Mobile sensor layout and active scheduling method based on shared bicycles
By laying sensors in the shared bicycle system and actively scheduling, the problem of data acquisition space coverage deviation in on-board perception technology is solved, and more efficient urban perception and sensor utilization are achieved.
Patent Information
- Application Number
- CN202411480617.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-23
- Publication Date
- 2025-02-25
AI Technical Summary
The existing on-board perception technology has spatial coverage deviations in the data acquisition process, especially the potential perception capabilities of shared bicycle systems have not been fully explored and optimized.
By arranging sensors in the shared bicycle system and actively scheduling, the distribution of sensors at shared bicycle parking points is optimized, and data collection efficiency is improved by using user guidance.
It effectively reduces spatial sampling deviation, improves the space-time coverage of urban perception, and improves the utilization rate of mobile sensors.
Smart Images
Figure BDA0005097271730000028 
Figure BDA0005097271730000031 
Figure BDA0005097271730000047
Abstract
Description
Technical Field
[0001] The present invention relates to vehicle-mounted sensing technology, and in particular to a mobile sensor deployment and active scheduling method based on a shared bicycle. Background Art
[0002] In-vehicle sensing technology has been widely used in the field of urban sensing in recent years, and has attracted widespread attention due to its low cost and wide spatial and temporal coverage. This technology can effectively cover a large area of urban space by collecting data while vehicles equipped with sensors are driving. However, the data collection quality of the in-vehicle sensing system depends largely on the mobility pattern of the carrier vehicle. For example, taxis, as a typical urban sensing carrier, are driven by profit and tend to be concentrated in high-demand areas, resulting in spatial deviations in the sampled data. Buses, due to their fixed routes, cannot cover areas beyond their operating network, resulting in spatial limitations in data collection. In addition, although dedicated sensor vehicles can conduct accurate regional surveys, they are difficult to deploy on a large scale due to their high deployment and operating costs. In contrast, the popularity of shared bicycle systems in cities has increased year by year, with wide coverage and dense parking points, making them a potential refined urban sensing platform. Despite the huge sensing potential of shared bicycle systems, current research has not fully explored their application and optimization in urban sensing. In particular, most existing studies focus on passive data collection, ignoring the issues of sensor deployment strategies and how to improve data collection efficiency through user guidance. Summary of the invention
[0003] The purpose of the present invention is to give full play to the perception ability of shared bicycles as urban sensors by reasonably deploying sensors and actively scheduling. To this end, the present invention provides a mobile sensor deployment and active scheduling method based on shared bicycles.
[0004] A mobile sensor deployment and active scheduling method based on a shared bicycle of the present invention comprises the following steps:
[0005] Step 1: Acquisition and processing of research data.
[0006] (1) Obtain historical riding order data from the shared bicycle system, including the starting point, end point, start time, and end time information of each ride; and obtain urban road network data from open source map services, including urban road nodes, road section connection relationships, and geographic location coordinates.
[0007] (2) Extract shared bicycle station information through shared bicycle order data.
[0008] (3) Generate the driving route between the starting point and the end point of each trip based on the road network information and the shortest path algorithm.
[0009] (4) Clean the shared bicycle travel data and filter out order data with a single riding time that is too long and a departure time that is not within the research range.
[0010] Step 2: Build a simulation system to simulate the running trajectory of shared bicycles.
[0011] Including determining the number of shared bicycles b at the initial moment of each station s s And simulate the running trajectory of each shared bicycle;
[0012] When determining the initial number of bicycles at each station, it is necessary to ensure that all travel orders can be met; to this end, it is achieved by simulating the flow of each station at each moment; first assume that at the initial moment t0, each station The number of shared bicycles is At each time step t, the number of shared bikes at each station is updated and recorded based on the order data
[0013]
[0014] In the formula, t0 represents the simulation start time, T represents the simulation end time, represents the set of all sites, represents the number of bicycles at the shared bicycle station s at time t; represents the number of bicycles returning to station s at time t; Represents the number of bicycles departing from station s at time t.
[0015] Finally, the number of shared bicycles b at the initial moment of each station s s Calculated by the following formula:
[0016]
[0017] With the number of vehicles at each station determined at the initial moment, the running trajectory of each vehicle is simulated, that is, the sequence of orders it serves in sequence. Specifically, it is assumed that when each order is served, one vehicle will be randomly selected from all available vehicles at its starting station.
[0018] Step 3: Solve the sensor-sharing bicycle station assignment problem.
[0019] Firstly, the relationship between the number of sensors deployed at each station and the number of times each road section is covered is found; then, a mixed integer linear programming model is proposed to solve the sensor-bike-sharing station allocation problem, with the goal of determining the optimal distribution of sensors at each station to maximize spatial coverage.
