CN112766766B - High-precision map crowdsourcing system based on optimal time-stop rule and data collection method thereof - Google Patents

High-precision map crowdsourcing system based on optimal time-stop rule and data collection method thereof Download PDF

Info

Publication number
CN112766766B
CN112766766B CN202110101606.4A CN202110101606A CN112766766B CN 112766766 B CN112766766 B CN 112766766B CN 202110101606 A CN202110101606 A CN 202110101606A CN 112766766 B CN112766766 B CN 112766766B
Authority
CN
China
Prior art keywords
crowdsourcing
time
precision map
time slice
module
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202110101606.4A
Other languages
Chinese (zh)
Other versions
CN112766766A (en
Inventor
唐洁
李东华
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
South China University of Technology SCUT
Original Assignee
South China University of Technology SCUT
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by South China University of Technology SCUT filed Critical South China University of Technology SCUT
Priority to CN202110101606.4A priority Critical patent/CN112766766B/en
Publication of CN112766766A publication Critical patent/CN112766766A/en
Application granted granted Critical
Publication of CN112766766B publication Critical patent/CN112766766B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06311—Scheduling, planning or task assignment for a person or group
    • G06Q10/063112—Skill-based matching of a person or a group to a task
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/10—Office automation; Time management
    • G06Q10/101—Collaborative creation, e.g. joint development of products or services

Landscapes

  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Engineering & Computer Science (AREA)
  • Strategic Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Economics (AREA)
  • Physics & Mathematics (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Development Economics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Educational Administration (AREA)
  • Game Theory and Decision Science (AREA)
  • Data Mining & Analysis (AREA)
  • Traffic Control Systems (AREA)

Abstract

The invention discloses a high-precision map crowdsourcing system based on an optimal time-stop rule and a data collection method thereof. The task issuing module is used for issuing a crowdsourcing task of the high-precision map; the member preference module enables selecting winning participants from the candidate participant set, receiving their data, and rewarding them; the optimal time stopping module controls stopping time of the crowdsourcing task of the high-precision map to maximize crowdsourcing utility. Based on the system, a data collection method of a high-precision map crowdsourcing system based on an optimal time-stop rule is provided. The invention provides sufficient data sources for updating the high-precision map, reduces the cost of data collection, and enables the updating of the high-precision map to be more real-time and efficient.

