Semantized parameter calibration method for urban cross-domain multi-mode space

Abstract

The invention discloses a semantic parameter calibration method for an urban cross-domain multi-mode space, and belongs to the technical field of traffic simulation. In order to realize semantic parameter calibration of the city cross-domain multi-mode space, the method comprises the following steps: constructing the city cross-domain multi-mode space based on BIM; performing multi-mode traffic network semantic parameter calibration, including self-driving traffic network characteristic parameter calibration, walking traffic network characteristic parameter calibration, bus traffic network characteristic parameter calibration, rail traffic network characteristic parameter calibration and hub space characteristic parameter calibration; the method comprises the following steps: performing simulation deduction on a traffic state, restoring a multi-mode traffic network operation state, performing statistics on traffic flow, travel distance, travel time and transfer frequency indexes, and comparing with actual geomagnetic data, checkpoint data, video monitoring data, vehicle GPS track data and rail bus card swiping data to complete multi-mode traffic network semantic parameter checking. The prediction capability and decision support capability of the traffic model are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of traffic simulation, and particularly relates to a method for semantic parameter calibration of urban cross-domain multi-mode space. Background Art

[0002] After spatially modeling the urban cross-domain multi-mode traffic space, how to endow traffic parameters with clear semantic information so that they can more accurately describe traffic phenomena and behaviors. Parameter semanticization can make the simulation model more accurately reflect the actual traffic situation and improve the model's prediction ability and decision-making support ability, which is a technical problem to be solved urgently. The semanticization of spatial parameters after cross-domain multi-mode collaborative modeling is in the exploratory stage, and the fusion of multi-source heterogeneous data (such as sensor data, video surveillance, BIM models) is the key.

[0003] The traffic network modeling based on GIS is to construct a two-dimensional road network topology using the GIS spatial database and realize path analysis by combining attribute data. Although OGC standards such as CityGML in the traffic network modeling based on GIS provide spatial data interoperability, they lack detailed building information. Parameter semanticization may rely on the structure of CityGML, but it is necessary to combine BIM data to supplement details. The advantage is strong scalability, but data compatibility issues need to be addressed.

[0004] Graph theory and complex network modeling. The traffic network is abstracted into a graph structure of nodes and edges to analyze network connectivity and robustness. Graph theory and complex network modeling usually use nodes and edges to represent the traffic network, and the parameters may include traffic flow, speed, etc., but the semanticization may be relatively basic and lack detailed attribute descriptions. The advantage is that the model is simple and suitable for macroscopic analysis, but it is difficult to handle microscopic details. Summary of the Invention

[0005] The problem to be solved by the present invention is to realize the semantic parameter calibration of urban cross-domain multi-mode space, and a method for semantic parameter calibration of urban cross-domain multi-mode space is proposed.

[0006] To achieve the above object, the present invention is realized through the following technical solutions: A method for semantic parameter calibration of urban cross-domain multi-mode space includes the following steps: S1. Collect geomagnetic sensor data, bayonet data, video detection data, vehicle GPS trajectory data, and rail transit card swiping data, and construct an urban cross-domain multi-mode space based on BIM; S2. Perform semantic parameter calibration on the urban cross-domain multi-mode space constructed in step S1 for multi-mode traffic network, including calibration of characteristic parameters of self-driving traffic network, calibration of characteristic parameters of pedestrian traffic network, calibration of characteristic parameters of bus traffic network, calibration of characteristic parameters of rail transit network, and calibration of characteristic parameters of hub space; S3. For the semantic parameters of the multi-modal traffic network obtained in step S2, simulate and deduce the traffic state, restore the operation state of the multi-modal traffic network, and count the indicators of traffic flow, travel distance, travel time, and transfer times. Compare them with the actual geomagnetic data, bayonet data, video monitoring data, vehicle GPS trajectory data, and rail transit card swiping data to complete the verification of the semantic parameters of the multi-modal traffic network.

