] 9. Speed ​​characteristic index:

[0157] like Fig. 20 As shown, there are two types of travel chain speed statistics;

[0158] like Fig.21 The above are the single trip speed-cumulative probability curves of two types of travel chains;

[0159] like Fig. 22As shown, it is the average speed-cumulative probability curve of each chain of the two types of travel chains;

[0160] Depend on Figure 20-22 As shown in Figure 2, the average speed of a single trip for both types of travel chains is about 30 km / h. Fig.21 It can also be seen that there are almost no trip chains with a single trip speed exceeding 60 km / h. Fig. 22 From this perspective, the travel intensity of urban distribution vehicles is relatively low. Considering that the average time for a single trip is less than 1 hour, long-term continuous long-distance driving is rare.

[0161] 10. Cross-regional characteristic indicators:

[0162] like Fig.23 The figure shows a statistical diagram of regional classification of closed travel chains;

[0163] like Fig.24 The figure shows a statistical diagram of regional classification of linear travel chains.

[0164] Depend on Fig.23 and 24 It can be seen that most of the two types of travel chains are freight travel chains in City A, which is in line with expectations. In other categories, the proportion of non-City A freight travel chains is relatively close. The reason is that in non-City A freight travel chains, there are also two types of travel chains: intra-city travel chains and cross-regional travel chains. In linear travel chains, the proportion of freight travel chains between City A and other cities in the province and between City A and other regions in the country is significantly higher than that of closed travel chains. This has a lot to do with the long chain nature of linear travel chains, and further verifies that closed travel chains are difficult to form in medium and long-distance transportation.

[0165] This application first constructs two types of urban distribution vehicle travel chains with urban freight distribution characteristics, and then establishes a multi-layer indicator system for urban distribution vehicle travel characteristics with three dimensional layers, 10 main indicator layers, and 21 sub-indicator layers. It can be seen that this application divides the urban distribution vehicle travel chain in City A into linear travel chains and closed travel chains, and compares and analyzes various travel chain indicators, which can well reflect the various significant characteristics of urban distribution vehicle travel.

[0166] If the above functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program code.

Claims

1. A method for extracting travel characteristics of urban distribution vehicles based on spatiotemporal and geographic information systems, characterized in that: include: Step S1: Obtain the trajectory of the city distribution vehicle; Step S2: construct one or more travel chains based on the trajectories of the city distribution vehicles; Step S3: Based on the constructed travel chains, corresponding travel chain features are extracted respectively.

2. According to the method for extracting travel characteristics of urban distribution vehicles based on time-space and geographic information system in claim 1, it is characterized in that: The attribute information of each track point in the track of the urban distribution vehicle includes at least time and longitude and latitude.

3. The method for extracting travel characteristics of urban distribution vehicles based on time-space and geographic information system according to claim 1 is characterized in that: The types of the travel chain include closed type and linear type.

4. The method for extracting travel characteristics of urban distribution vehicles based on time-space and geographic information system according to claim 3 is characterized in that: The process of constructing a travel chain based on the trajectory of a single city distribution vehicle in step S2 includes: Step S2-1: Perform motion detection on all track points in the track of the city distribution vehicle, extract all stop points, and generate the position, stop start and end time, and stop duration of each stop point; Step S2-2: taking the first stop point as the starting point of the first travel chain; Step S2-3: Select the next stop point according to the time sequence and calculate the distance with the starting point of the current travel chain; Step S2-4: Determine whether the distance between the current stop point and the current travel link is less than a first distance threshold, if yes, execute step S2-5, otherwise, execute step S2-7; Step S2-5: taking the current stop point as the end point of the current travel chain, and determining whether there are any stop points that have not been traversed, if yes, executing step S2-6, otherwise, executing step S2-10; Step S2-6: taking the current stop point as the starting point of the next travel chain and returning to step S2-3; Step S2-7: Determine whether the duration of stay at the current stay point exceeds a pre-configured first set threshold duration, if yes, execute step S2-8, otherwise, execute step S2-9; Step S2-8: taking the current stop point as the end point of the current travel chain, and determining whether there are any stop points that have not been traversed, if yes, executing step S2-6, otherwise, executing step S2-11; Step S2-9: Determine whether there are any untraversed stop points, if yes, execute step S2-10, otherwise, execute step S2-11; Step S2-10: adding the current stop point to the current travel chain, and returning to step S2-3; Step S2-11: Output the complete travel chain of all starting points and end points.