Description

High-precision map crowdsourcing system based on optimal time-stop rule and data collection method thereof
Technical Field
The invention relates to the technical field of urban intelligent transportation, in particular to a high-precision map crowdsourcing system based on an optimal time-stop rule and a data collection method thereof.
Background
In the current development of unmanned fire speed, perception is an important module in unmanned, and is a premise that unmanned vehicles can safely run on the road. Sensors such as cameras, lidar, etc. are the dominant means of unmanned vehicles to sense the surrounding environment. Sensing the surrounding environment by the sensor, however, presents a significant cost challenge. The high-precision map is used as an emerging sensing means to well overcome the defect of overhigh cost of the sensor, so that how to conveniently and rapidly acquire and update the high-precision map is a current main task. In order to collect high-precision map data, graphic manufacturers and unmanned vehicle manufacturers, such as hundred degrees, germany, tomtomcom and the like, have built a professional map collection vehicle team belonging to themselves, and street view image data and 3D laser point cloud data are obtained through the high-precision map collection vehicle provided with a camera and a laser radar. The stored source data is combined with manual error correction and labeling to produce and release a high-precision map through a background automatic map building process. Although the data acquisition operation precision of the professional acquisition vehicle is high and the road information acquisition is comprehensive, the cost is quite high, the number is small, the operation period is long, and the map output is limited. A method of updating a high-precision map in real time by a collection vehicle is not realistic.
Disclosure of Invention
The invention aims to overcome the defects and shortcomings of the prior art, and provides a high-precision map crowdsourcing system based on an optimal time-stop rule and a data collection method thereof, which utilize social vehicles to complete collection of road environment data in a daily driving process, solve the problem that the road environment data cannot be efficiently acquired and the high-precision map cannot be updated in real time due to the insufficient number of specialized collection vehicles, reduce the cost caused by collecting the road environment data, improve the generation efficiency of the high-precision map and enlarge the updating range of the high-precision map.
In order to achieve the above purpose, the technical scheme provided by the invention is as follows: high-precision map crowdsourcing system based on optimal time-stop rules, comprising:
the task issuing module is used for issuing a crowdsourcing task of the high-precision map;
a member preference module for enabling selection of winning participants from the candidate participant set, receiving their data, and rewarding them;
and the optimal time stopping module is used for controlling the stopping time of the crowdsourcing task of the high-precision map so as to maximize the crowdsourcing utility.
Further, the task issuing module comprises a traffic flow predicting module, a time dividing module and an issuing module, wherein:
the traffic flow prediction module is responsible for predicting the current traffic flow through the historical data of the road network;
the time division module is responsible for dispersing time into a time slice form, and calculating the time slice length and the longest execution time of the crowdsourcing task of the high-precision map according to a traffic flow prediction result;
the publishing module is responsible for publishing the high-precision map crowdsourcing task according to the time slice length, the longest execution time of the high-precision map crowdsourcing task and the expected selection number of the participants in the unit time slice, and then giving consideration.
Further, the member preference module includes a scoring module, a selection module, and a reward module, wherein:
the scoring module is responsible for calculating the score of each candidate participant according to the vehicle-mounted sensor equipment information and the credit value of the candidate participant in the current time slice;
the selection module is responsible for selecting a winning participant according to the score of each candidate participant in the current time slice and receiving the data collected by the winning participant;
the reward module is responsible for providing rewards to winning participants.
Further, the optimal time-stop module includes a crowdsourcing status update module and a stop decision module, wherein:
the crowdsourcing state updating module is responsible for updating the current accumulated and collected data quantity and updating the crowdsourcing state according to the current accumulated and collected data quantity;
and the stopping decision module is responsible for judging whether the current crowdsourcing state meets the stopping condition according to the updated crowdsourcing state and combining with the optimal time stopping rule, stopping the crowdsourcing task of the high-precision map if the current crowdsourcing state meets the stopping condition, and otherwise, continuing to enter the next time slice.
The invention also provides a data collection method of the high-precision map crowdsourcing system based on the optimal time-stop rule, which comprises the following steps:
s1, a task issuing module acquires road network historical data, wherein the road network historical data refers to historical traffic flow data, a gray prediction classical model GM (1, 1) is used for predicting current traffic flow, time is discretized into a time slice form, the time slice length and the longest execution time of a high-precision map crowdsourcing task are defined according to the traffic flow, and finally the high-precision map crowdsourcing task is issued according to the time slice length, the longest execution time of the high-precision map crowdsourcing task and the expected selection number of unit time slice participants, and a reward is given in addition;
