A method for planning electronic fences for shared bicycles based on plot segmentation and maximum regional coverage
By using the method based on plot segmentation and maximum area coverage, the problem of inaccurate urban division in the existing shared bicycle electronic fence planning is solved, and more accurate parking demand calculation and coverage effect is achieved, improving the planning quality of electronic fences.
Patent Information
- Application Number
- CN202111086544.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-09-16
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2041-09-16
AI Technical Summary
The existing electronic fence planning method for shared bicycles has inaccurate problems in urban division and parking demand calculation, and the road structure and administrative division cannot be effectively considered, resulting in unreasonable electronic fence setting and affecting the degree of coverage.
Using a method based on plot segmentation and maximum area coverage, by obtaining shared bicycle usage data, dividing more refined areas, calculating parking requirements for each area, and using the area maximum coverage model to optimize the setting location and capacity of electronic fences, considering Regional level coverage.
It realizes more accurate electronic fence planning, reduces calculation errors, meets parking needs in real scenarios, and improves the coverage effect of electronic fences.
Smart Images

Figure CN113850310B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of information technology, and in particular relates to a method for planning electronic fences for shared bicycles based on plot subdivision and maximum regional coverage. Background Art
[0002] In recent years, the emergence of shared bicycles has provided a more convenient way for residents to travel. People can borrow and return bicycles near the starting point and the end point, solving the "last mile" problem. However, with the popularity of shared bicycles, it has also caused related urban problems. A large number of users parked their bicycles in disorder, occupying public spaces such as sidewalks, hindering people's normal travel.
[0003] In order to regulate users' parking behavior, most existing studies have conducted theoretical research from the perspectives of social norms, institutional environment and reward mechanisms, and encouraged users to park their vehicles in appropriate places by formulating rules. Electronic fences are a digital infrastructure that specifies the geographical areas where users are allowed to park their vehicles. There is only one literature: Electronic fence planning for dockless bicycle sharing services [J]. Journal of Clean Production, 2019206:383-393. The electronic fence planning method based on big data analysis was studied. The authors first divided the city into grids, calculated people's parking needs based on the grids and the collected shared bicycle usage records, and then used the maximum coverage model to obtain the location of the electronic fence, in order to maximize the parking demand under the premise of setting limited fences. However, the implementation method in the literature has two defects: 1) They simply divided the city into grids and could not depict the precise parking demand. It is more meaningful to consider the road network structure and administrative divisions when calculating parking demand. 2) Their method is based on an ideal assumption that the geographic space (grid) can be abstracted as a point, so that the parking demand of a grid can be represented by a point. However, the parking spots for shared bikes are distributed throughout the area, and this assumption introduces uncertainty and errors in calculating distances and assessing coverage.
[0004] When planning electronic fences for shared bikes, if we can use areas with real geographical significance as the basic calculation unit for parking demand, and consider the coverage rate at the regional level when evaluating the coverage of electronic fences for demand areas, we can get more accurate electronic fence planning results. However, this poses greater challenges to both model construction and problem solving. Summary of the invention
[0005] The purpose of the present invention is to design a shared bicycle electronic fence planning method based on plot subdivision and maximum regional coverage to solve the above technical problems.
[0006] In order to solve the above technical problems, the specific technical solutions of the method for planning electronic fences for shared bicycles based on land parcel subdivision and maximum regional coverage of the present invention are as follows:
[0007] A method for planning electronic fences for shared bicycles based on plot subdivision and maximum regional coverage includes the following steps:
[0008] Step 1: Obtain the shared bicycle usage dataset and preprocess the data;
[0009] Step 2: Divide the urban land into finer areas based on usage records and find the parking demand of the area;
[0010] Step 3: Use the regional maximum coverage model to select the area where the electronic fence needs to be set up;
[0011] Step 4: Calculate the capacity of the electronic fence;
[0012] Step 5: Calculate the exact location to set up the geo-fence.