5. The method for extracting travel characteristics of urban distribution vehicles based on time-space and geographic information system according to claim 4 is characterized in that: In step S2-5, after the current stop point is taken as the end point of the current travel chain, the type of the current travel chain is marked as closed; In the step S2-8, after the current stop point is taken as the end point of the current travel link, the type of the current travel link is marked as a line type.

6. The method for extracting travel characteristics of urban distribution vehicles based on time-space and geographic information system according to claim 4 is characterized in that: The step S2-1 comprises: Step S2-1-1: Filter all trajectory points and remove outliers; Step S2-1-2: For the remaining trajectory points, calculate the distance between them and the previous trajectory point one by one. If the distance is less than a pre-configured second set threshold distance, mark the trajectory point as a stationary point, otherwise, mark it as a moving point; Step S2-1-3: Filter out 1 or 2 continuous moving points, calculate whether the distance between the front and rear stationary points is less than a preconfigured second set threshold distance, and the time interval is less than a preconfigured second time length, if yes, mark the moving point between the front and rear stationary points as a stationary point; Step S2-1-4: Aggregate all stationary points, and aggregate each group of continuous stationary points to obtain a preliminary stop point; Step S2-1-5: Calculate the average distance between the position of all static points in the preliminary stay point and the center of the stay point. If the average distance is less than the pre-configured third set threshold distance, the preliminary stay point is used as the stay point, and the position, stay start and end time, and stay duration of each stay point are generated. Otherwise, execute step S2-1-6; Step S2-1-6: Delete the marking status of all stationary points in the initial stay points, reduce the second set threshold distance, and then return to step S2-1-2.

7. The method for extracting travel characteristics of urban distribution vehicles based on time-space and geographic information system according to claim 1 is characterized in that: The travel chain characteristics include the spatiotemporal dynamic characteristics of urban distribution vehicle travel, the characteristics of urban distribution vehicle stop points, and the travel chain characteristics of urban distribution vehicles; The spatiotemporal dynamic characteristics of vehicle travel include travel time dynamic indicators and travel space dynamic indicators. The travel time dynamic indicators include travel intensity in different time periods within a day and travel intensity in different weeks within a week. The travel space dynamic indicators include the turnaround distance of urban distribution vehicles. The characteristics of the urban distribution vehicle stop points include regional distribution characteristic indicators, spatial aggregation characteristic indicators and stop point attribute characteristic indicators. The regional distribution characteristic indicators include the regional distribution statistics of the stop points. The spatial aggregation characteristic indicators include the average distance between adjacent stop points and the thermal distribution statistics of the stop points. The stop point attribute characteristic indicators include the spatial autocorrelation of the stop time and the distribution statistics of the POI types around the stop points. The travel chain characteristics of urban distribution vehicles include overall characteristic indicators, time characteristic indicators, spatial characteristic indicators, speed characteristic indicators and cross-regional characteristic indicators. The overall characteristic indicators include the average number of chains per vehicle and the average number of nodes per vehicle. The time characteristic indicators include the average time of a single trip, the average stay time at a stop point and the average total time of each travel chain. The spatial characteristic indicators include the average distance of a single trip and the average total mileage of each travel chain. The speed characteristic indicators include the average speed of a single trip and the average speed of each travel chain. The cross-regional characteristic indicators include intra-city freight travel chain characteristics, intra-city-provincial travel chain characteristics, intra-city-domestic travel chain characteristics and non-city travel chain characteristics.

8. A road optimization method based on the travel feature extraction method according to any one of claims 1 to 7, characterized in that: include: Extracting the characteristics of the trip chain; Set parking points based on travel chain characteristics.

9. A device comprising a memory, a processor, and a program stored in the memory, characterized in that: When the processor executes the program, the method according to any one of claims 1 to 8 is implemented.

10. A storage medium having a program stored thereon, characterized in that: When the program is executed, the method according to any one of claims 1 to 8 is implemented.

Images