s2, the member optimization module firstly acquires the vehicle-mounted sensor equipment information and the credit value of the candidate participants in the current time slice, calculates the score of each candidate participant according to the information, and then selects the winning participant according to the score of the candidate participant in the current time slice and the expected number of the candidate participants in the unit time slice, receives the data collected by the winning participant, and finally provides rewards for the winning participant;
and S3, the optimal time stopping module firstly updates the current accumulated data quantity, updates the crowdsourcing state according to the current accumulated data quantity, judges whether the current crowdsourcing state meets the stopping condition according to the optimal time stopping rule, stops the crowdsourcing task of the high-precision map if the stopping condition is met, and otherwise, continues to enter the next time slice.
Further, the step S1 includes the steps of:
s101, acquiring road network historical data, and predicting the current vehicle flow lambda through a gray prediction classical model GM (1, 1);
s102, discretizing the time into a time slice form, and defining a time slice length T based on the predicted current traffic flow lambda W The method comprises the following steps:
wherein n is 0 Selecting the number of participants for a unit time slice; defining the longest execution time T of the crowdsourcing task of the high-precision map as follows:
wherein T is 0 A reference value lambda of maximum execution time of crowdsourcing task of high-precision map predefined for system 0 Corresponding T predefined for a system 0 Is a vehicle flow reference value;
s103, according to the time slice length T W Maximum execution time T of high-precision map crowd-sourced task and number n of desired selections of participants per time slice 0 And defining reward b to issue high-precision map crowd-sourced task, and transmitting message formatIs less than T, T W ,n 0 B > represents that the longest execution time of the crowdsourcing task of the high-precision map is T, and T is defined by a length of T W Each time slice is intended to receive n 0 Data for each participant that wins will be rewarded b.
Further, the step S2 includes the steps of:
s201, acquiring vehicle-mounted sensor equipment information of candidate participants and credit values thereof, wherein the vehicle-mounted sensor equipment information comprises the quality and the quantity of cameras and the quality and the quantity of laser radars, and calculating the score of each candidate participant according to the quality and the quantity of the cameras, and the calculation formula of the score is as follows:
wherein s is the score of the candidate participant, r is the reputation value of the candidate participant, and represents the credibility of the data provided by the candidate participant, n c For the number of cameras to be used,for the quality of the ith camera, n l For the number of lidars, +.>The quality of the ith laser radar; in addition, the quality of the camera is defined as follows:
in the method, in the process of the invention,the pixel value of the ith camera; the quality of the lidar is defined as follows:
in the method, in the process of the invention,the number of lines of the ith laser radar;
s202, selecting the number n according to the scores of candidate participants in the current time slice and the expected number n of the participants in the unit time slice 0 To select the winning participant, which is the top n of the candidate participants in the current time slice that are most scored 0 A bit; then, receiving data collected by the winning participant;
s203, providing a reward b for each winning participant.
Further, the step S3 includes the steps of:
s301, updating the current accumulated and collected data quantity, wherein an updating formula of the current accumulated and collected data quantity is as follows:
wherein n represents a time slice number, D (n) and D (n-1) represent respectively an accumulated data amount collected until the nth time and an accumulated data amount collected until the nth-1 time, and D (n) is an accumulated data amount in the nth time slice; updating the crowdsourcing state according to the current accumulated data amount, wherein the crowdsourcing state is defined as follows:
wherein Y (n) is the crowdsourcing state of the nth time slice, e is the base number of natural logarithm, and θ is the system coefficient;
s302, defining an optimal time stopping rule as follows:
n * ＝min{n:1≤n≤M,Y(n)≥δ,ΔY(n)≤Φ}
wherein n is * The optimal stopping time of the crowdsourcing task of the high-precision map is given by the optimal stopping rule, and M is contained in the crowdsourcing task of the high-precision mapDelta is the minimum stop threshold of the crowdsourcing state of the crowdsourcing task of the high-precision map, delta Y (n) is the expected value of the crowdsourcing state gain of the next time slice under the condition that the crowdsourcing state of the nth time slice is known, phi is the cost generated by the crowdsourcing task of the high-precision map in a unit time slice, and
wherein n is 0 The number of participants desired to select for a unit time slice,an average of the amount of data provided to the winning participant, b a consideration provided to the winning participant; if the current time slice meets the optimal time-stop rule, stopping the crowdsourcing task of the high-precision map; otherwise, continuing the crowdsourcing task of the high-precision map, and entering the next time slice.
Compared with the prior art, the invention has the following advantages and beneficial effects:
1. according to the invention, road surface environment data required by updating the high-precision map is collected through crowdsourcing, the data is collected by utilizing the vehicles in society, the problem of small number of specialized collection vehicles is solved, the data collection efficiency is greatly improved, and the data collection efficiency is much smaller than that by utilizing the collection vehicles in cost.
