Flying bus parking apron site selection optimization method and device

Abstract

The invention provides a flight bus parking apron site selection optimization method and device, and the method comprises the steps: carrying out the rasterization of a target region based on a traffic flow network, and obtaining the user behavior data of the target region; calculating a risk value based on each grid of the target area to obtain a risk map of the target area; based on the total cost, the user behavior data and the risk map, obtaining a parking apron site selection optimization scheme of the flight bus through a bilevel planning model; wherein an upper-layer model of the bilevel planning model is a multi-target discrete site selection model and is used for determining a parking apron site selection set of the flying bus; and the lower model of the bilevel planning model is a traffic flow distribution model and is used for distributing traffic flow. Therefore, the construction and operation cost of the flight bus is effectively reduced, the risk of a flight path is reduced, the safety and the operation efficiency of a low-altitude three-dimensional traffic system are improved, and the method is suitable for infrastructure planning in a future urban low-altitude traffic system.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic layout, and in particular to a method and device for optimizing the site selection of a flying bus apron. Background Art

[0002] As urbanization intensifies, the load on the transportation system is also increasing, leading to increased traffic congestion, environmental pollution and reduced travel efficiency. Currently, the concept of flying cars has become a striking and potentially revolutionary technology in the field of transportation. Flying cars combine the advantages of traditional cars and airplanes, bringing revolutionary possibilities to urban transportation, reducing congestion and opening up new possibilities for personal transportation.

[0003] In the existing technology, the site selection of aircraft aprons usually only focuses on construction costs and distances, while ignoring important factors such as flight path risks, ground transportation connections, and dynamic changes in user travel behavior. These shortcomings lead to many limitations in the existing apron layout solutions when dealing with complex urban traffic environments. Summary of the invention

[0004] The present invention provides a method and device for optimizing the site selection of a flying bus apron, which are used to solve the problem that the apron layout scheme in the prior art only focuses on construction cost and distance, and has many limitations when dealing with complex urban traffic environments. The method can effectively reduce the construction and operation costs of flying buses, reduce the risks of flight paths, and improve the safety and operation efficiency of low-altitude three-dimensional transportation systems. The method is suitable for infrastructure planning in future urban low-altitude transportation systems.

[0005] The present invention provides a method for optimizing the site selection of a flying bus apron, comprising the following steps: Based on the traffic flow network, the target area is rasterized and user behavior data of the target area is obtained; Calculate the risk value based on each grid of the target area to obtain a risk map of the target area; Based on the total cost, the user behavior data and the risk map, an optimization plan for the apron location of the flying bus is obtained through a double-layer planning model; Among them, the upper model of the two-layer programming model is a multi-objective discrete site selection model, which is used to determine the apron site selection set of the flying bus; the lower model of the two-layer programming model is a traffic flow allocation model, which is used to allocate traffic flow.

[0006] According to a method for optimizing the location of a flying bus apron provided by the present invention, a traffic planning scheme for a flying bus is obtained through a double-layer planning model based on the total cost, the user behavior data and the risk map, specifically comprising: In the upper model, based on the total cost and the risk map, a multi-objective optimization algorithm is used to determine the apron site set; in the lower model, based on the apron site set and the user behavior data, a traffic flow result corresponding to the apron site set is determined through a traffic flow allocation algorithm; When the termination condition is met, the current apron site selection set is used as the apron site selection optimization plan and output; when the termination condition is not met, the apron site selection set is updated based on the traffic flow results corresponding to the apron site selection set, and it is continuously iterated until the termination condition is met, and the updated apron site selection set is used as the apron site selection optimization plan and output.

[0007] According to a method for optimizing the location of a flying bus apron provided by the present invention, based on the total cost and the risk map, a multi-objective optimization algorithm is used to determine the apron location set, which specifically includes: Based on the first objective function and the first constraint condition, a set of apron locations is determined by a multi-objective optimization algorithm; The first objective function is: In the formula, and is the weight coefficient; For flight segment operating costs; The construction cost of a single apron; is the transportation cost per unit distance on the ground; is the air transportation cost per unit distance; is the length of the ground section; is the length of the aerial segment; is the flow rate of the surface road section; is the flow rate of the aerial section; The first constraint condition includes: Restriction on the number of parking spaces; Maximum ground travel distance limits; At most one helipad can be built at a single location; The correlation conditions between traffic flow and apron construction.

[0008] According to a method for optimizing the location of a flying bus apron, based on the apron location set and the user behavior data, a traffic flow result corresponding to the apron location set is determined by a traffic flow allocation algorithm, specifically comprising: Based on the second objective function and the second constraint condition, a traffic flow result corresponding to the apron site selection set is determined by a traffic flow allocation algorithm; The second objective function is: In the formula, represents the total transportation travel cost; Indicates road segment The unit flow traffic cost function; Indicates road segment Total traffic volume on The entropy term representing the path selection; Represents user behavior data Path Traffic on represents the path selection entropy weight, The larger it is, the more the selection behavior is concentrated on the lowest cost path; The second constraint condition includes: Flow conservation; Flight segment traffic restrictions; User behavior restrictions determined based on the user behavior data.