[0013] Furthermore, the step 1 includes the following specific steps:
[0014] Step 1.1: Get the usage dataset of shared bicycles over a period of time and store them in the database;
[0015] A usage record R DBS It is expressed as follows:
[0016] R DBS =(bID,startT,startLoc,endT,endLoc),
[0017] Where bID is the shared bicycle ID, startT and endT are the borrowing time and return time, startLoc and endLoc are the borrowing location and return location, and the location includes longitude and latitude information; Step 1.2: Preprocess the shared bicycle data and remove invalid data;
[0018] For each usage record, the corresponding riding distance, time and speed are calculated; records with a riding distance less than 0.2 km or greater than 15 km, records with a riding time less than 1 minute or greater than 3 hours, and records with a riding speed greater than 400 m / min are deleted;
[0019] Step 1.3: Get all parking spots, calculate the number of vehicles and the total parking demand;
[0020] According to the usage records, all parking spots are obtained, that is, the startLoc and endLoc of all records, and the number of vehicles bikeNum and the total number of records tRecNum are calculated. The total parking demand tParkNum is 2*tRecNum.
[0021] Further, step 2 includes the following steps:
[0022] Step 2.1: Obtain plot data, map all parking points to the corresponding plots according to the plot boundaries, and calculate the daily average parking demand for each plot;
[0023] Step 2.2: Calculate the minimum area bounding rectangle for each plot (area), where the direction of the rectangle is the same as that of the plot (area). The plot passed in for the first time is called a "plot", and the area obtained by subdividing the plot is called an "area";
[0024] Step 2.3: Divide the plot into smaller areas based on the minimum area bounding rectangle and the daily average parking demand of the plot;
[0025] First, find the perpendicular bisector on the long side of the minimum area bounding rectangle, and divide the plot into two smaller areas; if the daily average parking demand of the area is less than the predefined threshold T subd , then the subdivision process ends, and the parking demand for this area is obtained at the same time; otherwise, return to step 2.2, and the area will be iteratively subdivided until the termination condition is met.
[0026] Further, step 3 includes the following steps:
[0027] Step 3.1: Define variables;
[0028] Define the following variables: n is the number of demand areas, m is the number of candidate areas for setting up electronic fences; u i represents the daily average parking demand of the i-th demand area (1 ≤ i ≤ n), p is the pre-given number of electronic fences to be set up, b ij represents the total number of services (parking volume) provided by the electronic fence located in area j to demand area i, N i represents the set of electronic fences that can provide some services to demand area i, x j ∈{0, 1}, when x j = 1, it means that area j is selected as the area for setting up an electronic fence, and when x j = 0, it means that no electronic fence is set up in area j, and λ i represents the total number of all services obtained by demand area i;
[0029] Step 3.2: Calculate the candidate areas for setting up electronic fences;
[0030] Sort the areas according to the daily average parking demand of the areas, and obtain the m (m << n) areas with the largest daily average parking demand as the candidate areas for setting up electronic fences;
[0031] Step 3.3: Use the regional maximum coverage model to optimize the area where the electronic fence needs to be set up; the optimization objective function is:
[0032]
[0033] satisfy:
[0034]
[0035]
[0036]
[0037]
[0038] The objective function (1) aims to maximize the overall coverage provided by the geo-fences, and the constraint (2) limits λ by summing the total coverage obtained by the demand area i from all geo-fences that meet the conditions. i , constraint (3) defines the upper bound u on the total coverage that can be obtained for each demand region i i , constraint (4) limits the number of electronic fences to be set, and constraint (5) sets the decision variable x j Binary restrictions are performed; the model finally calculates the area number where the electronic fence needs to be set up and the number of parking requirements pdNum assigned to each electronic fence j .
[0039] Furthermore, step 4 comprises the following steps:
[0040] Define the number of electronic fences set, and calculate the actual capacity pdNum of the jth electronic fence according to the total number of vehicles bikeNum, the total parking demand tParkNum and the number of parking demands assigned to each electronic fence j *(bikeNum / tParkNum), according to the actual capacity, select the closest value from the preset quantity set to get the final capacity of the electronic fence.