2. According to the invention, the score of each candidate participant is calculated through the vehicle-mounted sensor equipment information and the reputation value of the candidate participant, and the winning participant is screened out according to the score, so that the quality of the collected pavement environment data is ensured, and the credibility of the collected pavement environment data is ensured.
3. According to the invention, the stopping time of the high-precision map crowdsourcing task is judged through the optimal time stopping rule, so that the benefit of the high-precision map crowdsourcing task is maximized on the premise that the data volume of the collected road surface environment data meets the requirement of the high-precision map crowdsourcing task.
Drawings
Fig. 1 is a block diagram of the various modules of the system of the present invention.
FIG. 2 is a schematic diagram of the call process of each module of the system of the present invention.
FIG. 3 is a schematic flow chart of the method of the present invention.
FIG. 4 is a schematic diagram of the operation flow of the present invention.
Detailed Description
The invention will be further illustrated with reference to specific examples.
The high-precision map crowdsourcing system based on the optimal time-stop rule provided by the embodiment is a data collection system for updating a high-precision map, and fig. 1 and fig. 2 respectively show modules contained in the system and calling processes among the modules. It comprises the following steps:
the task issuing module is used for issuing a crowdsourcing task of the high-precision map;
a member preference module for enabling selection of winning participants from the candidate participant set, receiving their data, and rewarding them;
and the optimal time stopping module is used for controlling the stopping time of the crowdsourcing task of the high-precision map so as to maximize the crowdsourcing utility.
The task issuing module comprises a traffic flow predicting module, a time dividing module and an issuing module, wherein:
the traffic flow prediction module is responsible for predicting the current traffic flow through the historical data of the road network;
the time division module is responsible for dispersing time into a time slice form, and calculating the time slice length and the longest execution time of the crowdsourcing task of the high-precision map according to a traffic flow prediction result;
the publishing module is responsible for publishing the high-precision map crowdsourcing task according to the time slice length, the longest execution time of the high-precision map crowdsourcing task and the expected selection number of the participants in the unit time slice, and then giving consideration.
The member preference module includes a scoring module, a selection module, and a reward module, wherein:
the scoring module is responsible for calculating the score of each candidate participant according to the vehicle-mounted sensor equipment information and the credit value of the candidate participant in the current time slice;
the selection module is responsible for selecting a winning participant according to the score of each candidate participant in the current time slice and receiving the data collected by the winning participant;
the reward module is responsible for providing rewards to winning participants.
The optimal time stopping module comprises a crowdsourcing state updating module and a stopping decision module, wherein:
the crowdsourcing state updating module is responsible for updating the current accumulated and collected data quantity and updating the crowdsourcing state according to the current accumulated and collected data quantity;
and the stopping decision module is responsible for judging whether the current crowdsourcing state meets the stopping condition according to the updated crowdsourcing state and combining with the optimal time stopping rule, stopping the crowdsourcing task of the high-precision map if the current crowdsourcing state meets the stopping condition, and otherwise, continuing to enter the next time slice.
As shown in fig. 3, the present embodiment also provides a data collection method of the high-precision map crowd-sourcing system based on the optimal time-stop rule, which includes the following steps:
s1, firstly, a task publishing module acquires road network historical data, wherein the road network historical data refers to historical traffic flow data, predicts current traffic flow through a gray prediction classical model GM (1, 1), then discretizes time into a time slice form, defines the time slice length and the longest execution time of a high-precision map crowdsourcing task according to the traffic flow, and finally publishes the high-precision map crowdsourcing task according to the time slice length, the longest execution time of the high-precision map crowdsourcing task and the expected selection number of unit time slice participants, and then gives consideration to the following specific processes:
s101, acquiring road network historical data, and predicting the current vehicle flow lambda through a gray prediction classical model GM (1, 1);
s102, discretizing the time into a time slice form, and defining a time slice length T based on the predicted current traffic flow lambda W The method comprises the following steps:
wherein n is 0 Selecting the number of participants for a unit time slice; defining the longest execution time T of the crowdsourcing task of the high-precision map as follows:
wherein T is 0 A reference value lambda of maximum execution time of crowdsourcing task of high-precision map predefined for system 0 Corresponding T predefined for a system 0 Is a vehicle flow reference value;
s103, according to the time slice length T W Maximum execution time T of high-precision map crowd-sourced task and number n of desired selections of participants per time slice 0 And defining reward b to issue high-precision map crowd-sourced task, wherein the transmitted message format is less than T and T W ,n 0 B > represents that the longest execution time of the crowdsourcing task of the high-precision map is T, and T is defined by a length of T W Each time slice is intended to receive n 0 Data for each participant that wins will be rewarded b.
