CN106934490B - AGV call prediction method and device - Google Patents

AGV call prediction method and device Download PDF

Info

Publication number
CN106934490B
CN106934490B CN201710096372.2A CN201710096372A CN106934490B CN 106934490 B CN106934490 B CN 106934490B CN 201710096372 A CN201710096372 A CN 201710096372A CN 106934490 B CN106934490 B CN 106934490B
Authority
CN
China
Prior art keywords
call
calling
point
predicted
probability
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
CN201710096372.2A
Other languages
Chinese (zh)
Other versions
CN106934490A (en
Inventor
韩志
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.)
Guangzhou Shiyuan Electronics Thecnology Co Ltd
Original Assignee
Guangzhou Shiyuan Electronics Thecnology Co Ltd
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 Guangzhou Shiyuan Electronics Thecnology Co Ltd filed Critical Guangzhou Shiyuan Electronics Thecnology Co Ltd
Priority to CN201710096372.2A priority Critical patent/CN106934490B/en
Publication of CN106934490A publication Critical patent/CN106934490A/en
Application granted granted Critical
Publication of CN106934490B publication Critical patent/CN106934490B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

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
    • G06Q10/04—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
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • 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
    • G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Entrepreneurship & Innovation (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)
  • Game Theory and Decision Science (AREA)
  • Educational Administration (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses an AGV calling prediction method, which comprises the following steps: inquiring a call probability database according to the interval duration between the last call time of each call point and the prediction time to obtain the call probability of each call point at the prediction time; and predicting the calling points at the predicted time according to the calling probability of each calling point at the predicted time. Correspondingly, the invention also discloses an AGV calling prediction device. The method can intelligently predict which calling point the AGV calls at a certain moment can be initiated by, so that the AGV can move to the vicinity of the calling point in advance to prepare, the time for reaching the calling point is shortened, and the working efficiency is improved.