[0020] The coverage of a certain road section by a bicycle equipped with sensors at each station is described by binomial distribution: Assume that a certain shared bicycle station s has n s If a shared bicycle is equipped with a sensor, then the shared bicycles equipped with sensors at this station will The coverage is represented by c s,e ~B(n s ,p s,e ), where p s,e represents the probability that a bicycle at shared bicycle station s visits road segment e during this time period; therefore, during this simulation time period, the average number of times N that shared bicycles at station s cover road segment e is s,e It can be expressed as the expected value of the binomial distribution, as follows:
[0021]
[0022] Based on the above rules, the total number of given sensors N s , the following mixed integer linear programming model is established to solve the sensor-sharing bicycle station allocation problem to maximize spatial coverage:
[0023]
[0024]
[0025] In the formula, l e represents the length of section e; p s,e represents the probability that the shared bicycles at station s cover road section e; b s represents the number of shared bicycles at station s; N s Indicates the total number of sensors; N e is an auxiliary variable, indicating the total number of times the shared bicycles equipped with sensors cover the road section e; y e is a 0-1 auxiliary variable, which takes the value 1 if the number of coverages of section e exceeds K times; n s is a decision variable, indicating the number of sensors assigned to the shared bicycle station s; the parameter M can be a sufficiently large positive integer; the parameter K indicates that each road section can be covered K times to meet the monitoring requirements.
[0026] Step 4: Actively dispatch vehicles equipped with sensors.
[0027] When users choose to ride a bicycle at a bicycle station, the application software will give priority to recommending shared bicycles equipped with sensors, and give monetary incentives to users who accept the guidance.
[0028] The beneficial technical effects of the present invention compared with the prior art are:
[0029] (1) The present invention optimizes the distribution of sensors at shared bicycle parking spots, reduces spatial sampling bias, and improves the spatiotemporal coverage of overall urban perception.
[0030] (2) The present invention recommends shared bicycles to users and guides them to give priority to bicycles equipped with sensors, thereby improving the utilization rate of mobile sensors. BRIEF DESCRIPTION OF THE DRAWINGS
[0031] Figure 1 This is a schematic diagram of the process flow of the mobile sensor deployment and active scheduling method based on shared bicycles of the present invention. DETAILED DESCRIPTION
[0032] The present invention is further described in detail below with reference to the accompanying drawings and specific implementation methods.
[0033] The process of the mobile sensor deployment and active scheduling method based on shared bicycles of the present invention is as follows: Figure 1 As shown, the specific steps include:
[0034] Step 1: Acquisition and processing of research data.
[0035] (1) Obtain historical riding order data from the shared bicycle system, including the starting point, end point, start time, and end time information of each ride; and obtain urban road network data from open source map services (such as OpenStreetMap), including the connection relationships and geographic location coordinates of urban road nodes and road sections.
[0036] (2) Extract shared bicycle station information through shared bicycle order data.
[0037] (3) Generate the driving route between the starting point and the end point of each trip based on the road network information and the shortest path algorithm.
[0038] (4) Clean the shared bicycle travel data and filter out order data with a single riding time that is too long and a departure time that is not within the research range.
[0039] Step 2: Build a simulation system to simulate the running trajectory of shared bicycles.
[0040] Including determining the number of shared bicycles b at the initial moment of each station s s And simulate the running trajectory of each shared bicycle;
[0041] When determining the initial number of bicycles at each station, it is necessary to ensure that all travel orders can be met; to this end, it is achieved by simulating the flow of each station at each moment; first assume that at the initial moment t0, each station The number of shared bicycles is At each time step t, the number of shared bikes at each station is updated and recorded based on the order data
[0042]
[0043] In the formula, t0 represents the simulation start time, T represents the simulation end time, S represents the set of all sites, represents the number of bicycles at the shared bicycle station s at time t; represents the number of bicycles returning to station s at time t; Represents the number of bicycles departing from station s at time t.
[0044] Finally, the number of shared bicycles b at the initial moment of each station s s Calculated by the following formula:
[0045]
[0046] With the number of vehicles at each station determined at the initial moment, the running trajectory of each vehicle is simulated, that is, the sequence of orders it serves in sequence. Specifically, it is assumed that when each order is served, one vehicle will be randomly selected from all available vehicles at its starting station.
[0047] Step 3: Solve the sensor-sharing bicycle station assignment problem.
[0048] This step solves the problem of how to reasonably allocate sensors to each shared bicycle station to optimize the spatial coverage of the sensor network on the road network. First, find the relationship between the number of sensors deployed at each station and the number of times each road section is covered; then, a mixed integer linear programming model is proposed to solve the sensor-shared bicycle station allocation problem, with the goal of determining the optimal distribution of sensors at each station to maximize spatial coverage.