S2, the member optimization module firstly acquires the vehicle-mounted sensor equipment information and the credit value of the candidate participants in the current time slice, calculates the score of each candidate participant according to the information, and then selects the winning participant and receives the data collected by the winning participant according to the score of the candidate participant in the current time slice and the expected selection number of the candidate participants in the unit time slice, and finally provides rewards for the winning participant, wherein the specific process is as follows:
s201, acquiring vehicle-mounted sensor equipment information of candidate participants and credit values thereof, wherein the vehicle-mounted sensor equipment information comprises the quality and the quantity of cameras and the quality and the quantity of laser radars, and calculating the score of each candidate participant according to the quality and the quantity of the cameras, and the calculation formula of the score is as follows:
wherein s is the score of the candidate participant, r is the reputation value of the candidate participant, and represents the credibility of the data provided by the candidate participant, n c For the number of cameras to be used,for the quality of the ith camera, n l For the number of lidars, +.>The quality of the ith laser radar; in addition, the quality of the camera is defined as follows:
in the method, in the process of the invention,the pixel value of the ith camera; the quality of the lidar is defined as follows:
in the method, in the process of the invention,the number of lines of the ith laser radar;
s202, selecting the number n according to the scores of candidate participants in the current time slice and the expected number n of the participants in the unit time slice 0 To select the winning participant, which is the top n of the candidate participants in the current time slice that are most scored 0 A bit; then, receiving data collected by the winning participant;
s203, providing a reward b for each winning participant.
S3, the optimal time stopping module firstly updates the current accumulated data quantity, updates the crowdsourcing state according to the current accumulated data quantity, then judges whether the current crowdsourcing state meets the stopping condition according to the optimal time stopping rule, stops the crowdsourcing task of the high-precision map if the current crowdsourcing state meets the stopping condition, and otherwise, continues to enter the next time slice, wherein the specific process is as follows:
s301, updating the current accumulated and collected data quantity, wherein an updating formula of the current accumulated and collected data quantity is as follows:
wherein n represents a time slice number, D (n) and D (n-1) represent respectively an accumulated data amount collected until the nth time and an accumulated data amount collected until the nth-1 time, and D (n) is an accumulated data amount in the nth time slice; updating the crowdsourcing state according to the current accumulated data amount, wherein the crowdsourcing state is defined as follows:
wherein Y (n) is the crowdsourcing state of the nth time slice, e is the base number of natural logarithm, and θ is the system coefficient;
s302, defining an optimal time stopping rule as follows:
n * ＝min{n:1≤n≤M,Y(n)≥δ,ΔY(n)≤Φ}
wherein n is * The optimal stopping time of the crowdsourcing task of the high-precision map is given by the optimal time-stop rule, M is the number of time slices contained in the crowdsourcing task of the high-precision map, delta is the minimum stopping threshold value of the crowdsourcing state of the crowdsourcing task of the high-precision map, delta Y (n) is the expected value of the crowdsourcing state gain of the next time slice under the condition that the crowdsourcing state of the nth time slice is known, phi is the cost generated by the crowdsourcing task of the high-precision map in a unit time slice, and
wherein n is 0 The number of participants desired to select for a unit time slice,an average of the amount of data provided to the winning participant, b a consideration provided to the winning participant; if the current time slice meets the optimal time-stop rule, stopping the crowdsourcing task of the high-precision map; otherwise, continuing the crowdsourcing task of the high-precision map, and entering the next time slice.
As shown in fig. 4, the data collection operation flow of the above-mentioned high-precision map crowdsourcing system of the present embodiment is specifically as follows:
step 1: acquiring road network historical data, wherein the road network historical data refers to historical traffic flow data, and predicting current traffic flow through a gray prediction classical model GM (1, 1);
step 2: defining a time slice length and the longest execution time of a high-precision map crowdsourcing task based on the current traffic flow, wherein the time slice is a discretized representation of time;
step 3: combining the time slice length and the longest execution time of the high-precision map crowdsourcing task, and further defining the expected number of unit time slice participants and rewards, and distributing the high-precision map crowdsourcing task by using the four elements;
step 4: entering a time slice, and calculating the score of the candidate participants in the current time slice;
step 5: sequentially selecting candidate participants with highest scores as winning participants by taking the scores as a judgment standard, wherein the selection number of the winning participants is the expected selection number of the participants in a unit time slice, and then receiving data collected by the winning participants;
step 6: providing compensation for winning participants;
step 7: updating the current accumulated data volume and the crowdsourcing state according to the data volume collected by the current time slice;
step 8: judging whether the current crowdsourcing state meets the stopping condition according to the optimal time stopping rule, and if so, ending the crowdsourcing task of the high-precision map; otherwise, entering the next time slice, and continuing to carry out the high-precision map crowdsourcing task.
The above embodiments are only preferred embodiments of the present invention, and are not intended to limit the scope of the present invention, so variations in shape and principles of the present invention should be covered.