Description

AGV call prediction method and device
Technical Field
The invention relates to the field of automatic transportation, in particular to an AGV calling prediction method and an AGV calling prediction device.
Background
An AGV (Automated Guided Vehicle) is a Vehicle capable of performing automatic navigation and obstacle avoidance according to a scheduling requirement. Can be widely applied to various enterprises such as factories, trades, transportation and the like. Compared with other equipment commonly used in material conveying, the AGV has the advantages that fixing devices such as rails and supporting frames do not need to be laid in the moving area of the AGV, and the AGV is not limited by sites, roads and spaces. Therefore, in the automatic logistics system, the automation and the flexibility can be fully embodied, and the efficient, economical and flexible unmanned production is realized.
The AGV dispatching is generally initiated according to the requirements of a calling area and has the characteristic of variable time, and the AGV dispatching area is always away from the calling area by a certain distance, so that the time for reaching a calling point is long, and the working efficiency is influenced.
Disclosure of Invention
The invention aims to provide an AGV call prediction method and an AGV call prediction device, which can intelligently predict which call point an AGV call is initiated at a certain moment, so that the AGV can move to the vicinity of the call point in advance to prepare, the time for reaching the call point is shortened, and the working efficiency is improved.
In order to achieve the above object, an aspect of the present invention provides an AGV call prediction method, including:
inquiring a call probability database according to the interval duration between the last call time of each call point and the prediction time to obtain the call probability of each call point at the prediction time; the call probability database is configured according to the historical data of the AGV work and is used for recording the mapping relation between the call interval duration and the call probability;
and predicting the calling points at the predicted time according to the calling probability of each calling point at the predicted time.
The embodiment of the invention has the following beneficial effects:
according to the AGV call prediction method provided by the embodiment of the invention, the mapping relation between the call interval duration and the call probability is obtained by counting the historical data, so that the call point at the prediction moment is predicted according to the call interval duration, the AGV can move to the vicinity of the call point in advance to prepare, the time for reaching the call point is shortened, the AGV operation efficiency is improved, and the enterprise work efficiency is improved.
Further, after the predicting the call points at the predicted time according to the call probability of each call point at the predicted time, the method further includes:
judging whether the predicted calling point is the calling point called at the predicted moment;
if the predicted calling point is the calling point called at the predicted time, increasing the predicted correct times of the predicted calling point once, and increasing the predicted times of the predicted calling point once;
if the predicted calling point is not the calling point called at the predicted time, increasing the predicted times of the predicted calling point once;
predicting the call points at the predicted time according to the call probability of each call point at the predicted time, which specifically comprises:
calculating the prediction accuracy of each calling point according to the accumulated predicted times and the accumulated correct predicted times of each calling point;
and predicting the calling points at the predicted time by combining the calling probability and the prediction accuracy of each calling point at the predicted time.
Further, the predicting the call point at the predicted time by combining the call probability and the prediction accuracy of each call point at the predicted time specifically includes:
respectively calculating the product of the calling probability and the prediction accuracy of each calling point at the prediction moment as the weighted probability of each calling point;
and comparing the weighted probability of each calling point with a preset threshold value, and selecting the calling point with the weighted probability greater than the preset threshold value as a predicted calling point.
Further, the predicting the call point at the predicted time by combining the call probability and the prediction accuracy of each call point at the predicted time specifically includes:
respectively calculating the product of the calling probability and the prediction accuracy of each calling point at the prediction moment as the weighted probability of each calling point;
judging whether the maximum weighted probability is greater than a preset threshold value or not; and if so, selecting the calling point with the highest weighted probability as the predicted calling point.
Further, the method for configuring the call probability database includes:
extracting the call time record of each call point from the historical data of the AGV work;
according to the call time records, respectively counting the call times corresponding to all call interval durations for each call point; wherein, the number of calls with the interval duration j of the ith call point is Nij；
Accumulating the sum of the calling times of all calling points with the same calling interval duration; wherein, the sum of the calling times of all calling points with the calling interval duration being j is Mj；
Will Nij/MjRecording the call probability of the ith call point call interval duration j.
Another aspect of an embodiment of the present invention provides an AGV call prediction apparatus, including:
the query module is used for querying a call probability database according to the interval duration between the last call time of each call point and the prediction time to obtain the call probability of each call point at the prediction time; the call probability database is configured according to the historical data of the AGV work and is used for recording the mapping relation between the call interval duration and the call probability;