[0009] According to a method for optimizing the location of a flying bus apron provided by the present invention, a target area is rasterized based on a traffic flow network, and user behavior data of the target area is obtained, which specifically includes: Based on the traffic flow network, the target area is divided into multiple square grids, and the grid position is represented by the grid centroid coordinates; Based on the grid centroid coordinates, the user's travel start and end points and their corresponding travel demand data are mapped to each grid to obtain the user behavior data of the target area.

[0010] According to a flying bus apron site selection optimization method provided by the present invention, the upper model of the double-layer programming model is solved by calling gurobi by python; the lower model of the double-layer programming model is solved by the Frank-Wolfe algorithm; the double-layer programming model is iteratively solved between the upper model and the lower model by a taboo search algorithm.

[0011] The present invention also provides a flying bus apron site selection optimization device, comprising the following modules: A processing module, used for rasterizing a target area based on a traffic flow network and obtaining user behavior data of the target area; A calculation module, used for calculating the risk value based on each grid of the target area to obtain a risk map of the target area; A planning module, for obtaining an optimal site selection scheme for the apron of the flying bus through a double-layer planning model based on the total cost, the user behavior data and the risk map; Among them, the upper model of the two-layer programming model is a multi-objective discrete site selection model, which is used to determine the apron site selection set of the flying bus; the lower model of the two-layer programming model is a traffic flow allocation model, which is used to allocate traffic flow.

[0012] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for optimizing the location of a flying bus apron as described above is implemented.

[0013] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-described methods for optimizing the location of a flying bus apron.

[0014] The present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements any of the above-mentioned methods for optimizing the location of a flying bus apron.

[0015] The method and device for optimizing the site selection of the apron for a flying bus provided by the present invention rasterize the target area, obtain the user behavior data of the target area, then calculate the risk value according to each grid of the target area, obtain the risk map of the target area, and finally determine the apron site selection set of the flying bus according to the total cost, the user behavior data and the risk map through the upper model of the double-layer programming model, allocate the traffic flow through the lower model of the double-layer programming model, iteratively solve the upper and lower models of the double-layer programming model, and finally obtain the apron site selection optimization plan for the flying bus. The method can effectively reduce the construction and operation costs of the flying bus, reduce the risks of the flight path, and improve the safety and operation efficiency of the low-altitude three-dimensional transportation system, and is suitable for the infrastructure planning in the future urban low-altitude transportation system. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 A schematic diagram of the flow chart of the method for optimizing the site selection of the flying bus apron provided by the present invention.

[0018] Figure 2 A schematic diagram of a multi-layer transportation network provided by the present invention.

[0019] Figure 3This is a schematic diagram of the Sioux Falls network provided by the present invention.

[0020] Figure 4 This is a schematic diagram of the network topology structure and node and section numbering provided by the present invention.

[0021] Figure 5 This is a schematic diagram of the regional gridding provided by the present invention.

[0022] Figure 6 This is a schematic diagram of the Sioux Falls network area risk map provided by the present invention.

[0023] Figure 7 A schematic diagram of flight paths considering travel risks in the Sioux Falls network area provided by the present invention.

[0024] Figure 8 This is a schematic diagram of the algorithm design process provided by the present invention.

[0025] Fig. 9 This is a schematic diagram comparing the site selection results of different models in the Sioux Falls network area provided by the present invention.

[0026] Fig.10 A schematic diagram of the structure of the flying bus apron site selection optimization device provided by the present invention.

[0027] Fig.11 This is a schematic structural diagram of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] Figure 1 The schematic diagram of the process of the method for optimizing the site selection of the flying bus apron provided by the present invention is as follows: Figure 1 As shown, the method comprises the following steps: Step 100: rasterize the target area based on the traffic flow network and obtain user behavior data of the target area.

[0030] Specifically, in an embodiment of the present invention, a traffic flow network topology map of the target area can be first obtained, such as the Sioux Falls network, which includes multiple nodes (resident travel demand points) and road sections (roads connecting nodes), wherein each node is both a resident travel demand point and an alternative point for the flying bus apron.

[0031] After obtaining the traffic flow network topology map of the target area, the target area can be rasterized and the user behavior data of the target area can be obtained.

[0032] User behavior data includes user travel origin-destination (OD) data, travel time, travel distance and other data.

[0033] According to a method for optimizing the location of a flying bus apron provided by the present invention, a target area is rasterized based on a traffic flow network, and user behavior data of the target area is obtained, which specifically includes: Based on the traffic flow network, the target area is divided into multiple square grids, and the grid position is represented by the grid centroid coordinates; Based on the grid centroid coordinates, the user's travel start and end points and their corresponding travel demand data are mapped to each grid to obtain the user behavior data of the target area.

[0034] Specifically, after obtaining the traffic flow network topology of the target area, the target area can be divided into several square grids for discretizing the spatial area. The grid size depends on the analysis requirements (such as 10000 × 10000).

[0035] The grid centroid coordinates of each grid can be used as the position of the grid, and the user's travel starting and ending points and their corresponding travel demand data can be mapped to each grid to determine the grid number to which the user's travel starting and ending points belong. Then, the user's travel time, travel distance and other data can be analyzed to obtain the user behavior data of the target area.