Claims (5)

1. High-precision map crowdsourcing system based on optimal time-stop rule, which is characterized by comprising:
the task issuing module is used for issuing a crowdsourcing task of the high-precision map;
a member preference module for enabling selection of winning participants from the candidate participant set, receiving their data, and rewarding them;
the optimal time stopping module is used for controlling the stopping time of the crowdsourcing task of the high-precision map so as to maximize crowdsourcing utility;
the task issuing module comprises a traffic flow predicting module, a time dividing module and an issuing module, wherein:
the traffic flow prediction module is responsible for predicting the current traffic flow through the historical data of the road network;
the time division module is responsible for dispersing time into a time slice form, and calculating the time slice length and the longest execution time of the crowdsourcing task of the high-precision map according to a traffic flow prediction result;
the issuing module is responsible for issuing the high-precision map crowdsourcing task according to the time slice length, the longest execution time of the high-precision map crowdsourcing task and the expected selection number of the participants in a unit time slice, and giving consideration in addition;
the optimal time stopping module comprises a crowdsourcing state updating module and a stopping decision module, wherein:
the crowdsourcing state updating module is responsible for updating the current accumulated and collected data quantity and updating the crowdsourcing state according to the current accumulated and collected data quantity;
the stopping decision module is in charge of judging whether the current crowdsourcing state meets the stopping condition according to the updated crowdsourcing state and combining with the optimal time stopping rule, stopping the crowdsourcing task of the high-precision map if the current crowdsourcing state meets the stopping condition, and otherwise, continuing to enter the next time slice;
the optimal time stopping module firstly updates the current accumulated data quantity, updates the crowdsourcing state according to the current accumulated data quantity, then judges whether the current crowdsourcing state meets the stopping condition according to the optimal time stopping rule, stops the crowdsourcing task of the high-precision map if the current crowdsourcing state meets the stopping condition, and otherwise, continues to enter the next time slice, and comprises the following steps:
s301, updating the current accumulated and collected data quantity, wherein an updating formula of the current accumulated and collected data quantity is as follows:
wherein n represents a time slice number, D (n) and D (n-1) represent respectively an accumulated data amount collected until the nth time and an accumulated data amount collected until the nth-1 time, and D (n) is an accumulated data amount in the nth time slice; updating the crowdsourcing state according to the current accumulated data amount, wherein the crowdsourcing state is defined as follows:
wherein Y (n) is the crowdsourcing state of the nth time slice, e is the base number of natural logarithm, and θ is the system coefficient;
s302, defining an optimal time stopping rule as follows:
n * ＝min{n:1≤n≤M,Y(n)≥δ,ΔY(n)≤Φ}
wherein n is * The optimal stopping time of the crowdsourcing task of the high-precision map is given by the optimal time-stopping rule, M is the number of time slices contained in the crowdsourcing task of the high-precision map, delta is the minimum stopping threshold of the crowdsourcing state of the crowdsourcing task of the high-precision map, and delta Y (n) is knownUnder the condition of the crowdsourcing state of the nth time slice, the expected value of the crowdsourcing state gain of the next time slice, phi is the cost generated by the crowdsourcing task of the high-precision map in the unit time slice, and
wherein n is 0 The number of participants desired to select for a unit time slice,an average of the amount of data provided to the winning participant, b a consideration provided to the winning participant; if the current time slice meets the optimal time-stop rule, stopping the crowdsourcing task of the high-precision map; otherwise, continuing the crowdsourcing task of the high-precision map, and entering the next time slice.
2. The optimal time-stop rule-based high-precision map crowd-sourcing system of claim 1, wherein: the member preference module includes a scoring module, a selection module, and a reward module, wherein:
the scoring module is responsible for calculating the score of each candidate participant according to the vehicle-mounted sensor equipment information and the credit value of the candidate participant in the current time slice;
the selection module is responsible for selecting a winning participant according to the score of each candidate participant in the current time slice and receiving the data collected by the winning participant;
the reward module is responsible for providing rewards to winning participants.
3. The data collection method of the high-precision map crowd-sourcing system based on the optimal time-stop rule as claimed in claim 1 or 2, comprising the following steps:
s1, a task issuing module acquires road network historical data, wherein the road network historical data refers to historical traffic flow data, a gray prediction classical model GM (1, 1) is used for predicting current traffic flow, time is discretized into a time slice form, the time slice length and the longest execution time of a high-precision map crowdsourcing task are defined according to the traffic flow, and finally the high-precision map crowdsourcing task is issued according to the time slice length, the longest execution time of the high-precision map crowdsourcing task and the expected selection number of unit time slice participants, and a reward is given in addition;