[0007] Further, the specific implementation method of step S2 includes the following steps: S2.1. Calibrating the characteristic parameters of the self-driving traffic network includes calibrating the road capacity, calibrating the free-flow speed of the road section, calibrating the flow-density-speed relationship model, and calibrating the generalized travel cost of the self-driving road section; S2.2. Calibrating the characteristic parameters of the pedestrian traffic network includes calibrating the time of the pedestrian section and calibrating the generalized travel cost of the pedestrian section; S2.3. Calibrating the characteristic parameters of the bus traffic network includes calibrating the running time of the line section and calibrating the generalized travel cost of the bus section; S2.4. Calibrating the characteristic parameters of the rail transit network includes calibrating the in-out station time parameters, calibrating the in-station transfer time parameters, calibrating the running time parameters of the line section, and calibrating the generalized travel cost parameters of the rail transit network; S2.5. Calibrating the characteristic parameters of the hub space: including calibrating the walking time parameters in the hub, calibrating the elevator riding time parameters in the hub, and calibrating the generalized travel cost parameters of the hub.

[0008] Further, the specific implementation method of step S2.1 includes the following steps: S2.1.1. Road capacity calibration: Group the historical section flow data by road type and calibrate the road capacity; S2.1.2. Free-flow speed calibration of the road section: Based on the vehicle GPS data, use the hidden Markov map matching algorithm to match the GPS trajectory to the self-driving road section. Take the GPS data of the past six months, divide the time period from 6:00 to 24:00 into 5-minute time slices, statistically calculate the arithmetic mean of the average travel speed of each time slice, sort the calculated averages from large to small, and take the average of the first 24 time slices of the sorting result to calibrate the free-flow speed parameters of the road section; S2.1.3. Flow-density-speed relationship model calibration: Select the S3 model to construct the traffic flow density-speed model. The expression of the S3 model is: ; Where is the free-flow speed, is the critical density, mis the maximum flow inertia coefficient, l is the first positive parameter, k is the density, v is the speed; Then, based on the conservation law , the flow function is obtained, and the expression is: ; Among them, is the flow; Based on the continuity of the flow function with respect to density, the flow function is differentiated to obtain the expression: ; Then, based on the fact that in the steady state of traffic flow, the change of flow with respect to density approaches zero, the expression is obtained: ; By substituting the l = m / 2 relationship into the expression of the S3 model, and then setting the speed equal to the speed at the critical density, the maximum flow inertia coefficient is solved, and the calculation formula is as follows: ; Among them, is the speed at the critical density; S2.1.4. Calibration of the generalized travel cost of the self-driving road section: The various expenses incurred by cars include vehicle purchase expenses, operating expenses, insurance premiums, vehicle depreciation expenses. Vehicle purchase expenses include vehicle purchase price, purchase tax, value-added tax, vehicle and vessel use tax, license plate fee, vehicle inspection fee, driving training fee. Operating expenses include fuel power expenses, community parking fees, parking fees at point D, parking space fees, maintenance and repair expenses, tolls, road maintenance fees, annual inspection fees, insurance premiums, vehicle depreciation expenses. The calculation formula is obtained as follows: ; Among them, is the provider cost of the car; is the various expenses incurred by the car, is the annual operating kilometers of the car, q' is one of the various expenses such as purchase fee, road maintenance fee, parking fee, maintenance and repair fee, insurance premium, annual inspection fee, mc is the total of the various expenses such as purchase fee, road maintenance fee, parking fee, maintenance and repair fee, insurance premium, annual inspection fee.

[0009] Furthermore, the specific implementation method of step S2.2 includes the following steps: S2.2.1. Calibration of the walking section time: Based on the walking section length and the average walking speed Ratio calibration of walking time parameters for walking sections , where i is the walking section number; S2.2.2. Generalized travel cost calibration for walking sections, the calculation formula is as follows: ; Among them, is the generalized travel cost of the walking section, is the value time coefficient of walking travel, is the value of urban unit time, is the walking time of the walking section.