[0041] Furthermore, the step 5 comprises the following steps:
[0042] The DBSCAN method is used to cluster all parking spots falling in an area, and the cluster with the largest number of parking spots among all clusters is found. The center of this cluster is selected as the precise location for setting the fence.
[0043] The shared bicycle electronic fence planning method based on plot subdivision and regional maximum coverage of the present invention has the following advantages: The characteristics and innovation of the method of the present invention are that a new shared bicycle electronic fence planning method based on big data analysis is proposed. First, the urban plots are subdivided into regions. The division of the regions not only takes into account the road structure and administrative boundaries, but also takes into account the parking needs of the regions. Compared with the existing methods, the division of urban space is more reasonable and can obtain more meaningful parking needs; then the regional maximum coverage model is used to calculate the area where the fence needs to be set. The existing method abstracts the area as a point. When optimizing the model, the parking needs of an area can only be fully met or not met. The present invention takes into account the spatial shape of the area when planning the electronic fence, and can calculate the partial coverage of the area to obtain the electronic fence service, thereby reducing potential errors in the calculation process and better meeting the needs of electronic fence planning in real scenarios. BRIEF DESCRIPTION OF THE DRAWINGS
[0044] Figure 1 This is an overall flow chart of the method for planning electronic fences for shared bicycles based on land parcel subdivision and maximum regional coverage of the present invention;
[0045] Figure 2 It is a schematic diagram of the land plot subdivision process of the present invention. DETAILED DESCRIPTION
[0046] In order to better understand the purpose, structure and function of the present invention, the following is a further detailed description of a shared bicycle electronic fence planning method based on plot subdivision and maximum regional coverage of the present invention in conjunction with the accompanying drawings.
[0047] like Figure 1 As shown, a method for planning electronic fences for shared bicycles based on plot subdivision and maximum regional coverage includes the following steps:
[0048] Step 1: Obtain the shared bicycle usage dataset and preprocess the data.
[0049] Step 1.1: Get the usage dataset of shared bicycles over a period of time and store them in the database.
[0050] A usage record R DBS It is expressed as follows:
[0051] R DBS =(bID,startT,startLoc,endT,endLoc)
[0052] Among them, bID is the shared bicycle number, startT and endT are the borrowing time and return time, startLoc and endLoc are the borrowing location and return location, and the location includes longitude and latitude information.
[0053] Step 1.2: Preprocess the shared bicycle data and remove invalid data.
[0054] For each usage record, the corresponding riding distance, time and speed are calculated. Since a short trip may be caused by vehicle failure or GPS positioning error, and a long trip may be caused by people forgetting to return the vehicle or vehicle maintenance, the records with a riding distance of less than 0.2km or more than 15km are deleted, and the records with a time of less than 1 minute or more than 3 hours are deleted. In addition, the records with a riding speed greater than 400 meters per minute are deleted.
[0055] Step 1.3: Get all parking spots, calculate the number of vehicles and the total parking demand.
[0056] According to the usage records, all parking points are obtained, that is, the startLoc and endLoc of all records. The number of vehicles bikeNum and the total number of records tRecNum are calculated. The total parking demand tParkNum is 2*tRecNum.
[0057] Step 2: Divide the urban land into finer areas based on usage records and find the parking demand of the area. Figure 2 As shown, step 2 includes the following steps:
[0058] Step 2.1: Obtain the plot data, map all parking spots to the corresponding plots according to the plot boundaries, and calculate the average daily parking demand for each plot.
[0059] Step 2.2: Calculate the minimum area bounding rectangle of each plot (region), the direction of which is consistent with the direction of the plot (region). The first plot passed in is called a "plot", and the area obtained by subdividing the plot is called a "region".
[0060] Step 2.3: Divide the plot into smaller areas based on the minimum area bounding rectangle and the average daily parking demand of the plot.
[0061] First, find the perpendicular bisector of the minimum area enclosing rectangle on the long side and divide the plot into two smaller areas. If the average daily parking demand of the area is less than the predefined threshold T subd , then the subdivision process ends and the parking demand of the area is obtained. Otherwise, return to step 2.2, and the area will be iteratively subdivided until the termination condition is met.
[0062] Step 3: Use the regional maximum coverage model to select the area where the electronic fence needs to be set up.
[0063] Step 3.1: Define variables.
[0064] Define the following variables: n is the number of demand regions, and m is the number of candidate regions for setting up electronic fences. u i represents the average daily parking demand of the i-th demand region (1 ≤ i ≤ n). p is the pre-given number of electronic fences to be set up. b ij represents the total number of services (parking volume) provided by the electronic fence located in region j to demand region i. N i represents the set of electronic fences that can provide some services to demand region i. x j ∈ {0, 1}, when x j = 1, it means that region j is selected as the region for setting up an electronic fence, and when x j = 0, it means that no electronic fence is set up in region j. λ i represents the total number of all services obtained by demand region i.
[0065] Step 3.2: Calculate the candidate regions for setting up electronic fences.
[0066] Sort the regions according to the average daily parking demand of the regions, and obtain the m (m << n) regions with the largest average daily parking demand as the candidate regions for setting up electronic fences.
[0067] Step 3.3: Use the maximum coverage model of the region to optimize and obtain the regions where electronic fences need to be set up.
[0068] The objective function for optimization is:
[0069]
[0070] Subject to:
[0071]
[0072]
[0073]
[0074]
[0075] The objective function (1) aims to maximize the overall coverage provided by the electronic fences. Constraint (2) restricts λ by summing up the total coverage obtained by demand region i from all eligible electronic fences i . Constraint (3) defines the upper bound u of the total coverage that each demand region i can obtain i . Constraint (4) limits the number of electronic fences to be set up. Constraint (5) performs a binary restriction on the decision variable x j . The model finally calculates the region numbers where electronic fences need to be set up and the number of parking demands pdNum assigned to each electronic fence j .
[0076] Step 4: Calculate the capacity of the electric fence.
[0077] Define the number of electronic fences that can be set, such as {10, 20, 30, 50}. According to the total number of vehicles bikeNum, the total parking demand tParkNum and the number of parking demands assigned to each electronic fence, calculate the actual capacity pdNum of the jth electronic fence j *(bikeNum / tParkNum). According to the actual capacity, the closest value is selected from the preset quantity set to obtain the final capacity of the electronic fence.
[0078] Step 5: Calculate the exact location to set up the geo-fence.
[0079] The DBSCAN method is used to cluster all parking spots falling in an area, and the cluster with the largest number of parking spots among all clusters is found. The center of this cluster is selected as the precise location for setting the fence.
[0080] It is to be understood that the present invention is described by some embodiments, and it is known to those skilled in the art that various changes or equivalent substitutions may be made to these features and embodiments without departing from the spirit and scope of the present invention. In addition, under the teachings of the present invention, these features and embodiments may be modified to adapt to specific circumstances and materials without departing from the spirit and scope of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of this application are within the scope of protection of the present invention.
Claims
1. A method for planning electronic fences for shared bicycles based on plot subdivision and maximum regional coverage, characterized in that: It includes the following steps: Step 1: Obtain the shared bike usage dataset and preprocess the data; Step 2: Divide the urban plots into finer regions based on the usage records and calculate the parking demand of the regions; Step 2.1: Obtain the plot data, map all parking points to the corresponding plots according to the plot boundaries, and calculate the daily average parking demand of each plot; Step 2.2: Calculate the minimum area bounding rectangle of each plot or region. The direction of this rectangle is the same as that of the plot or region. The plot passed in for the first time is called a "plot", and the region obtained by subdividing the plot is called a "region"; Step 2.3: Divide the plots into smaller regions based on the minimum area bounding rectangle and the daily average parking demand of the plots; First, find the perpendicular bisector of the minimum area enclosing rectangle on the long side to divide the plot into two smaller areas; if the average daily parking demand of the area is less than the predefined threshold T subd , then the subdivision process ends and the parking demand of the area is obtained; otherwise, return to step 2.2, and the area will be iteratively subdivided until the termination condition is met; Step 3: Use the maximum coverage model of regions to select the regions where electronic fences need to be set; Step 3.1: Define variables; Define the following variables: n is the number of required areas, m is the number of candidate areas for setting up electronic fences; u i represents the average daily parking demand of the i-th demand area, 1≤i≤n, p is the number of electronic fences set in advance, b ij N represents the total number of services provided by the electronic fence located in area j to the demand area i, that is, the parking volume, i represents the set of electronic fences that can provide some services for demand area i, x j ∈{0,1}, when x j =1, it means that area j is selected as the area for setting up the electronic fence. j =0 means that no electronic fence is set in area j, λ i represents the total number of all services obtained by demand area i; Step 3.2: Calculate the candidate regions for setting electronic fences; Sort the regions according to the daily average parking demand of the regions, and obtain the m regions with the largest daily average parking demand, where m << n, as the candidate regions for setting electronic fences; Step 3.3: Optimize using the maximum coverage model of regions to obtain the regions where electronic fences need to be set; The objective function for optimization is: Satisfy: The objective function (1) aims to maximize the overall coverage provided by the geo-fences, and the constraint (2) limits λ by summing the total coverage obtained by the demand area i from all geo-fences that meet the conditions. i , constraint (3) defines the upper bound u on the total coverage that can be obtained for each demand region i i , constraint (4) limits the number of electronic fences to be set, and constraint (5) sets the decision variable x j Binary restrictions are performed; the model finally calculates the area number where the electronic fence needs to be set up and the number of parking requirements pdNum assigned to each electronic fence j ; Step 4: Calculate the capacity of the electronic fence; Step 5: Calculate the exact location of setting the electronic fence.
2. The method for planning electronic fences for shared bicycles based on plot subdivision and maximum regional coverage according to claim 1 is characterized in that: The said Step 1 includes the following specific steps: Step 1.1: Obtain the shared bike usage dataset within a period of time and store them in the database; A usage record R DBS It is expressed as follows: R DBS =(bID,startT,startLoc,endT,endLoc), where bID is the shared bike number, startT and endT are the borrowing time and returning time, startLoc and endLoc are the borrowing location and returning location, and the location includes longitude and latitude information; Step 1.2: Preprocess the shared bike data and remove invalid data; For each usage record, calculate the corresponding riding distance, time consumption, and speed; delete the records with a riding distance less than 0.2 km or greater than 15 km, delete the records with a time consumption less than 1 minute or greater than 3 hours, and delete the records with a riding speed greater than 400 meters per minute; Step 1.3: Obtain all parking points, calculate the number of vehicles, and the total parking demand; According to the usage records, obtain all parking points, that is, startLoc and endLoc of all records, calculate the number of vehicles bikeNum and the total number of records tRecNum, and the total parking demand tParkNum is 2 * tRecNum.
3. The method for planning electronic fences for shared bicycles based on plot subdivision and maximum regional coverage according to claim 1 is characterized in that: The said Step 4 includes the following steps: Define the number of electronic fences set, and calculate the actual capacity pdNum of the jth electronic fence according to the total number of vehicles bikeNum, the total parking demand tParkNum and the number of parking demands assigned to each electronic fence j *(bikeNum / tParkNum), according to the actual capacity, select the closest value from the preset quantity set to get the final capacity of the electronic fence.
4. The method for planning electronic fences for shared bicycles based on plot subdivision and maximum regional coverage according to claim 1 is characterized in that: The said Step 5 includes the following steps: Use the DBSCAN method to cluster all parking points falling in a region, find the cluster with the largest number of parking points in all clusters, and select the center of this cluster as the exact location for setting the fence.
Citation Information
Patent Citations
Hadoop-based shared bicycle parking site selection method
CN108764555A
Shared bicycle parking area layout method based on clustering algorithm
CN111881939A