and the prediction module is used for predicting the calling points at the prediction time according to the calling probability of each calling point at the prediction time.
According to the AGV call prediction device provided by the embodiment of the invention, the mapping relation between the call interval duration and the call probability is obtained by counting the historical data, so that the call point at the prediction moment is predicted according to the call interval duration, the AGV can move to the vicinity of the call point in advance to prepare, the time for reaching the call point is shortened, the AGV operation efficiency is improved, and the enterprise work efficiency is improved.
Further, still include:
the judging module is used for judging whether the predicted calling point is the calling point called at the predicted moment;
a first accumulation module for increasing a predicted correct number of times of a predicted call point and increasing a predicted number of times of the predicted call point, if the predicted call point is a call point called at a predicted time;
a second accumulation module for increasing the predicted number of times of the predicted call point once if the predicted call point is not the call point called at the predicted time;
the prediction module comprises:
the accuracy calculation unit is used for calculating the prediction accuracy of each calling point according to the accumulated predicted times and the accumulated correct prediction times of each calling point;
and the comprehensive prediction unit is used for predicting the calling points at the predicted time by combining the calling probability and the prediction accuracy of each calling point at the predicted time.
Further, the comprehensive prediction unit includes:
the product calculating subunit is used for calculating the product of the calling probability and the prediction accuracy of each calling point at the prediction moment as the weighted probability of each calling point;
and the first prediction subunit is used for comparing the weighted probability of each calling point with a preset threshold value and selecting the calling point with the weighted probability greater than the preset threshold value as the predicted calling point.
Further, the comprehensive prediction unit includes:
the product calculating subunit is used for calculating the product of the calling probability and the prediction accuracy of each calling point at the prediction moment as the weighted probability of each calling point;
the second prediction subunit is used for judging whether the maximum weighted probability is greater than a preset threshold value or not; and if so, selecting the calling point with the highest weighted probability as the predicted calling point.
Further, the configuration device of the call probability database includes:
the extracting unit is used for extracting the call time record of each call point from the history data of the AGV work;
the statistical unit is used for respectively counting the calling times corresponding to all the calling interval durations for each calling point according to the calling time records; wherein, the number of calls with the interval duration j of the ith call point is Nij；
The accumulation unit is used for accumulating the sum of the calling times of all calling points in the same calling interval duration; wherein, the sum of the calling times of all calling points with the calling interval duration being j is Mj；
A recording unit for recording Nij/MjRecording the call probability of the ith call point call interval duration j.
Drawings
FIG. 1 is a flowchart of an AGV call prediction method according to an embodiment of the present invention;
fig. 2 is a block diagram showing an AGV call prediction apparatus according to a second embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, a flowchart of an AGV call prediction method according to an embodiment of the present invention is shown; the method comprises the following steps:
s11, inquiring a call probability database according to the interval duration between the last call time and the prediction time of each call point, and obtaining the call probability of each call point at the prediction time; the call probability database is configured according to the historical data of the AGV work and is used for recording the mapping relation between the call interval duration and the call probability;
and S12, predicting the calling points at the predicted time according to the calling probability of each calling point at the predicted time.
By the AGV calling prediction method, the calling point at the prediction moment is predicted, the AGV can move to the position near the predicted calling point in advance, the time for reaching the calling point is shortened, the AGV running efficiency is improved, and the working efficiency of a factory is indirectly improved. And the AGV knows the next destination of operation in advance, can use more economic speed motion, can improve AGV continuation of the journey mileage when the energy saving, reduces AGV input quantity. When the calling probability of the AGV is low in a few minutes, the AGV can move to a charging area, automatic charging is started, and the endurance mileage of the AGV is improved. After the carrying efficiency of an enterprise is improved, the utilization rate of the corresponding working section storage area is also improved, the site investment cost of the enterprise can be reduced, and the average output value per square meter is improved.
The configuration method of the call probability database specifically comprises the following steps:
extracting the call time record of each call point from the historical data of the AGV work;
according to the call time records, respectively counting the call times corresponding to all call interval durations for each call point; wherein, the number of calls with the interval duration j of the ith call point is Nij；
Accumulating the calling times of all calling points in the same calling interval durationThe sum of the numbers; wherein, the sum of the calling times of all calling points with the calling interval duration being j is Mj；
Will Nij/MjRecording the call probability of the ith call point call interval duration j.