[0010] Furthermore, the specific implementation method of step S2.3 includes the following steps: S2.3.1. Line section running time calibration: Collect the running time between adjacent stations of the bus line based on the bus arrival and departure data , where i' , j' are the starting station number and the ending station number of the section respectively, k' is the line number, s is the time period, n' is the bus number, and calculate the average running time of the bus line section by line and by time period , the calculation formula is as follows: ; Among them, n is the total number of buses; S2.3.2. Generalized travel cost calibration for bus sections: Based on the urban unit time value , the average running time of the bus line section , calculate the generalized travel cost of the bus network , the calculation formula is as follows: ; Among them, is the value time coefficient of bus travel.

[0011] Furthermore, the specific implementation method of step S2.4 includes the following steps: S2.4.1. In-out station time parameter calibration: The in-out station walking time parameters include the walking in-out station time parameter and the elevator in-out station time parameter , the calculation formula is as follows: ; ; Among them, is the walking distance for entering and leaving the station by rail; is the travel distance of the elevator for entering and leaving the station by rail, is the average speed of the elevator; S2.4.2. Calibration of in-station transfer time parameters: In-station transfer walking time The calculation formula is as follows: ; where, is the in-station transfer walking distance by rail; S2.4.3. Calibration of line section operation time: Calculate the operation time between adjacent stations on the rail line based on the rail operation schedule data where i' , j' are the starting station number and the ending station number of the section respectively, k' is the line number, n '' is the rail operation schedule number, and calculate the average operation time of the rail line section by line and by time period The calculation formula is as follows: ; S2.4.4. Calibration of the generalized travel cost parameters of the rail transit network: including the walking in-and-out station cost parameter , the elevator in-and-out station cost parameter , the in-station transfer walking cost and the line section operation cost , the calculation formula is as follows: ; where, is the walking in-and-out station time, is the time value time coefficient of taking the elevator, is the elevator in-and-out station time, is the time value coefficient of in-station transfer walking by rail, is the in-station transfer walking time by rail, is the time value coefficient of the rail.

[0012] Furthermore, the specific implementation method of step S2.5 includes the following steps: S2.5.1. Calibration of the walking time parameters in the hub: The walking time in the hub has the following calculation formula: ; where, is the walking arc distance inside the hub; S2.5.2. Calibration of the elevator riding time parameters in the hub: The elevator riding time in the hub has the following calculation formula: ; Among them, is the travel distance of the elevator inside the hub; S2.5.3. Calibration of the general travel cost parameters of the hub: including the walking cost parameters inside the hub and the elevator ride cost parameters inside the hub , and the calculation formula is as follows: .

[0013] Furthermore, the specific implementation method of step S3 includes the following steps: S3.1. Load data under the simulation software: Load traffic network data, OD demand data, and flow-density-speed detector data to form a multi-mode simulation environment; S3.2. Path search: Generate a simulation path set based on the OD trajectory; S3.3. Multi-mode traffic simulation: Simulate and deduce the traffic state to restore the operation state of the multi-mode traffic network; S3.4. Statistical simulation indicators: Statistically traffic flow, travel distance, travel time, and transfer times; S3.5. Network parameter verification: Compare the simulation statistical data with the actual geomagnetic, checkpoint, video monitoring data, vehicle trajectory data, and rail transit card swiping data, and statistically calculate the relevant errors. The calculation formula is as follows: ; Among them, is the true value of section i , is the simulation value of section i output by the simulation model If the error indicator is less than 15%, it is considered that the parameters of this section are correct; otherwise, the network parameters need to be adjusted; S3.6. If the parameters are verified to be correct in step S3.5, it is considered that the parameter verification is completed; otherwise, according to the parameter comparison of the road network, rail network, and walking network, all parameters are adjusted again, especially the network parameters of the mode with larger errors, and then step S3.3 is performed again.

[0014] Advantages of the present invention: The semantic parameter calibration method of a city's cross-domain multi-modal space described in the present invention calibrates the multi-modal traffic network parameters through multi-source traffic big data, which can make the traffic model more accurately reflect the traffic conditions in the real world. The fusion of geomagnetic, checkpoint, video traffic detection data, vehicle GPS trajectory data, IC card swiping data and other multi-source data can fully capture the dynamic characteristics and spatiotemporal distribution of traffic flow, effectively improve the model's prediction ability and decision support ability, and provide a more scientific and accurate basis for cross-domain urban governance.