[0049] The coverage of a certain road section by a bicycle equipped with sensors at each station is described by binomial distribution: Assume that a certain shared bicycle station s has n s If a shared bicycle is equipped with a sensor, then the shared bicycles equipped with sensors at this station will The coverage is represented by c s,e ~b(n s ,p s,e ), where p s,e represents the probability that a bicycle at shared bicycle station s visits road section e during this time period (this parameter can be obtained by fitting the simulated shared bicycle trajectory); therefore, during this simulated time period, the average number of times N that shared bicycles at station s cover road section e is s,e It can be expressed as the expected value of the binomial distribution, as follows:
[0050]
[0051] Based on the above rules, the total number of given sensors N s , the following mixed integer linear programming model is established to solve the sensor-sharing bicycle station allocation problem to maximize spatial coverage:
[0052]
[0053] In the formula, l e represents the length of section e; p s,e represents the probability that the shared bicycles at station s cover road section e; b s represents the number of shared bicycles at station s; N s Indicates the total number of sensors; N e is an auxiliary variable, indicating the total number of times the shared bicycles equipped with sensors cover the road section e; y e is a 0-1 auxiliary variable, which takes the value 1 if the number of coverages of section e exceeds K times; n s is a decision variable, indicating the number of sensors assigned to the shared bicycle station s; the parameter M can be a sufficiently large positive integer; the parameter K indicates that each road section can be covered K times to meet the monitoring requirements.
[0054] Step 4: Actively dispatch vehicles equipped with sensors.
[0055] When users choose to ride a bicycle at a bicycle station, the application software will give priority to recommending shared bicycles equipped with sensors, and give monetary incentives to users who accept the guidance.
[0056] Through experiments, it is verified that when the monitoring frequency requirement is low (for example, once a day), the strategy of optimizing sensor allocation in the present invention is very effective, which can increase the sensor utility by about 13%-28%. Active scheduling can further increase the sensor benefit by 11.7-24.1%. This shows that actively guiding users to use bicycles equipped with sensors can further maximize the value of sensors.
Claims
1. A mobile sensor deployment and active scheduling method based on shared bicycles, characterized in that: The following steps are involved: Step 1: Acquisition and processing of research data; (1) Obtain historical riding order data from the shared bicycle system, including the starting point, end point, start time, and end time information of each ride; And obtain urban road network data from open source map services, including urban road nodes, road section connection relationships and geographic location coordinates; (2) Extracting shared bicycle station information through shared bicycle order data; (3) Generate the driving route between the starting point and the end point of each trip based on the road network information and the shortest path algorithm; (4) Clean the shared bicycle travel data and filter out the order data with a single riding time that is too long and the departure time that is not within the research range; Step 2: Build a simulation system to simulate the running trajectory of shared bicycles; Including determining the number of shared bicycles b at the initial moment of each station s s And simulate the running trajectory of each shared bicycle; When determining the initial number of bicycles at each station, it is necessary to ensure that all travel orders can be met; to this end, it is achieved by simulating the flow of each station at each moment; first assume that at the initial moment t0, each station The number of shared bicycles is At each time step t, the number of shared bikes at each station is updated and recorded based on the order data In the formula, t0 represents the simulation start time, T represents the simulation end time, represents the set of all sites, represents the number of bicycles at the shared bicycle station s at time t; represents the number of bicycles returning to station s at time t; represents the number of bicycles departing from station s at time t; Finally, the number of shared bicycles b at the initial moment of each station s s Calculated by the following formula: When the number of vehicles at each station is determined, the running trajectory of each vehicle is simulated, that is, the sequence of orders it serves in sequence. Specifically, it is assumed that when each order is served, one vehicle will be randomly selected from all available vehicles at its starting station. Step 3: Solve the sensor-sharing bicycle station allocation problem; Firstly, the relationship between the number of sensors deployed at each station and the number of times each road section is covered is found. Then, a mixed integer linear programming model is proposed to solve the sensor-bike sharing station allocation problem, with the goal of determining the optimal distribution of sensors at each station to maximize spatial coverage. The coverage of a certain road section by a bicycle equipped with sensors at each station is described by binomial distribution: Assume that a certain shared bicycle station s has n s If a shared bicycle is equipped with a sensor, then the shared bicycles equipped with sensors at this station will The coverage is represented by c s,e ~B(n s ,p s,e ), where p s,e represents the probability that a bicycle at shared bicycle station s visits road segment e during this time period; therefore, during this simulation time period, the average number of times N that shared bicycles at station s cover road segment e is s,e It can be expressed as the expected value of the binomial distribution, as follows: Based on the above rules, the total number of given sensors N s , the following mixed integer linear programming model is established to solve the sensor-sharing bicycle station allocation problem to maximize spatial coverage: In the formula, l e represents the length of section e; p s,e represents the probability that the shared bicycles at station s cover road section e; b s represents the number of shared bicycles at station s; N s Indicates the total number of sensors; N e is an auxiliary variable, indicating the total number of times the shared bicycles equipped with sensors cover the road section e; y e is a 0-1 auxiliary variable, which takes the value 1 if the number of coverages of section e exceeds K times; n s is a decision variable, indicating the number of sensors assigned to the shared bicycle station s; parameter M takes a sufficiently large positive integer; parameter K indicates that each road section is covered K times to meet the monitoring requirements; Step 4: Actively dispatch vehicles equipped with sensors; When users choose to ride a bicycle at a bicycle station, the application software will give priority to recommending shared bicycles equipped with sensors, and give monetary incentives to users who accept the guidance.