[0036] Optionally, after obtaining the user behavior data, the user behavior data can be stored as a csv file to provide data support for the subsequent cost calculation of ground travel and air travel.

[0037] It is understandable that according to the actual land use situation, there are several alternative points for the apron location in each grid area, and the apron location is selected and laid out based on these alternative points. The construction of the flying bus apron station at the given apron alternative point requires full coverage of all demand points, considering each demand point. can be matched with the corresponding parking area Service, one demand point can be served by multiple aprons.

[0038] Step 101: Calculate risk values ​​based on each grid in the target area to obtain a risk map of the target area.

[0039] Specifically, a risk map can be drawn based on each grid in the target area, combined with the ground population, electromagnetic environment, obstacle environment, etc., to study the impact of risk factors in flying bus operations on the operating path and the final site selection results.

[0040] The risks of the entire process of flying buses operating in low-altitude urban airspace include the risks of flying buses themselves, the risks to third parties on the ground when flying buses have accidents, and the potential risks of flying buses to the public below the airspace. Therefore, in order to comprehensively and scientifically measure all potential risks of logistics drone operations, the route confidence index, the measurement standards of each index and its weight distribution method in the "Specifications for the Planning of Logistics Routes for Light and Small Unmanned Aerial Vehicles in Urban Scenarios" issued by the Civil Aviation Administration of China in 2022 can be used as risk measurement standards for logistics drone path planning. Considering risk factors such as electromagnetic radiation and buildings, the comprehensive operation risk of each grid is calculated, and a risk map of the flight area is drawn.

[0041] It is understandable that no-fly zones need to be avoided when planning the routes of flying buses. The risk map can clearly show how to determine no-fly zones. For example, when the risk value of a grid exceeds 1, the grid is considered a no-fly zone. It is necessary to stipulate that the operating path between the aprons of flying cars must not pass through the no-fly zone, and the flight path planning is carried out on this basis.

[0042] Step 102: Based on the total cost, user behavior data and risk map, an optimal site selection scheme for the apron of the flying bus is obtained through a double-layer planning model.

[0043] Among them, the upper model of the two-level planning model is a multi-objective discrete site selection model, which is used to determine the apron site selection set for the flying bus; the lower model of the two-level planning model is a traffic flow allocation model, which is used to allocate traffic flow.

[0044] Specifically, Figure 2 A schematic diagram of a multi-layer transportation network provided by the present invention, such as Figure 2 As shown, in the embodiment of the present invention, considering the situation of the multi-layer transportation network, it is necessary to reasonably allocate the flow of ground transportation and air transportation, optimize the operating efficiency of multi-modal transportation, and reduce the total travel cost.

[0045] Therefore, in the embodiment of the present invention, a two-layer planning model is designed, and an upper-layer planning model and a lower-layer planning model need to be constructed separately.

[0046] First, an upper-level planning model is constructed. The upper-level model is a multi-objective flying bus apron location selection model. It is necessary to comprehensively consider the construction cost, operating cost and flying bus path risk of the flying bus apron to realize the apron site selection.

[0047] Then, a lower-level planning model is constructed. The lower-level model is a multimodal traffic flow allocation model under capacity constraints for the supporting operation of community-oriented unmanned minibuses and flying cars. It is necessary to optimize the coupling of ground traffic and air traffic based on user travel behavior and reasonably allocate traffic flow to minimize travel costs.

[0048] Therefore, based on the total cost, user behavior data and risk map, a two-layer planning model designed in an embodiment of the present invention can be used to iteratively solve the problem, and the upper-layer planning model and the lower-layer planning model can be collaboratively optimized to obtain an optimized apron site selection solution for the flying bus.

[0049] Optionally, the upper model of the two-layer programming model is solved by calling gurobi by python to optimize the apron location; the lower model of the two-layer programming model is solved by the Frank-Wolfe algorithm to allocate traffic flow; the two-layer programming model is iteratively solved between the upper model and the lower model by the taboo search algorithm to perform global optimality optimization, thereby obtaining the apron location optimization plan for the flying bus.

[0050] The method for optimizing the site selection of the apron for a flying bus provided by the present invention rasterizes the target area, obtains the user behavior data of the target area, then calculates the risk value according to each grid of the target area, obtains the risk map of the target area, and finally determines the apron site selection set of the flying bus according to the total cost, the user behavior data and the risk map through the upper model of the double-layer programming model, allocates the traffic flow through the lower model of the double-layer programming model, iterates and solves the upper and lower models of the double-layer programming model, and finally obtains the apron site selection optimization scheme for the flying bus. The method can effectively reduce the construction and operation costs of the flying bus, reduce the risks of the flight path, and improve the safety and operation efficiency of the low-altitude three-dimensional transportation system, and is suitable for the infrastructure planning in the future urban low-altitude transportation system.

[0051] According to a method for optimizing the location of a flying bus apron provided by the present invention, a traffic planning scheme for a flying bus is obtained through a double-layer planning model based on total cost, user behavior data and risk map, specifically including: In the upper model, based on the total cost and risk map, a multi-objective optimization algorithm is used to determine the apron site set; in the lower model, based on the apron site set and user behavior data, a traffic flow allocation algorithm is used to determine the traffic flow results corresponding to the apron site set; When the termination condition is met, the current apron site selection set is used as the apron site selection optimization plan and output; when the termination condition is not met, the apron site selection set is updated based on the traffic flow results corresponding to the apron site selection set, and it is continuously iterated until the termination condition is met, and the updated apron site selection set is used as the apron site selection optimization plan and output.

[0052] Specifically, the two-layer planning model needs to input risk maps, user behavior data, and candidate apron location sets (alternative points), etc. It is necessary to pre-set the parameters of the upper and lower model (cost coefficients determined based on the total cost, constraints, traffic allocation algorithm parameters, etc.), and generate the initial apron location set by random generation or by heuristic rules.

[0053] In the upper-level model, it is necessary to input the current iteration of the apron site set, risk map and total cost related parameters, use a multi-objective optimization algorithm (such as weight method, goal programming method, genetic algorithm, etc.) to solve the multi-objective problem, determine the optimized apron site set, and output the updated apron site set and the corresponding target values ​​(such as total cost, risk level, etc.).

[0054] In the lower-level model, it is necessary to input the apron site selection set, user behavior data, and traffic network structure output by the upper-level model. Based on the traffic flow allocation algorithm (such as the Frank-Wolfe algorithm), the ground and air traffic flow distribution results corresponding to the apron site selection set and the total travel cost of the current plan are calculated.

[0055] If the termination condition is met, the current apron site selection set can be output as the final optimization solution. The termination condition may be reaching a preset number of iterations, or whether the change in the objective function value of the current iteration from the result of the previous iteration is lower than a preset threshold, etc., which is not limited in the embodiment of the present invention.

[0056] If the termination condition is not met, the apron site selection set can be updated through the upper-level model according to the lower-level traffic flow distribution results and the total travel cost, and the next round of iteration process can be entered to further optimize the coverage and flow distribution.

[0057] According to a method for optimizing the location of a flying bus apron provided by the present invention, a multi-objective optimization algorithm is used based on a total cost and risk map to determine the apron location set, specifically including: Based on the first objective function and the first constraint condition, a set of apron locations is determined by a multi-objective optimization algorithm; The first objective function is: In the formula, and is the weight coefficient; For flight segment operating costs; The construction cost of a single apron; is the transportation cost per unit distance on the ground; is the air transportation cost per unit distance; is the length of the ground section; is the length of the aerial segment; is the flow rate of the surface road section; is the flow rate of the aerial section; The first constraints include: Restriction on the number of parking spaces; Maximum ground travel distance limits; At most one helipad can be built at a single location; The correlation conditions between traffic flow and apron construction.

[0058] Specifically, in the upper-level model, the goal is to optimize the apron construction and flight route opening decisions and minimize the total cost (including construction cost, operating cost and transportation cost) while meeting the urban traffic demand and flying bus service requirements.

[0059] The cost of apron construction, rationality of construction land and operation management of flying buses can be considered to limit the number of aprons in the city to a reasonable range and select the locations of alternative apron points from a reasonable range.

[0060] Let J be the number of apron alternative points in the whole domain, J be a subset of N, and the aprons correspond to different travel demand points. Make basic assumptions: (1) After arriving at the air bus station by road transportation, residents take the air bus directly from the station to the target apron, ignoring the endogenous gathering time.

[0061] (2) A maximum of W apron sites can be built in the main urban area.

[0062] (3) The apron can cover the travel needs of the entire region.

[0063] (4) The location of the alternative apron point is known.

[0064] Based on the above assumptions, the decision variables can be set: (1) ——Surface section Traffic flow; (2) ——Flight segment Traffic flow; (3) ——0-1 variable, decision whether at the node Construction of aprons; (4) ——Decision whether to open a flight route , when the traffic flow on the road segment hour, ; otherwise, 0.

[0065] Then, the full model can be constructed based on the above decision variables: First, you need to set the first objective function: In the formula, and is the weight coefficient; For flight segment operating costs; The construction cost of a single apron; is the transportation cost per unit distance on the ground; is the air transportation cost per unit distance; is the length of the ground section; is the length of the aerial segment; is the flow rate of the surface road section; is the traffic volume of the aerial section.

[0066] Secondly, you need to set the first constraint. The first constraint may include the following: (1) Restriction on the number of apron sites. For example, the number of apron sites to be built must be greater than 1 and less than W: (2) Maximum ground travel distance limit. For example, limit the maximum travel distance between the demand point and the apron: (3) A maximum of one apron can be built at a single point. It is stipulated that a maximum of one apron can be built at a single apron alternative point: (4) The association between traffic and apron construction. For example, it is set that only the road sections with aprons built at the start and end points can be allocated traffic: In addition, it is necessary to specify that the decision variable is a 0-1 variable: It should be noted that the upper-level planning (U) and Obtained from the lower-level planning (L).

[0067] Therefore, the apron site selection set can be determined according to the first objective function and the first constraint condition through a multi-objective optimization algorithm.

[0068] According to a method for optimizing the location of a flying bus apron, based on the apron location set and user behavior data, a traffic flow result corresponding to the apron location set is determined by a traffic flow allocation algorithm, specifically including: Based on the second objective function and the second constraint condition, the traffic flow result corresponding to the apron site selection set is determined by the traffic flow allocation algorithm; The second objective function is: In the formula, represents the total transportation travel cost; Indicates road segment The unit flow traffic cost function; Indicates road segment Total traffic volume on The entropy term representing the path selection; Represents user behavior data Path Traffic on represents the path selection entropy weight, The larger it is, the more the selection behavior is concentrated on the lowest cost path; The second constraint includes: Flow conservation; Flight segment traffic restrictions; User behavior restrictions determined based on user behavior data.

[0069] Specifically, based on the apron layout given by the upper model, in the process of traffic flow allocation, the second objective function needs to be set first. The second objective function can be: In the formula, represents the total transportation travel cost; Indicates road segment The unit flow traffic cost function; Indicates road segment Total traffic volume on The entropy term representing the path selection; Represents user behavior data Path Traffic on represents the path selection entropy weight, The larger it is, the more the selection behavior is concentrated on the lowest cost path.

[0070] Secondly, the second constraint condition needs to be set. The second constraint condition includes: (1) Flow conservation. Flow conservation can be specified by the following formula: In the formula, Represents user behavior data Total traffic demand; Indicates road segment Total traffic volume on is an indicator variable, when the path Included road segments When , the value is 1, otherwise it is 0; Indicates ground or flight mode.

[0071] (2) Flight route flow restrictions. Since flight routes are subject to airspace capacity restrictions, the flow rates on the flight routes can be regulated. Cannot exceed the maximum flow limit , nor can it be lower than the minimum flow limit : (3) User behavior restrictions based on user behavior data. Travelers make travel choices based on the apron site selection scheme given by the upper-level model, thus generating traffic demand on each road segment in the network.

[0072] The flying bus apron site selection optimization method provided by the present invention is further explained below through embodiments in specific application scenarios.

[0073] First, this embodiment takes the Sioux Falls network as an example to perform apron site selection planning and design. Figure 3 The Sioux Falls network diagram provided by the present invention is as follows: Figure 4 The network topology structure and node and section numbering diagram provided by the present invention are as follows: Figure 4 As shown, the road network contains 24 nodes and 76 road segments.

[0074] Step S01: regional data preprocessing and regional rasterization.

[0075] Figure 5 The regional grid diagram provided by the present invention needs to be explained as follows: Figure 5This is just an example. In actual application, the number of grids is relatively large. Specifically, in actual application, the Sioux Falls network can be divided into several 10000×10000 square grids. Assume that the 24 nodes on the network are travel demand points for residents and are also alternative points for flying bus aprons. The travel distance and travel path between each node are calculated and stored as a csv file to provide data support for subsequent ground travel and air travel cost calculations. The travel OD data between the Sioux Falls network nodes are selected as the basis for travel demand calculations. Based on the above OD data, the traffic flow distribution of travel OD between flying bus aprons is studied.

[0076] Step S02: drawing a regional risk map.

[0077] Specifically, Figure 6 This is a schematic diagram of the Sioux Falls network area risk map provided by the present invention. Figure 6 As shown in the figure, the route confidence index, the measurement criteria of each index and its weight allocation method in the "Specifications for the Planning of Logistics Routes for Light and Small Unmanned Aerial Vehicles in Urban Scenarios" issued by the Civil Aviation Administration of China in 2022 are used as risk measurement criteria for flying bus route planning. According to the definition of the risk map, the risk value of [-2, 1] is assigned to each indicator of each grid based on the evaluation and assignment methods of various indicators such as meteorological environment, electromagnetic environment, obstacle environment, navigation performance, communication performance, surveillance performance, ground personnel safety, ground facility safety, search difficulty, secondary damage, privacy factors, and noise factors in the specifications. The comprehensive operation risk of each grid can be calculated according to the relevant formula and the risk map of the flight area can be drawn.

[0078] Step S03, flight path planning algorithm.

[0079] Depend on Figure 6 It can be seen that the risk value of some grids is relatively high, and the risk of flying buses operating in the airspace of such grids is relatively high. Therefore, such grids should be avoided when planning the flight path. Figure 7 The flight path diagram for the Sioux Falls network area considering travel risks provided by the present invention is as follows: Figure 7 As shown ( Figure 7 Only the flight paths between 5 nodes are shown). The Dijkstra algorithm can be used to plan the flight paths between the nodes on the Sioux Falls network. The flight distance must be the shortest while satisfying the risk constraints, and the flight trajectory between the nodes can be obtained.

[0080] Step S04, solving the site selection optimization model of the present invention.

[0081] Specifically, the upper layer is a multi-objective discrete site selection model, which is solved by calling gurobi in python; the lower layer is a traffic flow allocation model, and the commonly used one is the Frank-Wolfe algorithm. Traffic flow allocation also involves the search for effective paths. Considering the improvement of the subpath cost function in this paper, an effective path search algorithm based on subpath cost is proposed. Finally, the taboo search algorithm is used to achieve the solution of the combination of the upper and lower models.

[0082] Next, the embodiment of the present application introduces the upper and lower layer iterative solution algorithm for the two-layer programming model in step S03, and the process may include: Step S041: An effective path search algorithm based on sub-path costs.

[0083] Step 1: Initialize and assign values ​​to related parameters; , , , ; Step 2: Based on the Dijkstra algorithm, search for the shortest feasible path between OD pairs and calculate ; Step 3: Starting from the current node i, check the unmarked arcs adjacent to i, such as arc (i, j), and mark arc (i, j): , go to Step 4; Step 4: Order .like judge Is it true? If yes, go to step 5; if no, go to step 8; ,judge Is it true? If yes, go to step 5; if no, go to step 8; , go to step5; Step 5: Calculate according to the set parameter values , , go to Step 6; Step 6: Judgement Is it established? If yes, go to Step 7; if no, go to Step 8; Step 7: Judgement Is it true? If not, update the current node to j and go to Step 3. If yes, record the valid path and go to Step 8. Step 8: Use the backtracking method to return to the immediately preceding node and determine whether there is an unmarked arc adjacent to the current node i. If yes, proceed to Step 3; if not, continue to backtrack to the previous layer. If you backtrack to the starting point r and there is no unmarked arc, the algorithm ends.

[0084] Step S042: Frank-Wolfe algorithm processes the lower layer traffic flow distribution.

[0085] Step 1: Initialization. Based on , according to the effective path search algorithm, the shortest path between RS is obtained , and the traffic flow between rs All Assign to Path On, that is ,get ,make .

[0086] Step 2: Update. .

[0087] Step 3: Descending direction. Based on , check each OD pair rs in turn, and find the shortest path between rs according to the effective path search algorithm , and the traffic flow between rs All Assign to Path On, that is ,get .

[0088] Step 4: Search step length. Use methods such as binary search to find the solution that satisfies of .

[0089] Step 5: Move. .

[0090] Step 6: End condition. If The program ends, otherwise set n=n+l and go to Step 2.

[0091] Step S043, Figure 8 The algorithm design flow diagram provided by the present invention is as follows: Figure 8 As shown, the taboo search algorithm implements the upper and lower layer loop iterative solution.

[0092] Step 1: Select an initial feasible solution , and initialize the taboo table .

[0093] Step 2: If the termination condition is met, then end; otherwise, in the feasible solution Field Select the candidate set that meets the taboo requirements .

[0094] Step 3: From the candidate set Select a solution with the best evaluation value ,make , update the taboo table H and repeat step 2.

[0095] Step 4: Output the calculation results and stop.

[0096] Step S05: comparative analysis of different site selection models.

[0097] Specifically, the unit time operating costs and apron construction costs of different modes of transportation are specified, as shown in Table 1. (The reference flight charge standard opened in the Guangdong-Hong Kong-Macao Greater Bay Area is 300 yuan / person / 40 kilometers. The cost of driverless minibus travel refers to the cost of customized public transportation.) Table 1 Unit cost

[0098] Taking the Sioux Falls network as an example, the traditional P-medium model, the simple site selection model, the risk-based site selection model, and the bi-level programming model of the present invention are solved and analyzed. Among them, the P-medium model only considers straight-line distance and full coverage, and each demand point is served by a single vertical take-off and landing airport. It does not consider the operating costs associated with air travel paths. The coverage of the vertical take-off and landing airport to the demand point is shown in Table 2.

[0099] Table 2 Coverage of apron to demand points

[0100] The simple site selection model only considers one station being served by one apron, and considers the user's travel cost on the ground, while the air travel cost is calculated only by distance. On the basis of the simple site selection model, the risk factor is taken into consideration, and the final site selection result is consistent with that without considering the risk factor. However, due to the influence of the added risk factor, the flight trajectory of the flying bus is no longer a point-to-point straight-line flight, so the distance of the air operation has increased. The objective function of the simple site selection model considering risk is a multi-objective function. The entropy weight method is used to weight the objective function to obtain the final weight coefficient value , . Fig. 9 This is a schematic diagram of the comparison of site selection results of different models in the Sioux Falls network area provided by the present invention, and the specific apron site selection location distribution is obtained as follows Fig. 9 shown.

[0101] When the number of apron construction M=5, the solution results of the above models are shown in Table 3.

[0102] Table 3 Site selection results of different models

[0103] The flying bus apron site selection optimization device provided by the present invention is described below. The flying bus apron site selection optimization device described below and the flying bus apron site selection optimization method described above can be referenced to each other.

[0104] Fig.10 The schematic diagram of the structure of the device for optimizing the location of the apron for the flying bus provided by the present invention is as follows: Fig.10 As shown, the device includes the following modules: The processing module 1000 is used to rasterize the target area based on the traffic flow network and obtain the user behavior data of the target area; A calculation module 1010 is used to calculate the risk value based on each grid of the target area to obtain a risk map of the target area; A planning module 1020 is used to obtain an optimal site selection plan for the apron of the flying bus through a double-layer planning model based on the total cost, user behavior data and risk map; Among them, the upper model of the two-level planning model is a multi-objective discrete site selection model, which is used to determine the apron site selection set for the flying bus; the lower model of the two-level planning model is a traffic flow allocation model, which is used to allocate traffic flow.