[0015] The semantic parameter calibration method of a city's cross-domain multimodal space described in the present invention verifies the spatial topological structure and computable spatial parameters of a multimodal computable spatial network based on depth-first search, map matching and traffic simulation algorithms, which can ensure the connectivity and rationality of the network, avoid problems such as incoherence and irrationality of the traffic network caused by data errors or model defects, and further enhance the prediction ability and decision support ability of the traffic model. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a flow chart of a semantic parameter calibration method for a city cross-domain multi-modal space according to the present invention; Figure 2 This is a structural block diagram of a semantic parameter calibration method for an urban cross-domain multi-modal space described in the present invention. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the present invention more clear, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. 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, that is, the specific embodiments described are only part of the embodiments of the present invention, rather than all of the specific embodiments. The components of the specific embodiments of the present invention described and shown in the drawings herein can be arranged and designed in various different configurations, and the present invention can also have other embodiments.

[0018] Therefore, the following detailed description of the specific embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents the selected specific embodiments of the present invention. Based on the specific embodiments of the present invention, all other specific embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present invention.

[0019] In order to further understand the content, features and effects of the present invention, the following specific implementation methods are given as examples, and the attached Figure 1 and attached Figure 2 The detailed instructions are as follows:

[0020] Example 1: A semantic parameter calibration method for urban cross - domain multi - mode space, comprising the following steps: S1. Collect geomagnetic sensor data, bayonet data, video detection data, vehicle GPS trajectory data, and rail transit card - swiping data, and construct an urban cross - domain multi - mode space based on BIM; S2. Perform semantic parameter calibration on the urban cross - domain multi - mode space constructed in step S1, including calibration of self - driving traffic network characteristic parameters, walking traffic network characteristic parameters, bus traffic network characteristic parameters, rail transit network characteristic parameters, and hub space characteristic parameters; Furthermore, the specific implementation method of step S2 includes the following steps: S2.1. Calibration of self - driving traffic network characteristic parameters includes calibration of road capacity, calibration of free - flow speed of road segments, calibration of flow - density - speed relationship model, and calibration of generalized travel cost of self - driving road segments; Furthermore, the specific implementation method of step S2.1 includes the following steps: S2.1.1. Calibration of road capacity: Group historical road segment flow data by road type and calibrate road capacity; S2.1.2. Calibration of free - flow speed of road segments: Based on vehicle GPS data, use the hidden Markov map - matching algorithm to match GPS trajectories to self - driving road segments. Take GPS data for the past six months, divide the period from 6:00 to 24:00 into 5 - minute time slices, statistically calculate the arithmetic mean of the average travel speed for each time slice, sort the calculated averages from largest to smallest, and take the average of the first 24 time slices in the sorted result to calibrate the free - flow speed parameter of the road segment; S2.1.3. Calibration of flow - density - speed relationship model: Select the S3 model to construct a traffic flow density - speed model. The expression of the S3 model is: ; Where, is the free - flow speed, is the critical density, m is the maximum flow inertia coefficient, which is the ratio of the free - flow speed to the critical density speed, l is the first positive parameter, k is the density, v is the speed; Then, based on the conservation law , obtain the flow function, and the expression is: ; Where, is the flow; Based on the continuity of the flow function with respect to relative density, differentiating the flow function gives the expression: ; The point of the highest flow on the curve corresponds to the maximum stable state of traffic flow. In the stable state, the change in flow with respect to density should approach zero, that is, the rate of change of flow is close to zero. This means that at the critical density, the derivative of the flow function should be zero.