In specific implementation, a statistical period is set first, for example, the statistical period is one week, which means that only data of the last 7 days are counted. For example, when the data of number 8 is generated from the data of number 1 to 7, only the data of number 2 to 8 are counted, and so on, as an independent statistical dimension each day. During the statistical period, the system automatically records the call data when generating a call, as shown in the following table:
date Time Starting point Terminal point
2016/11/2 1:20 A B
2016/11/2 1:30 B C
2016/11/2 1:50 A D
…… …… …… ……
According to the above table, the number of calls corresponding to all the call interval durations is counted for each call point, for example, the number of calls with the call interval duration of 10 minutes using the point a as the starting point is 0, the number of calls with the call interval duration of 20 minutes is 5, the number of calls with the call interval duration of 30 minutes is 30, and the like.
The call probability is obtained by normalizing the statistical results. The normalization methods are many, and the data of the same statistical dimension are accumulated to obtain the total number, and the total number is divided by each item in the dimension to obtain the normalized calling probability. In this embodiment, the number of times of the same interval duration is accumulated to obtain a total number, and the number of times of each call point is divided by the total number to obtain the call probability of the specific interval duration of each call point. In other embodiments, the number of times of different interval durations of the same call point may be accumulated to obtain a total number, and the number of times of each interval duration is divided by the total number to obtain the call probability of the specific interval duration of each call point.
In this embodiment, when the call probability database is configured, the case that the call probabilities of different call points at a specific interval duration are inconsistent is considered, two-dimensional data is used for the statistics of the call probabilities, further, based on the production rule of an enterprise, more statistical dimensions can be set, such as product types and different product types, different call points have different probabilities at the specific interval duration, the dimension of the product types is added in the process of configuring the database, and when the database is subsequently queried, the database is queried according to the interval duration of the last call time and the predicted time of each call point and the current product types of the call points, so as to obtain the call probabilities corresponding to each call point.
Further, in step S12, after predicting the call point at the predicted time based on the call probability of each call point at the predicted time, the method further includes:
s13, judging whether the predicted calling point is the calling point called at the predicted time;
s14, if the predicted calling point is the calling point calling at the predicted time, increasing the predicted correct times of the predicted calling point once, and increasing the predicted times of the predicted calling point once;
s15, if the predicted calling point is not the calling point called at the predicted time, increasing the predicted times of the predicted calling point;
predicting the call points at the predicted time according to the call probability of each call point at the predicted time, which specifically comprises:
calculating the prediction accuracy of each calling point according to the accumulated predicted times and the accumulated correct predicted times of each calling point;
and predicting the calling points at the predicted time by combining the calling probability and the prediction accuracy of each calling point at the predicted time.
Specifically, the accumulated predicted number of times and the accumulated predicted correct number of times may be an accumulated value of the predicted number of times and the predicted correct number of times in a statistical cycle of the configuration database, an accumulated value of the predicted number of times and the predicted correct number of times in a statistical cycle (that is, a period of time in a previous statistical cycle with a predicted time as an end point) of real-time scrolling, or an accumulated value of a statistical cycle with a length different from that of the configuration database.
Specifically, in an embodiment, the predicting the call point at the predicted time by combining the call probability and the prediction accuracy of each call point at the predicted time specifically includes:
respectively calculating the product of the calling probability and the prediction accuracy of each calling point at the prediction moment as the weighted probability of each calling point;
and comparing the weighted probability of each calling point with a preset threshold value, and selecting the calling point with the weighted probability greater than the preset threshold value as a predicted calling point.
In this embodiment, the call point at the predicted time is predicted synthetically by incorporating the prediction accuracy. For example, the predicted time is 3:15, the last call time at point a is 5 minutes from the predicted time, point B is 10 minutes, and point C is 15 minutes. Inquiring a call probability database to know that the call probability of the call interval duration of the point A is 5 minutes is 0.3, the call probability of the call interval duration of the point B is 10 minutes is 0.7, and the call probability of the call interval duration of the point C is 15 minutes is 0.8. If all the prediction accuracy rates are assumed to be 100%, the current prediction result is the point C, and if the prediction accuracy rate of the point C is 50%. The weighted probability is a: 0.3, B: 0.7, C: 0.48. if the weighted probability is higher than a preset threshold (e.g., 0.6), i.e., B point, the call point is regarded as the predicted call point. If no calling point with the weighted probability higher than the preset threshold value exists, the prediction does not generate the predicted calling point, and correspondingly, the prediction correct times and the predicted times are not accumulated, so that the prediction accuracy is controlled. This embodiment allows multiple call point calls to be predicted, which is suitable for situations where multiple call points may be called simultaneously during enterprise production.
In another embodiment, the predicting the call point at the predicted time by combining the call probability and the prediction accuracy of each call point at the predicted time specifically includes:
respectively calculating the product of the calling probability and the prediction accuracy of each calling point at the prediction moment as the weighted probability of each calling point;
judging whether the maximum weighted probability is greater than a preset threshold value or not; and if so, selecting the calling point with the highest weighted probability as the predicted calling point.
The implementation mode is suitable for the condition that only one calling point call is predicted in the enterprise production process, and for the condition that only one calling point call is predicted, the maximum weighting probability is extracted firstly, and whether the maximum weighting probability is larger than a preset threshold value or not is only judged, so that the calculation amount is reduced.
According to the AGV call prediction method provided by the embodiment of the invention, the mapping relation between the call interval duration and the call probability is obtained by counting the historical data, so that the call point at the prediction moment is predicted according to the call interval duration, the AGV can move to the vicinity of the call point in advance to prepare, the time for reaching the call point is shortened, the AGV operation efficiency is improved, and the enterprise work efficiency is improved.
Referring to fig. 2, a block diagram of an AGV call prediction apparatus according to a second embodiment of the present invention is shown. The AGV calling prediction apparatus includes:
the query module 21 is configured to query a call probability database according to the time interval between the last call time of each call point and the predicted time, and obtain the call probability of each call point at the predicted time; the call probability database is configured according to the historical data of the AGV work and is used for recording the mapping relation between the call interval duration and the call probability;
and the prediction module 22 is configured to predict the call point at the prediction time according to the call probability of each call point at the prediction time.
Further, still include:
the judging module is used for judging whether the predicted calling point is the calling point called at the predicted moment;
a first accumulation module for increasing a predicted correct number of times of a predicted call point and increasing a predicted number of times of the predicted call point, if the predicted call point is a call point called at a predicted time;
a second accumulation module for increasing the predicted number of times of the predicted call point once if the predicted call point is not the call point called at the predicted time;
the prediction module comprises:
the accuracy calculation unit is used for calculating the prediction accuracy of each calling point according to the accumulated predicted times and the accumulated correct prediction times of each calling point;
and the comprehensive prediction unit is used for predicting the calling points at the predicted time by combining the calling probability and the prediction accuracy of each calling point at the predicted time.
Further, the comprehensive prediction unit includes:
the product calculating subunit is used for calculating the product of the calling probability and the prediction accuracy of each calling point at the prediction moment as the weighted probability of each calling point;
and the first prediction subunit is used for comparing the weighted probability of each calling point with a preset threshold value and selecting the calling point with the weighted probability greater than the preset threshold value as the predicted calling point.
Further, the comprehensive prediction unit includes:
the product calculating subunit is used for calculating the product of the calling probability and the prediction accuracy of each calling point at the prediction moment as the weighted probability of each calling point;
the second prediction subunit is used for judging whether the maximum weighted probability is greater than a preset threshold value or not; and if so, selecting the calling point with the highest weighted probability as the predicted calling point.
Further, the configuration device of the call probability database includes:
the extracting unit is used for extracting the call time record of each call point from the history data of the AGV work;
the statistical unit is used for respectively counting the calling times corresponding to all the calling interval durations for each calling point according to the calling time records; wherein, the number of calls with the interval duration j of the ith call point is Nij；
The accumulation unit is used for accumulating the sum of the calling times of all calling points in the same calling interval duration; wherein, the sum of the calling times of all calling points with the calling interval duration being j is Mj；
A recording unit for recording Nij/MjRecording the call probability of the ith call point call interval duration j.
According to the AGV call prediction device provided by the embodiment of the invention, the mapping relation between the call interval duration and the call probability is obtained by counting the historical data, so that the call point at the prediction moment is predicted according to the call interval duration, the AGV can move to the vicinity of the call point in advance to prepare, the time for reaching the call point is shortened, the AGV operation efficiency is improved, and the enterprise work efficiency is improved.
The foregoing is a preferred embodiment of the present invention, and it should be noted that modifications and variations can be made by those skilled in the art without departing from the principle of the present invention, and these modifications and variations are also considered as the protection scope of the present invention.

Claims (8)

1. An AGV call prediction method, comprising:
inquiring a call probability database according to the interval duration between the last call time of each call point and the prediction time to obtain the call probability of each call point at the prediction time; the call probability database is configured according to the historical data of the AGV work and is used for recording the mapping relation between the call interval duration and the call probability;
predicting the calling points at the predicted time according to the calling probability of each calling point at the predicted time;
judging whether the predicted calling point is the calling point called at the predicted moment;
if the predicted calling point is the calling point called at the predicted time, increasing the predicted correct times of the predicted calling point once, and increasing the predicted times of the predicted calling point once;
if the predicted calling point is not the calling point called at the predicted time, increasing the predicted times of the predicted calling point once;
the predicting the call points at the predicted time according to the call probability of each call point at the predicted time specifically comprises the following steps:
calculating the prediction accuracy of each calling point according to the accumulated predicted times and the accumulated correct predicted times of each calling point;
and predicting the calling points at the predicted time by combining the calling probability and the prediction accuracy of each calling point at the predicted time.
2. The AGV call prediction method according to claim 1, wherein the predicting the call point at the predicted time in combination with the call probability and the prediction accuracy of each call point at the predicted time includes:
respectively calculating the product of the calling probability and the prediction accuracy of each calling point at the prediction moment as the weighted probability of each calling point;
and comparing the weighted probability of each calling point with a preset threshold value, and selecting the calling point with the weighted probability greater than the preset threshold value as a predicted calling point.
3. The AGV call prediction method according to claim 1, wherein the predicting the call point at the predicted time in combination with the call probability and the prediction accuracy of each call point at the predicted time includes:
respectively calculating the product of the calling probability and the prediction accuracy of each calling point at the prediction moment as the weighted probability of each calling point;
judging whether the maximum weighted probability is greater than a preset threshold value or not; and if so, selecting the calling point with the highest weighted probability as the predicted calling point.
4. An AGV call prediction method according to any of claims 1 to 3 wherein said method of configuring said call probability database comprises:
extracting the call time record of each call point from the historical data of the AGV work;
according to the call time records, respectively counting the call times corresponding to all call interval durations for each call point; wherein, the number of calls with the interval duration j of the ith call point is Nij；
Accumulating the sum of the calling times of all calling points with the same calling interval duration; wherein, the sum of the calling times of all calling points with the calling interval duration being j is Mj；
Will Nij/MjRecording the call probability of the ith call point call interval duration j.
5. An AGV call prediction apparatus comprising:
the query module is used for querying a call probability database according to the interval duration between the last call time of each call point and the prediction time to obtain the call probability of each call point at the prediction time; the call probability database is configured according to the historical data of the AGV work and is used for recording the mapping relation between the call interval duration and the call probability;
the prediction module is used for predicting the calling points at the prediction time according to the calling probability of each calling point at the prediction time;
the judging module is used for judging whether the predicted calling point is the calling point called at the predicted moment;
a first accumulation module for increasing a predicted correct number of times of a predicted call point and increasing a predicted number of times of the predicted call point, if the predicted call point is a call point called at a predicted time;
a second accumulation module for increasing the predicted number of times of the predicted call point once if the predicted call point is not the call point called at the predicted time;
wherein the prediction module comprises:
the accuracy calculation unit is used for calculating the prediction accuracy of each calling point according to the accumulated predicted times and the accumulated correct prediction times of each calling point;
and the comprehensive prediction unit is used for predicting the calling points at the predicted time by combining the calling probability and the prediction accuracy of each calling point at the predicted time.
6. The AGV call prediction device of claim 5 wherein the integrated prediction unit comprises:
the product calculating subunit is used for calculating the product of the calling probability and the prediction accuracy of each calling point at the prediction moment as the weighted probability of each calling point;
and the first prediction subunit is used for comparing the weighted probability of each calling point with a preset threshold value and selecting the calling point with the weighted probability greater than the preset threshold value as the predicted calling point.
7. The AGV call prediction device of claim 5 wherein the integrated prediction unit comprises:
the product calculating subunit is used for calculating the product of the calling probability and the prediction accuracy of each calling point at the prediction moment as the weighted probability of each calling point;
the second prediction subunit is used for judging whether the maximum weighted probability is greater than a preset threshold value or not; and if so, selecting the calling point with the highest weighted probability as the predicted calling point.
8. An AGV call prediction apparatus according to any of claims 5 to 7 wherein said means for configuring said call probability database includes:
the extracting module is used for extracting the call time record of each call point from the historical data of the AGV work;
the counting module is used for respectively counting the calling times corresponding to all the calling interval durations for each calling point according to the calling time records; wherein, the number of calls with the interval duration j of the ith call point is Nij；
The accumulation module is used for accumulating the sum of the calling times of all calling points in the same calling interval duration; wherein, the sum of the calling times of all calling points with the calling interval duration being j is Mj；
A recording module for converting Nij/MjRecording the call probability of the ith call point call interval duration j.
CN201710096372.2A 2017-02-22 2017-02-22 AGV call prediction method and device Active CN106934490B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710096372.2A CN106934490B (en) 2017-02-22 2017-02-22 AGV call prediction method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710096372.2A CN106934490B (en) 2017-02-22 2017-02-22 AGV call prediction method and device

Publications (2)

Publication Number Publication Date
CN106934490A CN106934490A (en) 2017-07-07
CN106934490B true CN106934490B (en) 2021-05-07

Family

ID=59424529

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710096372.2A Active CN106934490B (en) 2017-02-22 2017-02-22 AGV call prediction method and device

Country Status (1)

Country Link
CN (1) CN106934490B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107734200B (en) * 2017-11-03 2019-08-13 中国人民解放军信息工程大学 A kind of communication network users calling behavior prediction method and device based on maximum likelihood
CN114841624A (en) * 2022-06-16 2022-08-02 中国平安人寿保险股份有限公司 Scheduling model training method, device, equipment and medium based on artificial intelligence
CN116300979B (en) * 2023-05-26 2023-08-01 君华高科集团有限公司 Robot cruise path generation system and method

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105354648A (en) * 2015-12-12 2016-02-24 深圳力子机器人有限公司 Modeling and optimizing method for AGV dispatching management

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102081786A (en) * 2011-01-30 2011-06-01 北京东方车云信息技术有限公司 Vehicle scheduling method and system
CN103177575B (en) * 2013-03-07 2014-12-31 上海交通大学 System and method for dynamically optimizing online dispatching of urban taxies
CN105389975B (en) * 2015-12-11 2017-11-14 北京航空航天大学 Special train dispatching method and device

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105354648A (en) * 2015-12-12 2016-02-24 深圳力子机器人有限公司 Modeling and optimizing method for AGV dispatching management

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
Where to find my next passenger;Jing Yuan等;《Proceedings of the 13th international conference on Ubiquitous computing》;20110930;第109-118页 *
基于真实轨迹的出租车智能调度系统研究;邹庆楠;《中国优秀硕士学位论文全文数据库信息科技辑》;20150615;第I138-4页 *

Also Published As

Publication number Publication date
CN106934490A (en) 2017-07-07

Similar Documents

Publication Publication Date Title
CN106952017B (en) AGV scheduling method and device
CN106934490B (en) AGV call prediction method and device
US9726502B2 (en) Route planner for transportation systems
CN114118496B (en) Method and system for automatically scheduling queuing reservation based on big data analysis
CN110264123B (en) Distribution path generation method and equipment
US9865171B2 (en) Geographical positioning in time
CN103984993A (en) Rail transit passenger flow OD distribution real-time speculation method
CN114594744B (en) Distributed factory production distribution integrated scheduling method and system
CN108806302A (en) A kind of vehicle dispatching method and device
US20190035170A1 (en) Servicing schedule method based on prediction of degradation in electrified vehicles
WO2019085459A1 (en) Evaluation management method, application server, and computer-readable storage medium
CN115730754A (en) Method and device for preventing robot congestion under warehouse picking system
CN111343345B (en) Management method, system, electronic equipment and medium for outgoing call of hotel order
CN112498132B (en) Vehicle charging intention determining method and device and vehicle
CN111652407B (en) Task processing method, device, medium, electronic equipment and system in warehouse
CN112465384A (en) Transportation capacity scheduling method and device, computer equipment and computer readable storage medium
CN104504615A (en) Data processing method and system for monitoring electric power operation cost
CN114531374B (en) Network monitoring method, device, equipment and storage medium
CN113256212B (en) Order processing method, device, communication equipment and storage medium
CN111027909A (en) Method for calculating work efficiency of park operation vehicle
CN115049079A (en) Management method and management platform for power battery recovery
US20210133676A1 (en) Shipping Dock Detention Data
CN118043769A (en) Air conditioner and control system thereof
EP3460732B1 (en) Dispatching method and system based on multiple levels of steady state production rate in working benches
CN113192354B (en) Large-scale station waiting duration prediction method based on time pane state probability

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