s2, the member optimization module firstly acquires the vehicle-mounted sensor equipment information and the credit value of the candidate participants in the current time slice, calculates the score of each candidate participant according to the information, and then selects the winning participant according to the score of the candidate participant in the current time slice and the expected number of the candidate participants in the unit time slice, receives the data collected by the winning participant, and finally provides rewards for the winning participant;
s3, the optimal time stopping module firstly updates the current accumulated data quantity, updates the crowdsourcing state according to the current accumulated data quantity, judges whether the current crowdsourcing state meets the stopping condition according to the optimal time stopping rule, stops the crowdsourcing task of the high-precision map if the current crowdsourcing state meets the stopping condition, and otherwise, continues to enter the next time slice, and comprises the following steps:
s301, updating the current accumulated and collected data quantity, wherein an updating formula of the current accumulated and collected data quantity is as follows:
wherein n represents a time slice number, D (n) and D (n-1) represent respectively an accumulated data amount collected until the nth time and an accumulated data amount collected until the nth-1 time, and D (n) is an accumulated data amount in the nth time slice; updating the crowdsourcing state according to the current accumulated data amount, wherein the crowdsourcing state is defined as follows:
wherein Y (n) is the crowdsourcing state of the nth time slice, e is the base number of natural logarithm, and θ is the system coefficient;
s302, defining an optimal time stopping rule as follows:
n * ＝min{n:1≤n≤M,Y(n)≥δ,ΔY(n)≤Φ}
wherein n is * The optimal stopping time of the crowdsourcing task of the high-precision map is given by the optimal time-stop rule, M is the number of time slices contained in the crowdsourcing task of the high-precision map, delta is the minimum stopping threshold value of the crowdsourcing state of the crowdsourcing task of the high-precision map, delta Y (n) is the expected value of the crowdsourcing state gain of the next time slice under the condition that the crowdsourcing state of the nth time slice is known, phi is the cost generated by the crowdsourcing task of the high-precision map in a unit time slice, and
wherein n is 0 The number of participants desired to select for a unit time slice,an average of the amount of data provided to the winning participant, b a consideration provided to the winning participant; if the current time slice meets the optimal time-stop rule, stopping the crowdsourcing task of the high-precision map; otherwise, continuing the crowdsourcing task of the high-precision map, and entering the next time slice.
4. The data collection method of the high-precision map crowd-sourcing system based on the optimal time-stop rule according to claim 3, wherein the step S1 comprises the following steps:
s101, acquiring road network historical data, and predicting the current vehicle flow lambda through a gray prediction classical model GM (1, 1);
s102, discretizing the time into a time slice form, and defining a time slice length T based on the predicted current traffic flow lambda W The method comprises the following steps:
wherein n is 0 Selecting the number of participants for a unit time slice; defining the longest execution time T of the crowdsourcing task of the high-precision map as follows:
wherein T is 0 A reference value lambda of maximum execution time of crowdsourcing task of high-precision map predefined for system 0 Corresponding T predefined for a system 0 Is a vehicle flow reference value;
s103, according to the time slice length T W Maximum execution time T of high-precision map crowd-sourced task and number n of desired selections of participants per time slice 0 And defining reward b to issue high-precision map crowd-sourced task, wherein the transmitted message format is that<T,T W ,n 0 ,b>Representing that the longest execution time of the crowdsourcing task of the high-precision map is T, and T is represented by length T W Each time slice is intended to receive n 0 Data for each participant that wins will be rewarded b.
5. The data collection method of the high-precision map crowd-sourcing system based on the optimal time-stop rule of claim 3, wherein the step S2 comprises the following steps:
s201, acquiring vehicle-mounted sensor equipment information of candidate participants and credit values thereof, wherein the vehicle-mounted sensor equipment information comprises the quality and the quantity of cameras and the quality and the quantity of laser radars, and calculating the score of each candidate participant according to the quality and the quantity of the cameras, and the calculation formula of the score is as follows:
where s is the score of the candidate participant and r is the reputation value of the candidate participantRepresenting the credibility, n, of the data provided by the candidate participants c For the number of cameras to be used,for the quality of the ith camera, n l For the number of lidars, +.>The quality of the ith laser radar; in addition, the quality of the camera is defined as follows:
in the method, in the process of the invention,the pixel value of the ith camera; the quality of the lidar is defined as follows:
in the method, in the process of the invention,the number of lines of the ith laser radar;
s202, selecting the number n according to the scores of candidate participants in the current time slice and the expected number n of the participants in the unit time slice 0 To select the winning participant, which is the top n of the candidate participants in the current time slice that are most scored 0 A bit; then, receiving data collected by the winning participant;
s203, providing a reward b for each winning participant.
CN202110101606.4A 2021-01-26 2021-01-26 High-precision map crowdsourcing system based on optimal time-stop rule and data collection method thereof Active CN112766766B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110101606.4A CN112766766B (en) 2021-01-26 2021-01-26 High-precision map crowdsourcing system based on optimal time-stop rule and data collection method thereof