[0021] Then, based on the fact that in the stable state of traffic flow, the change in flow with respect to density approaches zero, the expression is obtained: ; By substituting the l = m / 2 relationship into the expression of the S3 model, and then setting the speed equal to the speed at the critical density, the maximum flow inertia coefficient is solved and the calculation formula is as follows: ; where, is the speed at the critical density; S2.1.4. Calibration of the generalized travel cost of the self-driving road section: The various expenses incurred by cars include vehicle purchase expenses, operating expenses, insurance premiums, vehicle depreciation expenses. Vehicle purchase expenses include vehicle purchase price, purchase tax, value-added tax, vehicle and vessel use tax, license plate fee, vehicle inspection fee, driving training fee. Operating expenses include fuel power expenses, community parking fees, parking fees at point D, parking space fees, maintenance and repair expenses, tolls, road maintenance fees, annual inspection fees, insurance premiums, vehicle depreciation expenses. The calculation formula is obtained as follows: ; where, is the provider cost of the car; are the various expenses incurred by the car, is the annual operating kilometers of the car, q' is one of the various expenses such as purchase fee, road maintenance fee, parking fee, maintenance and repair fee, insurance premium, annual inspection fee, etc., mc is the total of the various expenses such as purchase fee, road maintenance fee, parking fee, maintenance and repair fee, insurance premium, annual inspection fee, etc.

[0022] An example of car usage cost data is shown in Table 1: Table 1 Car usage cost data

[0023] S2.2. Calibration of the characteristic parameters of the pedestrian traffic network includes calibration of the walking section time and calibration of the generalized travel cost of the walking section; Further, the specific implementation method of step S2.2 includes the following steps: S2.2.1. Calibration of walking section time: Based on the length of the walking section and the average walking speed to calibrate the walking time parameter of the walking section , where i is the walking section number; S2.2.2. Calibration of the generalized travel cost of the walking section. The calculation formula is as follows: ; Among them, is the generalized travel cost of the walking section, is the time value coefficient of walking travel, with a value of 2, is the value of the city's unit time, is the walking time of the walking section; S2.3. Calibration of the characteristic parameters of the bus transportation network includes calibration of the running time of the line section and calibration of the generalized travel cost of the bus section; Further, the specific implementation method of step S2.3 includes the following steps: S2.3.1. Calibration of the running time of the line section: Collect the running time between adjacent stations of the bus line based on the data of the bus arriving and departing stations , where i' , j' are the numbers of the starting and ending stations of the section respectively, k' is the line number, s is the time period to which it belongs, n' is the bus number, and calculate the average running time of the bus line section by line and by time period . The calculation formula is as follows: ; Among them, n is the total number of buses; 1 is the morning peak, 2 is the evening peak, and 3 is the flat peak in the time period to which it belongs; S2.3.2. Calibration of the generalized travel cost of the bus section: Based on the value of the city's unit time , the average running time of the bus line section , calculate the generalized travel cost of the bus network . The calculation formula is as follows:

[0024] Among them, is the time value coefficient of bus travel, with a value of 1; S2.4. The calibration of rail transit network characteristic parameters includes the calibration of in-out station time parameters, in-station transfer time parameters, line section running time parameters, and the calibration of the generalized travel cost parameters of the rail transit network; Furthermore, the specific implementation method of step S2.4 includes the following steps: S2.4.1. Calibration of in-out station time parameters: The in-out station walking time parameters include the walking in-out station time parameters and the elevator in-out station time parameters , and the calculation formulas are as follows: ; ; Among them, is the walking distance of in-out station of the track; is the traveling distance of the elevator for in-out station of the track, is the average speed of the elevator; S2.4.2. Calibration of in-station transfer time parameters: The calculation formula of the in-station transfer walking time is as follows: ; Among them, is the walking distance of in-station transfer of the track; S2.4.3. Calibration of line section running time: Calculate the running time between adjacent stations of the track line based on the track operation shift data , where i' , j' are the starting station number and the ending station number of the section respectively, k' is the line number, n '' is the track operation shift number, and calculate the average running time of the track line section by line and by time period , and the calculation formula is as follows: ; S2.4.4. Calibration of the generalized travel cost parameters of the rail transit network: It includes the walking in-out station cost parameter , the elevator in-out station cost parameter , the in-station transfer walking cost and the line section running cost , and the calculation formula is as follows: ; Among them, is the walking in-out station time, is the time value time coefficient of taking the elevator, with a value of 1.5, is the elevator in-out station time, is the time value coefficient of in-station walking transfer, with a value of 2.5, is the in-station walking transfer time, is the rail time value coefficient, with a value of 1; S2.5. Calibration of hub space characteristic parameters: including calibration of in-hub walking time parameters, calibration of in-hub elevator riding time parameters, and calibration of hub generalized travel cost parameters.

[0025] Furthermore, the specific implementation method of step S2.5 includes the following steps: S2.5.1. Calibration of in-hub walking time parameters: The in-hub walking time is calculated as follows: ; where, is the in-hub walking arc distance; S2.5.2. Calibration of in-hub elevator riding time parameters: The in-hub elevator riding time is calculated as follows: ; where, is the in-hub elevator travel distance; S2.5.3. Calibration of hub generalized travel cost parameters: including in-hub walking cost parameter and in-hub elevator riding cost parameter , and the calculation formula is as follows: .

[0026] S3. For the multi-modal traffic network semantic parameters obtained in step S2, conduct simulation and deduction of traffic states, restore the operation state of the multi-modal traffic network, and count traffic flow, travel distance, travel time, and transfer times indicators, and compare them with actual geomagnetic data, checkpoint data, video monitoring data, vehicle GPS trajectory data, and rail transit card swiping data to complete the verification of multi-modal traffic network semantic parameters.

[0027] Furthermore, the specific implementation method of step S3 includes the following steps: S3.1. Load data under the simulation software: Load traffic network data, OD demand data, and flow-density-speed detector data to form a multi-modal simulation environment; S3.2. Path search: Generate a simulation path set based on OD trajectories; S3.3. Multi-modal traffic simulation: Conduct simulation and deduction of traffic states to restore the operation state of the multi-modal traffic network; S3.4. Statistical simulation indicators: Count traffic flow, travel distance, travel time, and transfer times; S3.5. Network parameter verification: Compare the simulation statistical data with the actual geomagnetic, bayonet, video monitoring data, vehicle trajectory data, and rail transit card swiping data, and calculate the relevant errors. The calculation formula is as follows: ; wherein, is the true value of section i , is the simulation value of section i output by the simulation model; If the error index is less than 15%, it is considered that the parameters of this section are correct; otherwise, the network parameters need to be adjusted. S3.6. If the parameters are verified to be correct in step S3.5, it is considered that the parameter verification is completed; otherwise, according to the parameter comparison of the road network, rail network, and pedestrian network, all parameters are readjusted, especially the network parameters of the mode with larger errors, and then step S3.3 is performed again.

[0028] It should be noted that relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including a..." does not exclude the existence of additional identical elements in the process, method, article or device including the element.

[0029] Although the present application has been described above with reference to specific embodiments, various improvements can be made to it and components therein can be replaced with equivalents without departing from the scope of the present application. In particular, as long as there is no structural conflict, the various features in the specific embodiments disclosed in the present application can be combined with each other in any way, and the cases of these combinations are not exhaustively described in this specification only for the sake of saving space and resources. Therefore, the present application is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A semantic parameter calibration method for urban cross-domain multi-modal space, characterized in that: The steps include: S1. Collect geomagnetic sensor data, checkpoint data, video detection data, vehicle GPS trajectory data, and rail bus card swiping data to build a cross-domain multi-modal space in the city based on BIM; S2. Calibrate the semantic parameters of the multimodal transportation network for the urban cross-domain multimodal space constructed in step S1, including the calibration of characteristic parameters of self-driving transportation network, characteristic parameters of pedestrian transportation network, characteristic parameters of bus transportation network, characteristic parameters of rail transportation network, and characteristic parameters of hub space; S3. For the semantic parameters of the multimodal transportation network obtained in step S2, simulate the traffic status, restore the operating status of the multimodal transportation network, count the traffic flow, travel distance, travel time, and number of transfers, and compare them with the actual geomagnetic data, checkpoint data, video monitoring data, vehicle GPS trajectory data, and rail bus card swiping data to complete the verification of the semantic parameters of the multimodal transportation network.

2. According to claim 1, the semantic parameter calibration method of urban cross-domain multi-modal space is characterized in that: The specific implementation method of step S2 includes the following steps: S2.

1. Calibration of characteristic parameters of self-driving traffic network includes calibration of road capacity, calibration of free flow speed of road sections, calibration of flow-density-speed relationship model, and calibration of generalized travel cost of self-driving road sections; S2.

2. The calibration of pedestrian traffic network characteristic parameters includes the calibration of walking section time and the calibration of generalized travel cost of walking section; S2.

3. The calibration of bus network characteristic parameters includes the calibration of line section operation time and the calibration of bus section generalized travel cost; S2.

4. The calibration of rail transit network characteristic parameters includes the calibration of station entry and exit time parameters, station transfer time parameters, line section operation time parameters, and rail transit network generalized travel cost parameters; S2.

5. Calibration of hub spatial characteristic parameters: including calibration of walking time parameters within the hub, calibration of elevator riding time parameters within the hub, and calibration of hub generalized travel cost parameters.

3. According to claim 2, a semantic parameter calibration method for urban cross-domain multi-modal space is characterized in that: The specific implementation method of step S2.1 includes the following steps: S2.1.

1. Road capacity calibration: group historical road traffic data by road type and calibrate road capacity; S2.1.

2. Calibration of free-flow speed of road sections: Based on vehicle GPS data, use the hidden Markov map matching algorithm to match the GPS trajectory to the self-driving road section, take the GPS data of the past six months, divide the period from 6:00 to 24:00 into 5-minute time slices, calculate the arithmetic mean of the average travel speed of each time slice, sort the calculated average values ​​from large to small, take the average of the first 24 time slices of the sorting results, and calibrate the free-flow speed parameters of the road section; S2.1.

3. Flow-density-velocity relationship model calibration: Select the S3 model to build a traffic flow density model. The expression of the S3 model is: ； in, is the free flow speed, is the critical density, m is the maximum flow inertia coefficient, l is the first positive parameter, k is the density, v It’s speed; Then based on the conservation law , the flow function is obtained, the expression is: ； in, It is the flow rate; Based on the fact that the relative density of the flow function is continuous, the flow function is differentiated to obtain the expression: ； Then, based on the fact that the traffic flow is in a stable state, the change of flow to density approaches zero, and the expression is obtained as follows: ； By l = m / 2 is substituted into the expression of the S3 model, and then the speed is set equal to the speed at the critical density, and the maximum flow inertia coefficient is obtained by solving the formula as follows: ； in, is the velocity at the critical density; S2.1.

4. Calibration of generalized travel cost for self-driving road sections: The various expenses of a car include vehicle purchase cost, operating cost, insurance premium, and vehicle depreciation. Vehicle purchase cost includes vehicle purchase price, purchase tax, value-added tax, vehicle and vessel use tax, license fee, vehicle inspection fee, and driving training fee. Operating cost includes fuel power cost, community parking fee, D point parking fee, parking space fee, repair and maintenance cost, toll, road maintenance fee, annual inspection fee, insurance premium, and vehicle depreciation fee. The calculation formula is as follows: ； in, is the provider cost of the car; It is the various expenses of car expenditure. is the annual mileage of the car, q' It is one of the various expenses including purchase fee, road maintenance fee, parking fee, repair and maintenance fee, insurance premium, annual inspection fee, etc. mc It is the total amount of various expenses such as purchase fee, road maintenance fee, parking fee, repair and maintenance fee, insurance premium, annual inspection fee, etc.

4. According to claim 3, the semantic parameter calibration method of urban cross-domain multi-modal space is characterized in that: The specific implementation method of step S2.2 includes the following steps: S2.2.

1. Walking segment time calibration: based on walking segment length and average walking speed The walking time parameters of walking sections are calibrated by the ratio of ,in i Number the walking sections; S2.2.

2. The generalized travel cost of walking sections is calibrated using the following calculation formula: ； in, is the generalized travel cost of the walking section, is the walking trip value time coefficient, is the city unit time value, It is the walking time of the walking section.

5. According to claim 4, a semantic parameter calibration method for urban cross-domain multi-modal space is characterized in that: The specific implementation method of step S2.3 includes the following steps: S2.3.

1. Line section operation time calibration: collect bus arrival and departure data to calculate the operation time between adjacent stops on the bus line ,in i' , j' They are the starting station number and the ending station number of the section respectively. k' is the line number, s is the time period to which it belongs, n' Number the buses and calculate the average running time of the bus route sections by route and time period , the calculation formula is as follows: ； in, n is the total number of buses; S2.3.

2. Calibration of generalized travel cost of bus sections: based on the urban unit time value , average running time of bus line sections , calculate the generalized travel cost of the bus network , the calculation formula is as follows: ； in, is the time coefficient of bus travel value.

6. The semantic parameter calibration method of a city cross-domain multi-modal space according to claim 5 is characterized in that: The specific implementation method of step S2.4 includes the following steps: S2.4.

1. Station entry and exit time parameter calibration: Station entry and exit walking time parameters include walking entry and exit time parameters and elevator entry and exit time parameters , the calculation formula is as follows: ； ； in, Walking distance for track entry and exit; is the distance traveled by the elevator entering and leaving the station. is the average speed of the elevator; S2.4.

2. Calibration of transfer time parameters within the station: Walking time for transfer within the station The calculation formula is as follows: ； in, The walking distance for transfer within the rail station; S2.4.

3. Line section operation time calibration: Calculate the operation time between adjacent stations on the track based on the track operation schedule data ,in i' , j' They are the starting station number and the ending station number of the section respectively. k' is the line number, n '' is the track operation shift number, and the average running time of the track section is calculated by line and time period , the calculation formula is as follows: ； S2.4.

4. Calibration of generalized travel cost parameters for rail transit network: including walking entry and exit cost parameters , Elevator entry and exit cost parameters , Walking cost of transferring within the station and line section operating costs , the calculation formula is as follows: ； in, It is the time for walking in and out of the station. is the time coefficient of the time value of taking the elevator, It is the time when the elevator enters and leaves the station. is the value coefficient of the transfer time within the rail station, is the transfer time within the rail station. is the orbital time value coefficient.

7. The semantic parameter calibration method of a city cross-domain multi-modal space according to claim 6 is characterized in that: The specific implementation method of step S2.5 includes the following steps: S2.5.

1. Parameter calibration of walking time within the hub: Walking time within the hub The calculation formula is: ； in, is the walking arc distance inside the hub; S2.5.

2. Parameter calibration of elevator ride time in the hub: Elevator ride time in the hub The calculation formula is: ； in, is the distance travelled by the elevator inside the hub; S2.5.

3. Calibration of New York's generalized travel cost parameters: including walking cost parameters within the hub and the cost parameters of taking the elevator within the hub , the calculation formula is as follows: 。 8. The semantic parameter calibration method of a city cross-domain multi-modal space according to claim 7 is characterized in that: The specific implementation method of step S3 includes the following steps: S3.

1. Load data in the simulation software: load traffic network data, OD demand data, and flow-density-speed detector data to form a multi-mode simulation environment; S3.

2. Path search: Generate simulation path set based on OD trajectory; S3.

3. Multimodal traffic simulation: simulate the traffic status and restore the operation status of the multimodal traffic network; S3.

4. Simulation indicator statistics: statistics on traffic flow, travel distance, travel time, and number of transfers; S3.

5. Network parameter verification: Compare the simulation statistical data with the actual geomagnetic, checkpoint, video monitoring data, vehicle trajectory data and rail bus card swiping data, and calculate the relevant errors. The calculation formula is as follows: ； in, For road section i The true value of Road segments output for simulation models i The simulation value of If the error If the index is less than 15%, the parameters of the road section are considered correct, otherwise the network parameters need to be adjusted; S3.

6. If the parameters verified in step S3.5 are correct, the parameter verification is considered complete; otherwise, based on the comparison of the parameters of the road network, rail network, and pedestrian network, all parameters are readjusted, especially the network parameters of this mode with large adjustment errors, and then step S3.3 is repeated.

Images