[0105] According to a flying bus apron location optimization device provided by the present invention, based on total cost, user behavior data and risk map, a traffic planning scheme for the flying bus is obtained through a double-layer planning model, which specifically includes: In the upper model, based on the total cost and risk map, a multi-objective optimization algorithm is used to determine the apron site set; in the lower model, based on the apron site set and user behavior data, a traffic flow allocation algorithm is used to determine the traffic flow results corresponding to the apron site set; When the termination condition is met, the current apron site selection set is used as the apron site selection optimization plan and output; when the termination condition is not met, the apron site selection set is updated based on the traffic flow results corresponding to the apron site selection set, and it is continuously iterated until the termination condition is met, and the updated apron site selection set is used as the apron site selection optimization plan and output.

[0106] According to a flying bus apron location optimization device provided by the present invention, based on the total cost and risk map, a multi-objective optimization algorithm is used to determine the apron location set, specifically including: Based on the first objective function and the first constraint condition, a set of apron locations is determined by a multi-objective optimization algorithm; The first objective function is: In the formula, and is the weight coefficient; For flight segment operating costs; The construction cost of a single apron; is the transportation cost per unit distance on the ground; is the air transportation cost per unit distance; is the length of the ground section; is the length of the aerial segment; is the flow rate of the surface road section; is the flow rate of the aerial section; The first constraints include: Restriction on the number of parking spaces; Maximum ground travel distance limits; At most one helipad can be built at a single location; The correlation conditions between traffic flow and apron construction.

[0107] According to a flying bus apron site selection optimization device provided by the present invention, based on the apron site selection set and user behavior data, a traffic flow result corresponding to the apron site selection set is determined by a traffic flow allocation algorithm, specifically including: Based on the second objective function and the second constraint condition, the traffic flow result corresponding to the apron site selection set is determined by the traffic flow allocation algorithm; The second objective function is: In the formula, represents the total transportation travel cost; Indicates road segment The unit flow traffic cost function; Indicates road segment Total traffic volume on The entropy term representing the path selection; Represents user behavior data Path Traffic on represents the path selection entropy weight, The larger it is, the more the selection behavior is concentrated on the lowest cost path; The second constraint includes: Flow conservation; Flight segment traffic restrictions; User behavior restrictions determined based on user behavior data.

[0108] According to a flying bus apron location optimization device provided by the present invention, based on the traffic flow network, the target area is rasterized and the user behavior data of the target area is obtained, which specifically includes: Based on the traffic flow network, the target area is divided into multiple square grids, and the grid position is represented by the grid centroid coordinates; Based on the grid centroid coordinates, the user's travel start and end points and their corresponding travel demand data are mapped to each grid to obtain the user behavior data of the target area.

[0109] According to a flying bus apron site selection optimization device provided by the present invention, the upper model of the double-layer programming model is solved by calling gurobi by python; the lower model of the double-layer programming model is solved by the Frank-Wolfe algorithm; the double-layer programming model is iteratively solved between the upper model and the lower model by the taboo search algorithm.

[0110] Fig.11 A schematic diagram of the structure of an electronic device provided by the present invention, such as Fig.11 As shown, the electronic device may include: a processor 1110, a communications interface 1120, a memory 1130 and a communication bus 1140, wherein the processor 1110, the communications interface 1120 and the memory 1130 communicate with each other through the communication bus 1140. The processor 1110 may call the logic instructions in the memory 1130 to execute the flying bus apron location optimization method, which includes: Based on the traffic flow network, the target area is rasterized and the user behavior data of the target area is obtained; Calculate the risk value based on each grid in the target area to obtain a risk map of the target area; Based on the total cost, user behavior data and risk map, the optimal location selection scheme for the apron of the flying bus is obtained through a double-layer planning model; Among them, the upper model of the two-level planning model is a multi-objective discrete site selection model, which is used to determine the apron site selection set for the flying bus; the lower model of the two-level planning model is a traffic flow allocation model, which is used to allocate traffic flow.

[0111] In addition, the logic instructions in the above-mentioned memory 1130 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention can be essentially or partly embodied in the form of a software product that contributes to the prior art. 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. Various media that can store program codes.

[0112] On the other hand, the present invention further provides a computer program product, the computer program product comprising a computer program, the computer program can be stored on a non-transitory computer-readable storage medium, and when the computer program is executed by a processor, the computer can execute the method for optimizing the location of the flying bus apron provided by the above methods, the method comprising: Based on the traffic flow network, the target area is rasterized and the user behavior data of the target area is obtained; Calculate the risk value based on each grid in the target area to obtain a risk map of the target area; Based on the total cost, user behavior data and risk map, the optimal location selection scheme for the apron of the flying bus is obtained through a double-layer planning model; Among them, the upper model of the two-level planning model is a multi-objective discrete site selection model, which is used to determine the apron site selection set for the flying bus; the lower model of the two-level planning model is a traffic flow allocation model, which is used to allocate traffic flow.

[0113] In another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented when the computer program is executed by a processor to execute the method for optimizing the location of the apron for a flying bus provided by the above methods, the method comprising: Based on the traffic flow network, the target area is rasterized and the user behavior data of the target area is obtained; Calculate the risk value based on each grid in the target area to obtain a risk map of the target area; Based on the total cost, user behavior data and risk map, the optimal location selection scheme for the apron of the flying bus is obtained through a double-layer planning model; Among them, the upper model of the two-level planning model is a multi-objective discrete site selection model, which is used to determine the apron site selection set for the flying bus; the lower model of the two-level planning model is a traffic flow allocation model, which is used to allocate traffic flow.

[0114] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0115] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0116] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for optimizing the site selection of a flying bus apron, characterized in that: include: Based on the traffic flow network, the target area is rasterized and user behavior data of the target area is obtained; Calculate the risk value based on each grid of the target area to obtain a risk map of the target area; Based on the total cost, the user behavior data and the risk map, an optimization plan for the apron location of the flying bus is obtained through a double-layer planning model; Among them, the upper model of the two-layer programming model is a multi-objective discrete site selection model, which is used to determine the apron site selection set of the flying bus; the lower model of the two-layer programming model is a traffic flow allocation model, which is used to allocate traffic flow.

2. The method for optimizing the location of the flying bus apron according to claim 1, characterized in that: Based on the total cost, the user behavior data and the risk map, a traffic planning scheme for the flying bus is obtained through a double-layer planning model, specifically including: In the upper model, based on the total cost and the risk map, a multi-objective optimization algorithm is used to determine the apron site set; in the lower model, based on the apron site set and the user behavior data, a traffic flow result corresponding to the apron site set is determined through a traffic flow allocation algorithm; When the termination condition is met, the current apron site selection set is used as the apron site selection optimization plan and output; when the termination condition is not met, the apron site selection set is updated based on the traffic flow results corresponding to the apron site selection set, and it is continuously iterated until the termination condition is met, and the updated apron site selection set is used as the apron site selection optimization plan and output.

3. The method for optimizing the location of the flying bus apron according to claim 2, characterized in that: Based on the total cost and the risk map, a multi-objective optimization algorithm is used to determine the apron site set, specifically including: Based on the first objective function and the first constraint condition, a set of apron locations is determined by a multi-objective optimization algorithm; The first objective function is: In the formula, and is the weight coefficient; For flight segment operating costs; The construction cost of a single apron; is the transportation cost per unit distance on the ground; is the air transportation cost per unit distance; is the length of the ground section; is the length of the aerial segment; is the flow rate of the surface road section; is the flow rate of the aerial section; The first constraint condition includes: Restriction on the number of parking spaces; Maximum ground travel distance limits; At most one helipad can be built at a single location; The correlation conditions between traffic flow and apron construction.

4. The method for optimizing the location of the flying bus apron according to claim 2, characterized in that: Based on the apron site selection set and the user behavior data, a traffic flow result corresponding to the apron site selection set is determined by a traffic flow allocation algorithm, specifically including: Based on the second objective function and the second constraint condition, a traffic flow result corresponding to the apron site selection set is determined by a traffic flow allocation algorithm; The second objective function is: In the formula, represents the total transportation travel cost; Indicates road segment The unit flow traffic cost function; Indicates road segment Total traffic volume on The entropy term representing the path selection; Represents user behavior data Path Traffic on represents the path selection entropy weight, The larger it is, the more the selection behavior is concentrated on the lowest cost path; The second constraint condition includes: Flow conservation; Flight segment traffic restrictions; User behavior restrictions determined based on the user behavior data.

5. The method for optimizing the location of the flying bus apron according to claim 1, characterized in that: Based on the traffic flow network, the target area is rasterized and the user behavior data of the target area is obtained, including: Based on the traffic flow network, the target area is divided into multiple square grids, and the grid position is represented by the grid centroid coordinates; Based on the grid centroid coordinates, the user's travel start and end points and their corresponding travel demand data are mapped to each grid to obtain the user behavior data of the target area.

6. The method for optimizing the location of the apron for flying buses according to any one of claims 1 to 4, characterized in that: The upper model of the two-level programming model is solved by calling gurobi by python; the lower model of the two-level programming model is solved by the Frank-Wolfe algorithm; the two-level programming model is iteratively solved between the upper model and the lower model by the taboo search algorithm.

7. A device for optimizing the site selection of a flying bus apron, characterized in that: include: A processing module, used for rasterizing a target area based on a traffic flow network and obtaining user behavior data of the target area; A calculation module, used for calculating the risk value based on each grid of the target area to obtain a risk map of the target area; A planning module, for obtaining an optimal site selection scheme for the apron of the flying bus through a double-layer planning model based on the total cost, the user behavior data and the risk map; Among them, the upper model of the two-layer programming model is a multi-objective discrete site selection model, which is used to determine the apron site selection set of the flying bus; the lower model of the two-layer programming model is a traffic flow allocation model, which is used to allocate traffic flow.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method for optimizing the location of the flying bus apron is implemented as described in any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for optimizing the location of a flying bus apron is implemented as described in any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for optimizing the location of a flying bus apron is implemented as described in any one of claims 1 to 6.

Images