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110101606.4A CN112766766B (en) 2021-01-26 2021-01-26 High-precision map crowdsourcing system based on optimal time-stop rule and data collection method thereof

Publications (2)

Publication Number Publication Date
CN112766766A CN112766766A (en) 2021-05-07
CN112766766B true CN112766766B (en) 2024-04-09

Family

ID=75707422

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110101606.4A Active CN112766766B (en) 2021-01-26 2021-01-26 High-precision map crowdsourcing system based on optimal time-stop rule and data collection method thereof

Country Status (1)

Country Link
CN (1) CN112766766B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113503869B (en) * 2021-06-25 2023-06-20 华南理工大学 High-precision map crowdsourcing vehicle selection method based on regional perceptibility

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107103408A (en) * 2017-03-24 2017-08-29 扬州大学 Complex task distribution method under a kind of mass-rent environment
CN107784561A (en) * 2017-11-10 2018-03-09 东南大学 The implementation method of online incentive mechanism in a kind of mobile mass-rent system
CN108364125A (en) * 2018-01-30 2018-08-03 华南理工大学 A kind of vehicle moving crowdsourcing multiple target task distribution method
CN109902903A (en) * 2017-12-11 2019-06-18 华东师范大学 A kind of crowdsourcing execution method and crowdsourcing system based on wechat
CN110110962A (en) * 2019-04-02 2019-08-09 华南理工大学 A kind of task gunz executes the preferred method of team
CN110160544A (en) * 2019-06-12 2019-08-23 北京深思敏行科技有限责任公司 A kind of high-precision map crowdsourcing more new system based on edge calculations
CN111428932A (en) * 2020-03-27 2020-07-17 中国民航大学 Medium-and-long-term air traffic flow prediction method based on wavelet transformation and gray prediction
CN111915185A (en) * 2020-07-31 2020-11-10 湖北大学 Space-time crowdsourcing task allocation method and device based on path planning strategy

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140278634A1 (en) * 2013-03-15 2014-09-18 Microsoft Corporation Spatiotemporal Crowdsourcing
US20180260902A1 (en) * 2017-03-11 2018-09-13 Michael John Curran System and method for determining an optimal stopping time

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107103408A (en) * 2017-03-24 2017-08-29 扬州大学 Complex task distribution method under a kind of mass-rent environment
CN107784561A (en) * 2017-11-10 2018-03-09 东南大学 The implementation method of online incentive mechanism in a kind of mobile mass-rent system
CN109902903A (en) * 2017-12-11 2019-06-18 华东师范大学 A kind of crowdsourcing execution method and crowdsourcing system based on wechat
CN108364125A (en) * 2018-01-30 2018-08-03 华南理工大学 A kind of vehicle moving crowdsourcing multiple target task distribution method
CN110110962A (en) * 2019-04-02 2019-08-09 华南理工大学 A kind of task gunz executes the preferred method of team
CN110160544A (en) * 2019-06-12 2019-08-23 北京深思敏行科技有限责任公司 A kind of high-precision map crowdsourcing more new system based on edge calculations
CN111428932A (en) * 2020-03-27 2020-07-17 中国民航大学 Medium-and-long-term air traffic flow prediction method based on wavelet transformation and gray prediction
CN111915185A (en) * 2020-07-31 2020-11-10 湖北大学 Space-time crowdsourcing task allocation method and device based on path planning strategy

Also Published As

Publication number Publication date
CN112766766A (en) 2021-05-07

Similar Documents

Publication Publication Date Title
JP2021531462A (en) Intelligent navigation methods and systems based on topology maps
CN108458716A (en) A kind of electric vehicle charging air navigation aid based on the prediction of charging pile dynamic occupancy
CN111081020B (en) Vehicle-mounted traffic accident early warning device based on cloud edge combination
JP5509666B2 (en) Radio wave propagation characteristic estimation support system, radio wave propagation characteristic estimation support method, and radio wave propagation characteristic estimation support apparatus
CN104851280B (en) A kind of vehicle travel control method, device, system and relevant device
WO2022021062A1 (en) Remaining range prediction method and battery remote service system
CN112766766B (en) High-precision map crowdsourcing system based on optimal time-stop rule and data collection method thereof
CN114757807A (en) Multi-mode fused online accounting method for actual emission of atmospheric pollutants
CN112710798B (en) Water body blue algae identification system and method
CN116029411A (en) Smart city area management method and system based on Internet of things
US20130261946A1 (en) Vehicle information management server, vehicle information management system, and vehicle information management method
CN111341093B (en) Control method, device, equipment and storage medium for motorcade
CN114971290B (en) Park management system and method based on intelligent street lamp
CN112785162A (en) High-precision map crowdsourcing data quality assessment method and system based on intelligent vehicle semantics
JP2012066705A (en) System and method for providing information
CN116432985B (en) Intelligent community information processing method and system based on Internet of things
CN110019376B (en) Driving comparison method for electric vehicle rental system
CN110879862A (en) Vehicle information query system based on Internet of things
CN116229424A (en) Intelligent networking automobile pedestrian detection method based on improved YOLO v3
CN113503869B (en) High-precision map crowdsourcing vehicle selection method based on regional perceptibility
CN113269339B (en) Method and system for automatically creating and distributing network appointment tasks
WO2023060586A1 (en) Automatic driving instruction generation model optimization method and apparatus, device, and storage medium
CN110164127B (en) Traffic flow prediction method and device and server
CN114861220A (en) Automatic driving data processing method and system conforming to data privacy security
CN114187759A (en) Road side unit driving assistance method and device based on data driving model

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant