CN111598481B

CN111598481B - Shared bicycle flow system, automatic scheduling system and method based on subarea division

Info

Publication number: CN111598481B
Application number: CN202010455032.6A
Authority: CN
Inventors: 赵亮; 徐聪; 吴云凤; 白翰; 崔娜
Original assignee: Shandong Zhengqu Institute Of Transportation Engineering; Shandong Zhengqu Traffic Engineering Co ltd; Shandong Jiaotong University
Current assignee: Shandong Zhengqu Institute Of Transportation Engineering; Shandong Zhengqu Traffic Engineering Co ltd; Shandong Jiaotong University
Priority date: 2020-05-26
Filing date: 2020-05-26
Publication date: 2023-09-01
Anticipated expiration: 2040-05-26
Also published as: CN111598481A; WO2021238231A1

Abstract

The present disclosure provides a shared bicycle flow system, an automatic dispatch system based on division of subareas, and a method thereof, wherein the shared bicycle flow system comprises an overground conveying device arranged at each bicycle picking and placing point, an underground conveying device connected with the overground conveying device of each bicycle picking and placing point, and a multi-layer storage device capable of providing a bicycle for the overground conveying device or the underground conveying device; adjacent bicycle picking and placing points and ground storage devices are connected through a ground conveying device or an underground conveying device to form a mobile conveying network of the sharing bicycle. The provided flow system realizes the linkage of stations in a certain area, predicts the demand of each station through a comprehensive demand prediction method, then carries out dynamic subdivision to form a demand scheduling scheme of each station in the subdivision, and finally realizes the automatic transportation of the sharing bicycle according to the scheduling scheme, and provides high-efficiency and convenient bicycle storage service for users to the maximum extent when the users have demands.

Description

Shared bicycle flow system, automatic scheduling system and method based on subarea division

Technical Field

The disclosure relates to the technical field of sharing bicycles, in particular to a sharing bicycle flow system, an automatic scheduling system and a method based on subarea division.

Background

The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.

In recent years, the appearance of sharing single vehicles perfects a public transportation system, solves the problem of the last kilometer of citizens, accords with the environment-friendly living concept, and generates a series of problems along with the rapid increase of the number of sharing single vehicles. In the street and alley of the city, the problem of random parking of a bicycle is gradually highlighted, the phenomenon of no road can walk and no place can park is brought to the attention of various social circles, a series of measures such as parking point standardization and electronic fence popularization are implemented, the parking problem is standardized to a certain extent, but the convenience and the randomness of the sharing travel are relatively weakened, the problems of difficult bicycle searching and parking occur, and the user experience is poor; at present, all sharing bicycle stations are scattered, which is not beneficial to coordination control and has great difficulty in centralized management.

The inventor finds that the following problems mainly exist in the field of shared bicycle transportation at present:

firstly, all people need to participate in transportation, and the coverage area of a shared bicycle is wider, so that a large amount of human resources can be consumed, and the problem of untimely scheduling exists; although some inventions propose some shared bicycle carriers and some handling devices, there are significant unsafe factors for the carriers to be affected by the surrounding environment when automated transportation is achieved. Although the carrying device is arranged, people are required to realize the transportation between the shared bicycle stations, so that the labor is greatly consumed, the problems of limited dispatching personnel and untimely dispatching exist, and the user requirements can not be well met.

Secondly, the problem that the quantity of the shared bicycles put in each station is unbalanced with the actual demand is common, and the user demand cannot be met well. In the field of shared bicycle demand prediction, the prior art mainly utilizes some algorithms of machine learning to analyze and establish a model for the history data of the shared travel to predict, does not fully consider the combination of the essential demands of users, and can improve the accuracy of demand prediction to a certain extent from the viewpoint of the demands of the users; meanwhile, the consideration factors are incomplete, the fluctuation factors of the surrounding environment of the site are not fully considered, and the influence of suction points around the site on the fluctuation of the demand is ignored.

Thirdly, in the aspect of sharing a bicycle scheduling method, the intrinsic association degree among stations is mainly considered, the influence of intrinsic characteristics and surrounding environment characteristics of each station is not fully considered, the effectiveness of scheduling cannot be met, invalid scheduling work is continuously performed, and a large amount of resource waste is caused.

Disclosure of Invention

In order to solve the problems, the present disclosure provides a flow system of a shared bicycle, an automatic dispatching system based on subdivision, and a method thereof, wherein the proposed flow system realizes the linkage of stations in a certain area, predicts the demand of each station through a comprehensive demand prediction method, then performs dynamic subdivision to form a demand dispatching scheme of each station in the subdivision, and finally the flow system realizes the automatic transportation of the shared bicycle according to the dispatching scheme, and provides high-efficiency and convenient bicycle access service to the maximum extent for users when the users have demands.

In order to achieve the above purpose, the present disclosure adopts the following technical scheme:

a first object of the present disclosure is to provide a shared bicycle flow system including an above-ground conveying device provided at each bicycle pick-and-place point, an underground conveying device connected to the above-ground conveying device at each bicycle pick-and-place point, and a multi-layered storage device capable of providing a bicycle to the above-ground conveying device or the underground conveying device; adjacent bicycle picking and placing points and ground storage devices are connected through a ground conveying device or an underground conveying device to form a mobile conveying network of the sharing bicycle.

A second object of the present disclosure is to provide an automatic scheduling method based on division of subregions, comprising the steps of:

Acquiring bicycle picking and placing data of bicycle picking and placing points, and predicting the demand of each bicycle picking and placing point based on a method for fusing user demands and sharing travel attractive force by a random forest algorithm;

according to the prediction result of the demand, dividing the bicycle picking and placing points in a sub-region dynamic mode based on tree branches and combining internal and external factors, and generating a scheduling scheme according to the division result;

the control of the scheduling is performed according to a scheduling scheme.

A third object of the present disclosure is to provide an automatic dispatch system based on subdivision, including the above-mentioned shared bicycle flow system, and a control platform for sending dispatch instructions to the shared bicycle flow system; the control platform comprises a shared bicycle demand prediction system and a dynamic subarea division scheduling system;

the shared bicycle demand prediction system is configured to perform the shared bicycle demand prediction method in the automatic scheduling method based on division of subregions described above;

alternatively, the dynamic subdivision scheduling system is configured to perform a dynamic subdivision scheduling method of the above-described subdivision-based automatic scheduling method.

Compared with the prior art, the beneficial effects of the present disclosure are:

(1) The shared bicycle flow system disclosed by the invention combines and utilizes the space above ground and underground, a vehicle transport network is formed in conditional zones such as non-isolated areas and non-isolated areas, all shared bicycle demand areas are basically covered in a certain area, the vehicles on the storage device or the transport rail are transported to each bicycle picking and placing point according to the need, a large number of bicycles are not required to be stored for a long time by the bicycle picking and placing point, the area requirement of the bicycle picking and placing point can be met in a smaller area, a plurality of continuous bicycle picking and placing points are arranged in the flow system, and a user can conveniently and rapidly access the bicycle in the area along the line; meanwhile, the storage device is of a multi-layer structure, so that the occupied area for storing the bicycle can be reduced. Breaks through the limitation of the original manpower to participate in dispatching transportation, and can avoid the problems of limited dispatching personnel, untimely dispatching and the like; the parking behavior of the user can be effectively standardized, the problem of disordered parking is avoided, meanwhile, the vehicle is efficiently scheduled, and the problem of difficult vehicle and parking can be relieved.

(2) The sharing bicycle demand prediction method disclosed by the invention is based on the random forest algorithm to fuse the user demand and the sharing travel attraction, so that the user demand can be accurately obtained, the dispatching accuracy is improved, the execution of invalid dispatching is reduced or avoided, the dispatching times are reduced, and the dispatching execution efficiency of the system is improved.

(3) The dynamic subarea division scheduling method disclosed by the invention is used for carrying out dynamic subarea division scheduling based on the tree branch principle and integrating internal and external factors, and the shared bicycle demand prediction method based on the random forest algorithm and the shared travel attraction is combined with the second aspect of the disclosure, so that the shared bicycle flow system based on the human non-isolation or the machine non-isolation in the first aspect of the disclosure is subjected to complementary optimization, a demand scheduling scheme in each subarea can be formed, and the user scheduling demands are met to the maximum extent.

Drawings

The accompanying drawings, which are included to provide a further understanding of the disclosure, illustrate and explain the exemplary embodiments of the disclosure and together with the description serve to explain and do not limit the disclosure.

FIG. 1 is a schematic flow system architecture diagram of embodiment 1 of the present disclosure;

FIG. 2 is a schematic diagram of a storage device according to embodiment 1 of the present disclosure;

FIG. 3 is a schematic diagram of the structure of each storage layer in the storage device according to embodiment 1 of the disclosure;

FIG. 4 is a schematic view of a flow system apparatus setup position in a flow system of embodiment 1 of the present disclosure;

fig. 5 is a schematic structural view of an above-ground conveying device of embodiment 1 of the present disclosure;

FIG. 6 is a schematic view of the structure of an underground conveyance device of embodiment 1 of the present disclosure;

FIG. 7 is a schematic view of the structure of a bicycle carrier of embodiment 1 of the present disclosure;

FIG. 8 is a schematic diagram of the structure of an intelligent induction electronic lock in a bicycle carrier in accordance with embodiment 1 of the present disclosure;

FIG. 9 is a block diagram of an automatic dispatch system based on subdivision of embodiment 2 of the present disclosure;

FIG. 10 is a flow chart of a shared bicycle demand prediction method of embodiment 3 of the present disclosure;

FIG. 11 is a flow chart of a dynamic subdivision scheduling method in accordance with embodiment 4 of the present disclosure;

FIG. 12 is a control method of execution scheduling of embodiment 5 of the present disclosure;

wherein: 1. the system comprises a protective fence, 2, an ground transportation rail, 3, a position sensor, 4, an automatic telescopic door, 5, a door opening button, 6, a vehicle taking button, 7, a vehicle storage button, 8, a vehicle carrying device, 9, an inductive intelligent lock, 10, a reed switch, 11, a magnet, 12, a driving motor, 13, a positioning device, 14, an electronic lock locking base, 15, a solar panel, 16, an inductive device, 17, a storage device, 18, a vehicle inlet, 19, a vehicle outlet, 20, a first storage layer, 21, a second storage layer, 22, a third storage layer, 23, a spiral ascending rail, 24, an inlet and outlet rail, 25, a vehicle storage area, 26, an underground conveying device, 27, an underground transportation rail, 28, a fixed platform, 29, a telescopic driving power device, 30, a pressure telescopic device, 31, a vehicle carrying part, 32, a pressure sensor, 33, a non-isolated green belt, 34, a road, 35, a control module, 36, a stand column, 37, a device for fixing a vehicle in a vehicle storage area, 38, a telescopic device, 39 and a main controller.

The specific embodiment is as follows:

the disclosure is further described below with reference to the drawings and examples.

It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the present disclosure. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs.

It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments in accordance with the present disclosure. As used herein, the singular is also intended to include the plural unless the context clearly indicates otherwise, and furthermore, it is to be understood that the terms "comprises" and/or "comprising" when used in this specification are taken to specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof. It should be noted that, without conflict, the various embodiments and features of the embodiments in the present disclosure may be combined with each other. The embodiments will be described in detail below with reference to the accompanying drawings.

Example 1

In one or more embodiments disclosed herein, as shown in fig. 1-8, a shared bicycle flow system includes an above-ground conveyor disposed at each bicycle pick-and-place point, an underground conveyor connected to the above-ground conveyor at each bicycle pick-and-place point, and a multi-layered storage device 17 capable of providing a bicycle to the above-ground conveyor or the underground conveyor; adjacent bicycle picking and placing points and the ground storage device 17 are connected through the ground conveying device or the underground conveying device to form a mobile conveying network of the sharing bicycle.

The storage device 17 is used for storing the bicycles, and the above-ground conveying device and the underground conveying device are used for conveying the bicycles to the picking and placing points of the respective bicycles according to the bicycle requirements.

In this embodiment, the above-ground and underground spaces are combined and utilized to form a vehicle transportation network, and vehicles in the storage device 17 or the transportation track are transported to each bicycle picking and placing point according to the needs, so that the bicycle picking and placing point does not need to store a large number of bicycles for a long time, and the area requirement of the bicycle picking and placing point can be met in a small area. The storage device 17 is of a multi-layer structure, the occupied area for storing the bicycles can be reduced, meanwhile, the bicycles are stored uniformly through the overground storage device 17, and when the vehicles are required for each taking and placing point, a certain number of bicycles are put into each taking and placing point through the overground conveying device or the underground conveying device.

As shown in fig. 4, taking a conventional intersection as an example, the above-ground conveying device is arranged at the mechanically non-isolated green belt 33, the underground conveying device can be arranged below the road 34, and meanwhile, the above-ground conveying device and the underground conveying device can be arranged to effectively avoid the influence of the flow system on road traffic.

Alternatively, the storage device 17 is used for storing the bicycles, and may be disposed under the ground or on the ground, preferably, the storage device is disposed on the ground, so that construction costs may be effectively reduced.

Alternatively, the storage device 17 may be disposed at a location where it is desired, and the storage device 17 may be disposed near a vehicle access point having a large vehicle volume, and one or more vehicle access points may share one above-ground storage device 17. In particular, the device can be arranged in a suitable area near the non-isolated area or the non-isolated area of people, such as green belts on two sides of a road.

As shown in fig. 1, the above-ground conveying device is schematically connected to the storage device 17, and fig. 5 is a schematic structure of the above-ground conveying device alone.

The storage device 17 may be a structure as shown in fig. 1 and 2, including a device housing constituting a bicycle accommodating space, a plurality of storage layers provided in the housing, and a transportation rail capable of moving and transporting a bicycle between the respective storage layers, the transportation rail being provided with an inlet and an outlet of the transportation rail to the storage layers, respectively, at each storage layer.

The present embodiment may be provided with three layers as an example, and a first storage layer 20, a second storage layer 21, and a third storage layer 22 are provided.

It will be appreciated that the lowest storage level is provided with a trolley inlet 18 and a trolley outlet 19, respectively, connected to an above-ground conveyor, or an underground conveyor, for moving the trolley to the storage device 17, or for transporting the trolley of the storage device 17 to the respective trolley pick-and-place points.

In some embodiments, the specific structure of the transport track may be provided as: the spiral ascending and rotating track structure comprises a vertical column 36 arranged in the shell and a spiral ascending track 23 fixed on the vertical column.

Alternatively, as shown in fig. 3, each storage layer includes an entrance track 24 provided at the layer in connection with the spiral-type ascending track 23, and a vehicle storage area 25 in rail connection with the entrance track 24.

In order to increase the storage area of the storage area vehicles, as many bicycles as possible are placed, the vehicle storage area 25 is provided with an inclined plane with a certain radian and angle, and a bicycle fixing device 37 is arranged on the inclined plane; alternatively, the bicycle fixing device 37 may be provided as a clamping device, two clamping blocks being provided, the movement of which is controlled by pneumatic or electric means.

For transporting and transferring the single vehicle between the above-ground conveying device, the underground conveying device and the storage device 17, fixing devices for placing the single vehicle can be arranged in each device, for example, a fixing platform can be arranged on a track, a carrying device capable of moving independently can be arranged, and the single vehicle carrying device 8 capable of moving freely relative to each device can be arranged.

As one structure that can be realized, all the bicycle pick-and-place points are also provided with a control module 35, as shown in fig. 7 and 8, the bicycle carrying device 8 includes a bottom plate, a driving device and a bicycle fixing device that are disposed on the bottom plate and drive the bottom plate to move, a bicycle information identifying device, a main controller 39 and a wireless communication module, the main controller 39 is respectively connected with the driving device, the bicycle fixing device, the bicycle information identifying device and the wireless communication module, and the main controller 39 is wirelessly connected with the control module 35.

Specifically, the driving device may be electrically driven, and includes a driving motor 12, a power supply battery connected to the driving motor 12, and a moving mechanism connected to the driving motor 12, and preferably, a solar panel 15 may also be disposed, where the solar panel 15 is electrically connected to the power supply battery, and provides electric energy for the driving motor. The moving mechanism may be a wheel, track wheel or track.

Alternatively, the control module 35 may employ a single-chip microcomputer.

Alternatively, the bicycle fixing device may include an inductive smart lock 9 and an inductive device 16 fixed on the bottom plate, and the inductive smart lock 9 and the inductive device 16 are electrically connected to the main controller 39, respectively, for transmitting the inductive information to the control module 35. When a shared bicycle is sensed to be placed on the carrying device 8, the inductive intelligent lock 9 is automatically locked.

The induction type intelligent lock can be arranged in any shape, such as arc shape or polygon shape.

The sensing means 16 may be a pressure sensor for determining whether a bicycle is placed on the base plate, improving the reliability of the operation of the bicycle carrier 8 and providing the accuracy of the bicycle placement information.

It will be appreciated that the electronic lock fixing base 14 is further included for fixing the lock body of the induction type intelligent lock 9.

Optionally, the information identifying device includes a reed pipe 10 disposed on the sharing bicycle, a telescopic device 38 disposed on the bottom plate of the bicycle carrying device 8 and opposite to the reed pipe 10, and a magnet 11 disposed on the top end of the telescopic device 38, where the telescopic device 38 is electrically connected to a main controller 39, and the reed pipe 10 is wirelessly connected to the main controller.

After the sensing device 16 senses signals, the sensing signals are transmitted to the main controller 39 of the conveying device, the telescopic device 38 is controlled to pop up to a certain height, the magnet 11 at the top is made to approach the reed switch 10 in the bottom shell near the pedal of the sharing bicycle, the reed switch 10 is closed, the central control unit of the sharing bicycle can be communicated, the central control unit transmits locking signals to the main controller 39 through the wireless mobile communication module, the inductive intelligent lock 9 is controlled to be locked, and the telescopic assembly automatically contracts back into the conveying device.

Further, the locking information of the induction type intelligent lock 9 on the bicycle carrying device 8 is linked with the shared bicycle background system, the shared bicycle background system receives the locking information of the bicycle lock on the shared bicycle and receives the locking information of the induction type intelligent lock 9, and the bicycle is successfully returned; when the user closes the lock on the bicycle and does not receive the locking information of the induction type intelligent lock 9, the bicycle returning operation fails.

After the sharing bicycle is fixed, the sharing bicycle is controlled to be locked, the bicycle returning operation is realized, the parking behavior of a user can be standardized, the sharing bicycle is standardized to be parked on a transportation device in a flowing system, and the circulation of the sharing bicycle is realized.

Optionally, in order to determine the specific position of the bicycle carrier 8, the bicycle carrier 8 may further comprise a positioning module 13. The positioning module 13 may be a GPS positioning module.

In some embodiments, the above-ground conveying device may be configured as a track structure, as shown in fig. 1 or 5, and the above-ground conveying device includes a control module 35 and an above-ground conveying track 2 laid on the ground, guard rails 1 disposed on both sides of the above-ground conveying track 2, a bicycle access port disposed on the guard rails to provide a bicycle access path, and a bicycle interaction device disposed at the bicycle access port for receiving the interaction information, where the control module 35 is communicatively connected to the bicycle carrier device 8 and the bicycle interaction device, respectively, through a communication module.

The control module 35 receives the information of the access vehicle interaction device and controls the vehicle carrying device 8 to carry the vehicle to the corresponding position. The access vehicle interaction device is used for receiving information of a user accessing a bicycle.

Alternatively, the guard rail 1 may be a fence or a fence.

In some embodiments, the access vehicle interaction device comprises an automatic retractable door 4 arranged at a bicycle access port, and buttons for controlling the opening and closing of the retractable door, and may comprise a door opening button 5, a vehicle taking button 6 and a vehicle storing button 7.

The bicycle parking device also comprises a position sensor 3 arranged at the bicycle access port, the position sensor 3 is in wireless connection with the main controller, when the bicycle carrying device 8 moves to the position sensor 3, the position sensor 3 transmits an action signal to the main controller 39, and the main controller controls the bicycle carrying device 8 to stop. The position sensor 3 can also adopt an RFID tag, an RFID reader is arranged on the bicycle carrying device 8, and when the RFID reader detects corresponding tag information, the bicycle is parked.

The underground conveyor 26 is used to provide an underground transport path for the transfer of vehicles between individual vehicle pick-and-place points or between the storage device 17 and individual vehicle pick-and-place points, and is located at a location where it is inconvenient to locate the conveyor from the road surface, such as under the ground at some intersection.

The underground conveying device can realize the transportation of the sharing bicycle from the ground to the underground and from the underground to the ground. In some embodiments, as shown in fig. 6, the underground transportation device may include an underground transportation rail 27, a pressure telescoping device 30 disposed at the above-ground and underground connection port, an upper end surface of the pressure telescoping device 30 being flush with the ground and abutting when the pressure telescoping device 30 is extended to the first position, and an upper end surface of the pressure telescoping device 30 being flush with and abutting the underground transportation rail 27 when the pressure telescoping device 30 is compressed to the second position.

Alternatively, the pressure telescoping device 30 may include a bicycle carrying portion 31, a telescoping mechanism and a fixed platform 28 fixedly connected to the bicycle carrying portion 31, and a telescoping driving power device 29 disposed on the fixed platform 28 and electrically connected to the telescoping mechanism, which are sequentially disposed from top to bottom. The bicycle carrier portion 31 may be a carrier plate or a tracked carrier plate having a track shape configuration that matches the underground transportation track 27. The fixed platform 28 provides stable support. A pressure sensor 32 may also be provided on the bicycle carrier part 31 for detecting whether a bicycle or a bicycle carrier 8 is placed on the bicycle carrier part 31.

Specifically, the telescopic driving power device 29 may be a hydraulic driving device, and the telescopic mechanism is a pressure telescopic rod.

Further, the system may also include a control platform communicatively coupled to the control module 35 in the flow system.

The working principle of the flow system is as follows:

the vehicle-taking interaction device is used for acquiring the vehicle-taking requirement of a user, the vehicle can be directly taken in and taken out by pressing the door opening button, the vehicle-taking button can be pressed when the vehicle does not exist at the current station, and the vehicle-storing button can be pressed by the carrying device 8 which does not park at the current station; the storage device 17 comprises normal vehicle storage and vehicle storage to be maintained, and the vehicles in the system are necessary to be supplemented and stored according to the number of the shared bicycles in the system and the requirement condition; the control module 35 receives and analyzes the running information in the system in real time, and outputs control instructions to schedule the single vehicles and the single vehicle carrying devices 8 in the whole flow system; the bicycle carrier 8 carries the bicycle according to the control scheduling instruction of the control module 35. In addition, whether the vehicle is parked normally is judged by the vehicle carrying device 8, and when the user cannot directly access the vehicle, the control module 35 dispatches shared vehicles or parked vehicle carrying devices 8 of nearby sites or the storage device 17 in real time according to the conditions of each site and the user requirements, so that the vehicle access requirements of the user are met to the maximum extent.

Example 2

The present embodiment provides an automatic dispatching system based on division of subareas, which performs division of subareas of shared bicycle according to user requirements, and automatically dispatches the shared bicycle flow system described in embodiment 1, for dispatching each shared bicycle or bicycle carrier 8 in the flow system.

An automatic dispatching system based on subarea division, as shown in fig. 9, comprises the shared bicycle flow system described in embodiment 1 and a control platform for sending dispatching instructions to the shared bicycle flow system, wherein the control platform comprises a shared bicycle demand prediction system and a dynamic subarea division dispatching system;

shared bicycle demand prediction system: the system is configured to be used for predicting the user demand of different periods of time of each station to obtain the predicted demand of each bicycle pick-and-place point; and a theoretical basis is provided for the number of vehicles distributed in the whole flow system.

Dynamic subdivision scheduling system: the system is configured to divide dynamic subareas according to the obtained predicted demand of each bicycle picking and placing point, schedule according to bicycle demand proportion of each bicycle picking and placing point in the subareas, generate a scheduling scheme and send the scheduling scheme to the control module 35 so that the control module 35 controls the bicycle or bicycle carrying device 8 in the flow system.

Example 3

The method for predicting the demand of the shared bicycle can be realized on a control platform connected with a control module, specifically can be realized by a system for predicting the demand of the shared bicycle, and as shown in fig. 10, the method comprises the following steps:

step 1, user demand statistics: acquiring user travel information and travel reservation information, and counting first bicycle demand X of bicycle picking and placing points ₁ ；

Step 2, demand prediction based on sharing travel attraction: suction points of the area near the pick-and-place point of the bicycle are determined according to each suction pointCalculation of the attractive force of the points to obtain the second bicycle demand X of the bicycle pick-and-place points ₂ ；

Step 3, calculating and obtaining a third bicycle demand X of bicycle pick-and-place points based on a random forest algorithm ₃ ；

And step 4, weighting and summing the demand obtained in the step to obtain the demand of each bicycle pick-and-place point.

In step 1, travel information of a user can be acquired in an excitation feedback mode. The motivation feedback can be that an integral motivation mode is adopted, a travel questionnaire is sent, the questionnaire comprises a main riding path starting and ending point, a travel time period and user comments, questionnaire investigation information is received, integral is added for an account filled with the questionnaire, the reliability of data is ensured, and if the riding information of the user is seriously inconsistent with the filling content of the questionnaire, a certain integral is deducted for the user. Users mainly include fixed users holding a week card, month card, or year card, and some general users.

In step 2, determining suction points of the area near the bicycle pick-and-place point, and calculating to obtain a second bicycle demand according to the attractive force of each suction point, wherein the method comprises the following steps:

step 21, dividing the suction grade of the suction point;

the attraction points are public places with relatively large human flow, such as hospitals, schools, parks, bus stations, subway stations and the like. Suction points refer to places that attract people to share a trip.

Optionally, the method can be divided according to the size of the traffic, and the first level: bus stops, subway stops, second grade: district, supermarket, school, tertiary: restaurant, park square, four stages: others;

step 22, determining suction points in the set area of the pick-and-place points of the bicycle, and determining the suction force reduction coefficient lambda of each suction point according to the suction level _x ；

The setting area of the bicycle picking and placing point can be set as a range area of one kilometer around the bicycle picking and placing point, the larger the traffic flow is, the larger the reduction coefficient is, the larger the demand is, and the reduction coefficient lambda can be set according to the traffic flow proportion _x 。

Step 23, reducing coefficient lambda according to the attractive force _x Calculating and obtaining second bicycle demand X of shared bicycle pick-and-place points ₂ ；

Second bicycle demand X ₂ The following calculation formula can be adopted for solving:

Wherein X is ₂ The demand for sharing the bicycle; s is S _{Total (S)} Is the area of a kilometer region near the suction point; s is S _{i inhale} The floor area of a nearby suction point is occupied, i is the i-th bicycle picking and placing point; k is the slow travel proportion of the suction points, and the travel proportion can be divided according to age hierarchy for sites with obvious time characteristics and age characteristics; n is the number of people at the suction point; lambda (lambda) _x The coefficient of attraction determined for the travel attraction points is shared according to different grades.

Step 3, calculating and obtaining a third bicycle demand X of bicycle pick-and-place points based on a random forest algorithm ₃ Comprises the following steps:

step 31, acquiring a sample data set.

The method comprises the steps of obtaining historical vehicle number data of all bicycle picking and placing points in a whole regional system and corresponding relevant characteristic data, wherein the historical vehicle number data comprise characteristic data such as geographic positions, time, seasons, holidays, workdays, weather, temperature, humidity and wind speed, and the like, and the running track and starting and ending point data of the bicycles in the regional system are shared as an original data set.

And step 32, sample extraction is carried out on the sample data set, and a training subset of the plurality of decision trees is obtained.

Sampling S training sample subsets from the total samples by adopting a bootmap resampling method, and constructing S regression trees, wherein the sampled training samples are training sets, and samples which are not sampled in the total samples are used as test sets.

Step 33, decision tree construction: based on a loss minimization principle, each training subset is correspondingly trained to obtain a decision tree, in the decision tree training process, a set number of characteristic variables with larger relativity are selected to participate in decision tree node splitting, and a random forest regression model is obtained through training of a plurality of training subsets;

generating a decision tree based on a loss minimization principle by each training sample subset, generating S decision trees by S training sample subsets to form a random forest, and setting the selected characteristic variables not to exceed log in order to solve the overfitting phenomenon caused by excessive characteristic variables ₂ M+1, wherein M represents the number of associated feature variables, the participated feature variables are selected according to a correlation principle, and partial feature variables with larger correlation G are selected to participate in the decision tree node splitting process according to the correlation magnitude sequence.

The correlation determination method may be as follows:

wherein X is a demand variable, Y _i For a certain characteristic variable, G is the correlation between the demand variable and a certain characteristic value, A is the sum of the data numbers of all the demand variables X and all the characteristic variables Y, A _i All data corresponds to the number of data in a for a certain feature.

After the S decision trees are constructed, simulation is carried out by using test set data, the error of the decision trees is estimated, and the parameters of the decision trees are optimized. And (3) averaging the error estimates of the S decision trees to obtain a random forest generalization error estimated value, and optimizing model parameters.

Step 34, predicting results of the random forest regression model: acquiring bicycle travel data and corresponding characteristic variable data of bicycle pick-and-place points in real time, inputting the bicycle travel data and the corresponding characteristic variable data into a random forest regression model, obtaining voting results of each decision tree, and weighting to obtain random forest regression prediction results, namely third bicycle demand X of the bicycle pick-and-place points ₃ ；

The prediction result output by the random forest regression prediction model is generated by voting results of all decision trees. The random forest regression prediction results are as follows:

wherein Y is _i For data of relevant characteristic factors, H _ik S is the number of the constructed total decision trees and X is a single decision tree prediction model _Y And (5) sharing the bicycle demand regression prediction result.

In step 3, the number of feature variables for constructing the decision tree is limited according to the correlation magnitude, and a random forest algorithm is optimized, so that the demand X is predicted more accurately ₃ 。

The result of obtaining the whole demand prediction by the steps 1-3 is composed of three parts, and the demand obtained by the steps is weighted and summed, specifically as follows:

X＝λ ₁ X ₁ +λ ₂ X ₂ +λ ₃ X ₃

wherein lambda is ₁ 、λ ₂ And lambda (lambda) ₃ Representing the corresponding weights; x is the total demand of the site; x is X ₁ The first bicycle demand; x is X ₂ A second bicycle demand obtained for a shared travel attraction; x is X ₃ A third bicycle demand is obtained for a prediction based on a random forest algorithm.

The embodiment integrates the user demands, the shared traveling attraction and the random forest algorithm, deeply mines the user demands, starts from the economic and convenient angles of the users, provides excitation feedback service, mainly mines the vehicle demands of fixed users (week, month and year card users) within a certain time, synthesizes the preset information of some common users, and improves the accuracy of demand prediction; the index of sharing trip attraction is introduced, and the fluctuation influence of the suction points around the site on the demand is fully considered; the historical data of each site can be analyzed more conveniently, a high-precision random forest algorithm is selected, the number of characteristic variables for constructing a decision tree is limited through characteristic variable correlation analysis, and the accuracy of random forest prediction is improved, so that the accuracy of demand prediction is improved.

Example 4

The dynamic subarea division scheduling method can adjust the subarea range in real time through dynamic subarea division, and improves the scheduling flexibility and timeliness. The method can be realized on a control platform connected with a control module, and concretely can be realized by a dynamic subarea division scheduling system, as shown in fig. 11, and comprises the following steps:

Step 1, acquiring bicycle access data and bicycle track information of each bicycle access point;

step 2, classifying the bicycle picking and placing points according to different characteristics according to the acquired data;

step 3, dynamically dividing adjacent bicycle picking and placing points according to the dynamic change of the demand according to the classification result to form a plurality of subareas;

step 4, scheduling each sub-zone according to the sub-zone division result: and if the scheduling cannot meet the bicycle requirement of the subarea, executing the steps 1-3 to carry out subarea division again.

In step 1, historical data information and real-time dynamic access information of access vehicles of all stations are collected. By adopting the flow system of the embodiment 1, the data information of the picking and placing points of each bicycle can be acquired more conveniently and rapidly, so that the internal characteristics of the picking and placing point of each station can be accurately analyzed.

classifying the pick-and-place points of the bicycle according to different characteristics may include sorting by time characteristics, sorting by level as required, and the like.

Step 21, classifying according to time characteristics, may be divided into:

time-division site: the bicycle demand is relatively large during certain time periods, such as the early peak and late peak periods.

Full time segment site: a site with relatively large demand in all time periods;

common sites: there is no obvious temporal feature.

Step 22, grading the common sites according to the requirements: and dividing according to the predicted demand quantity of each bicycle pick-and-place point and the demand quantity grade.

Step 3, dynamically dividing adjacent bicycle picking and placing points according to dynamic changes of requirements according to classification results to form a plurality of dynamic subareas, wherein the method specifically can be as follows:

step 31, forming complementary subareas with surrounding stations and forming corresponding scheduling schemes by combining dynamic demand conditions of the surrounding stations according to time characteristics of the time-division stations and the full-time-division stations;

step 32, carrying out dynamic subarea division scheduling for a common station, which may include the following steps:

step 321, merging sites with high complementarity in common sites to serve as a picking and placing point; complementarity is that high demand is non-overlapping in time, and high demand periods are staggered.

Step 322, dynamically selecting a pick-and-place point with the largest real-time demand and stable demand in the set area range as a main station;

the picking and placing points are all continuous in geographic position, and according to the principle of layered sampling, a main station is selected at intervals of a certain number of stations, and the selection principle can be as follows: the demand change has no obvious temporal characteristics; in a certain period of time, the demand is higher than that of surrounding stations, and the bicycle picking and placing point with the largest demand is dynamically selected according to the real-time demand;

Step 333, dividing the sub-regions according to the tree branch principle: selecting the most suitable surrounding sites to be combined by taking the main site as a center and adopting a complementarity principle and a shared travel attraction principle to form a subarea;

the tree branch principle fully combines the characteristics of the linkage of the flow system stations, the main stations are selected as roots, then the tree branch grows according to the upward principle, and when the growth requirement cannot be met, other nodes in the tree branch are selected to continue to grow. The system of the embodiment is in a strip shape or a road net shape, and is divided into sub-areas by forming a line from a main station according to the dividing conditions, and when the dividing conditions are not met, other stations on the line are selected for division.

If the current site continues to merge downwards and does not meet the dividing principle, the merging of the site is ended, then whether the previously merged site has the possibility of continuing merging is searched, and when the number of the merged sites in the sub-region reaches the set number (for example, 8 times), the dividing of the current sub-region is ended, specifically, the method can be as follows:

the method for calculating the demand difference value of the site within a certain time comprises the following steps:

Q＝Q _{at present} +Q _{And also (3) the method} -Q _{Is required to}

Wherein Q is a demand difference; q (Q) _{At present} The number of vehicles existing for the station; q (Q) _{And also (3) the method} The number of vehicles returned to the station in a certain time; q (Q) _{Is required to} The number of vehicles is required for a certain period of time for a station.

The complementary principle is satisfied: after a certain site is combined with a main site, the total demand difference Q is in a decreasing trend. The specific method for judging the demand complementarity is as follows:

wherein K is a reduction coefficient; q (Q) ₁ The difference value of the requirements of the subareas before the station combination is obtained; q (Q) ₂ The difference value of the requirements of the sub-areas after the station combination is obtained; h is the complementarity of the requirements after the merging stations.

The principle of sharing travel attraction is satisfied: if the current subarea demand difference is negative, showing that the supply and the demand are not needed, selecting a site with small shared travel attraction when the sites are combined, and if the demand difference is positive, showing that the supply and the demand are needed, selecting a site with large shared travel attraction.

The method for judging the shared travel attraction force comprises the following steps:

wherein A represents the size of a sharing trip attractive force of a certain site; d (D) _i Representing the distance from the station to a suction point in a nearby suction area; lambda (lambda) _x The attractive force reduction coefficient is determined for sharing the travel attraction points according to different grades; r is R _{i inhale} Representing a half of an attraction point in the attraction zoneDiameter is as follows; s is S _{Total (S)} Representing the total area of the suction area.

In step 4, scheduling is performed according to the division result of the subareas: and (3) aiming at the obtained subareas, carrying out bicycle adjustment among all taking and placing points in the subareas, carrying out bicycle adjustment on subarea layers, namely the subareas when the internal adjustment cannot meet the requirement of the required quantity, and carrying out the steps 1-4 to carry out subarea division again when the subarea layers cannot meet the requirement of the required quantity.

In the embodiment, the scheduling is firstly performed in each sub-region, the relevance in the sub-region is fully considered, the scheduling accuracy is improved, and the scheduling efficiency is improved.

Scheduling between picking and placing points of the bicycle in the subarea is carried out, and the method specifically comprises the following steps: and (5) adjusting the distribution of the picking and placing points of each bicycle in the subarea according to the required proportion.

If the difference value of the requirements of a certain subarea exceeds a certain threshold value, scheduling among subareas is carried out, and if the current subarea cannot be effectively adjusted, the division of the current subarea is adjusted.

The current dispatching scheme can not meet the requirements of users, dynamic subareas are divided again to form a new dispatching adjustment scheme, dynamic adjustment of subarea division can be realized, and dispatching flexibility is improved.

According to the dynamic subarea division method, the characteristics of site linkage of a flow system are fully combined, the tree branch principle is adopted to carry out selection and division of the points, the influence of internal factors and external factors of the sites is fully considered in the division condition, the site selection is carried out by utilizing the demand complementarity principle and the shared travel attraction principle, the influence of surrounding fluctuation factors is considered, meanwhile, different internal time characteristics of each site are considered, the method is divided into full-time-period point positions, time-period-point positions and common point positions, corresponding subarea division scheduling is carried out, and demand proportion scheduling is carried out in each subarea, so that the user scheduling demands can be better met.

Example 5

The present embodiment provides an automatic scheduling method based on division of sub-areas, which is implemented in the control platform in the system described in embodiment 2, and performs division of sub-areas of shared bicycle according to user requirements, and performs automatic scheduling on the shared bicycle flow system described in embodiment 1, and schedules each shared bicycle or bicycle carrier 8 in the flow system.

The automatic scheduling method based on subarea division comprises the following steps:

s1, acquiring bicycle picking and placing data of bicycle picking and placing points, and predicting the demand of each bicycle picking and placing point by a method of fusing user demands and sharing travel attractive force based on a random forest algorithm;

s2, according to the prediction result of the demand, dividing the bicycle pick-and-place points in a sub-area mode based on the tree branch principle and combining internal and external factors, and generating a scheduling scheme according to the division result;

s3, executing scheduling control according to a scheduling scheme.

Step 1 adopts the shared bicycle demand prediction method described in embodiment 3, and step 2 adopts the dynamic subdivision scheduling method described in embodiment 4 to obtain a scheduling scheme; also included are control methods performed by the control module 35 according to a scheduling scheme.

The control method executed by the control module 35, as shown in fig. 12, includes the following:

Step 1, conveying control between bicycle storage points: according to the scheduling scheme, the number of the bicycles at each station is adjusted between adjacent stations by controlling the carrying device 8;

the carrier device 8 receives the signal from the control module 35 and moves to a specified position according to the above-ground and below-ground conveying devices 26.

When a vehicle taking or parking requirement exists, a driving motor 12 of the carrying device 8 is started, and then the shared bicycle is controlled to be transported to the corresponding telescopic door 4; when the system does not have the requirement of vehicle storage and taking, the control module 35 realizes equidistant emission of the shared bicycle in each station according to the positioning data of the bottom positioning device 3 of each carrying device 8 when the system detects that all doors are closed;

when the shared bicycle passes through the intersection and is dispatched by the underground conveying device 26, and the shared bicycle and the carrying device 8 thereof are conveyed to the pressure sensor 32 of the underground conveying device, the pressure sensor 32 transmits signals to the telescopic driving power device 29 to start the pressure telescopic rod 30, the bicycle carrying part 31 is further controlled to convey the shared bicycle downwards, finally, the pressure telescopic rod is fully contracted into the fixed platform 28, the shared bicycle is conveyed to the ground through the underground conveying rail 27 after being conveyed, and when the shared bicycle reaches a designated place, the shared bicycle is conveyed to the ground through the pressure telescopic device 30 by the same method as the downward conveying, and the dispatching task is continued.

Step 2, access control of the access vehicle: acquiring user demand information of a vehicle access interaction device;

when a user needs to fetch the vehicle, receiving the vehicle fetching information and controlling the telescopic door to be opened; dispatching the bicycle closest to the dispatching distance to the telescopic door position;

when a user needs to park, parking information is received, and the telescopic door is controlled to be opened; dispatching the carrying device 8 closest to the dispatching position to the telescopic door;

after a user parks, sensing devices (16) at two ends of the carrying device sense that the vehicle is placed, and the vehicle is automatically locked, so that the parking is completed.

Where the nearest bicycle or carrier 8 may be at the current site, where there is no bicycle or carrier 8 that may be scheduled, may be at an adjacent site or in the storage 17.

Step 3, storage control of the bicycle: counting the number data of the single vehicles in the subareas, wherein the number data comprises normal vehicles and vehicles to be maintained, and separating the normal vehicles and the vehicles to be maintained;

when the quantity does not meet the demand requirement, replenishing the storage device 17 of the sub-zone with bicycles;

when the number of the bicycles is larger than the demand requirement, the redundant bicycles in the storage device of the subarea are dispatched to the subarea which is not satisfied by the demand.

In order to facilitate maintenance of maintenance personnel, if the vehicle needs maintenance, the vehicle is stored in the first storage layer 20 of the storage device, and when the number of vehicles to be maintained exceeds the set value of the total stored number of vehicles in the storage device 17, a maintenance instruction is sent to the terminal of the maintenance personnel.

A first storage layer 20, a second storage layer 21 and a third storage layer 22 are provided.

When the vehicle does not need maintenance, the shared bicycle is transported and stored upwards along with the spiral ascending track 23, and is transported upwards through the spiral ascending track 23 when being stored, and enters the second layer through the inlet and outlet 24 connected with the spiral ascending track 23 through the second storage layer 21 when passing through the second storage layer 21, the shared bicycle storage and fixing device is arranged around the center in a circle, the vehicles arriving from the spiral ascending track 23 are stored through continuous rotation, and when the second storage layer 21 is full, the first storage layer 20 and the third storage layer 22 are stored successively; when the shared bicycle is dispatched from the storage device 17, the vehicle at the first storage layer 20 is dispatched first, and when the underlying vehicle is not enough to be dispatched, the shared bicycle of the upper storage device is dispatched in sequence.

The system timely performs scheduling adjustment according to the real-time user demands and the step method, and meets the user demands to the maximum extent.

The foregoing description of the preferred embodiments of the present disclosure is provided only and not intended to limit the disclosure so that various modifications and changes may be made to the present disclosure by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present disclosure should be included in the protection scope of the present disclosure.

While the specific embodiments of the present disclosure have been described above with reference to the drawings, it should be understood that the present disclosure is not limited to the embodiments, and that various modifications and changes can be made by one skilled in the art without inventive effort on the basis of the technical solutions of the present disclosure while remaining within the scope of the present disclosure.

Claims

1. The shared bicycle flow system is characterized in that: the device comprises an overground conveying device arranged at each bicycle picking and placing point, an underground conveying device connected with the overground conveying device at each bicycle picking and placing point, and a multi-layer storage device capable of providing bicycles for the overground conveying device or the underground conveying device; adjacent bicycle picking and placing points and ground storage devices are connected through a ground conveying device or an underground conveying device to form a mobile conveying network of the sharing bicycle;

The storage device comprises a device shell which forms a bicycle accommodating space, a plurality of storage layers arranged in the shell, and a transportation rail which can move and transport the bicycle between the storage layers, wherein the transportation rail is respectively provided with an inlet and an outlet which are communicated with the storage layers by the transportation rail at each storage layer;

the transportation track of the storage device comprises an upright post arranged in the shell and a spiral ascending track fixed on the upright post;

the ground conveying device comprises a control module, a ground conveying rail paved on the ground, protective rails arranged on two sides of the ground conveying rail, a bicycle access port arranged on the protective rails and used for providing a bicycle access channel, and a bicycle access interaction device arranged at the bicycle access port and used for receiving interaction information, wherein the control module is respectively in communication connection with the bicycle carrying device and the bicycle access interaction device through communication modules;

the vehicle storing and taking interaction device comprises an automatic telescopic door arranged at a vehicle storing and taking opening and a button for controlling the switch of the telescopic door, and can comprise a door opening button, a vehicle taking button and a vehicle storing button;

the underground conveying device comprises an underground conveying rail and a pressure telescoping device arranged at an underground connecting port on the ground, when the pressure telescoping device stretches to a first position, the upper end face of the pressure telescoping device is flush with the ground and is in butt joint, and when the pressure telescoping device is compressed to a second position, the upper end face of the pressure telescoping device is flush with the underground conveying rail and is in butt joint;

The pressure telescoping device comprises a bicycle bearing part, a telescoping mechanism and a fixed platform which are sequentially arranged from top to bottom, and a telescoping driving power device which is arranged on the fixed platform and is electrically connected with the telescoping mechanism;

access control of access vehicle: acquiring user demand information of a vehicle access interaction device; when a user needs to fetch the vehicle, receiving the vehicle fetching information and controlling the telescopic door to be opened; dispatching the bicycle closest to the dispatching distance to the telescopic door position; when a user needs to park, parking information is received, and the telescopic door is controlled to be opened; dispatching the carrying device closest to the dispatching position to the telescopic door; after a user parks, sensing devices at two ends of the carrying device sense that the vehicle is placed, and automatically lock the vehicle to finish parking; the nearest bicycle or carrier can be at the current site, and the bicycle or carrier which can be scheduled is not at the current site, and can be at the adjacent site or in the storage device;

based on the subarea division, the method comprises the following steps: acquiring bicycle picking and placing data of bicycle picking and placing points, and predicting the demand of each bicycle picking and placing point based on a method for fusing user demands and sharing travel attractive force by a random forest algorithm; according to the prediction result of the demand, dividing the bicycle picking and placing points in a sub-region dynamic mode based on tree branches and combining internal and external factors, and generating a scheduling scheme according to the division result; executing control of scheduling according to a scheduling scheme;

The method for fusing user demands and sharing travel attractive force based on the random forest algorithm predicts the demand of each bicycle pick-and-place point, and comprises the following steps: acquiring user travel information and travel reservation information, and counting the first bicycle demand of bicycle picking and placing points; determining suction points of the area near the bicycle pick-and-place point, and calculating to obtain a second bicycle demand of the bicycle pick-and-place point according to the suction force of each suction point; calculating and obtaining a third bicycle demand of bicycle pick-and-place points based on a random forest algorithm; and (3) weighting and summing the demand obtained in the steps to obtain the demand of each bicycle pick-and-place point.

2. The shared bicycle flow system of claim 1, wherein: the mobile system further comprises a bicycle carrying device which freely moves relative to each device, each bicycle storage point is further provided with a control module, each bicycle carrying device comprises a bottom plate, a driving device and a bicycle fixing device, the driving device and the bicycle fixing device are arranged on the bottom plate and used for driving the bottom plate to move, the bicycle information identifying device, the main controller and the wireless communication module are respectively connected with the driving device, the bicycle fixing device, the bicycle information identifying device and the wireless communication module, and the main controller is in wireless connection with the control module.

3. The shared bicycle flow system of claim 2, wherein: the bicycle fixing device comprises an induction type intelligent lock and an induction device which are fixed on the bottom plate, wherein the induction type intelligent lock and the induction device are respectively electrically connected with the main controller and used for transmitting induction information to the control module.

4. The shared bicycle flow system of claim 2, wherein: the information identification device comprises a reed pipe arranged on the sharing bicycle, a telescopic device arranged on a bottom plate of the bicycle carrying device and opposite to the reed pipe in position, and a magnet arranged at the top end of the telescopic device, wherein the telescopic device is electrically connected with the main controller, and the reed pipe is in wireless connection with the main controller.

5. An automatic scheduling method based on subdivision, implemented based on a system according to any of claims 1-4, characterized by comprising the steps of:

Executing control of scheduling according to a scheduling scheme;

the method for fusing user demands and sharing travel attractive force based on the random forest algorithm predicts the demand of each bicycle pick-and-place point, and comprises the following steps:

acquiring user travel information and travel reservation information, and counting the first bicycle demand of bicycle picking and placing points;

determining suction points of the area near the bicycle pick-and-place point, and calculating to obtain a second bicycle demand of the bicycle pick-and-place point according to the suction force of each suction point;

calculating and obtaining a third bicycle demand of bicycle pick-and-place points based on a random forest algorithm;

and (3) weighting and summing the demand obtained in the steps to obtain the demand of each bicycle pick-and-place point.

6. The automatic scheduling method based on subdivision scheme as claimed in claim 5, wherein: in the shared bicycle demand prediction method, suction points of areas near bicycle pick-and-place points are determined, and a second bicycle demand is calculated according to the attractive force of each suction point, and the method comprises the following steps:

dividing the suction grade of the suction point;

determining suction points in a bicycle pick-and-place point setting area, and determining an attraction reduction coefficient of each suction point according to the attraction grade;

And calculating to obtain the second bicycle demand of the shared bicycle pick-and-place point according to the attractive force reduction coefficient.

7. The automatic scheduling method based on subdivision scheme as claimed in claim 5, wherein: the method for calculating and obtaining the third bicycle demand of the bicycle pick-and-place point based on the random forest algorithm comprises the following steps:

acquiring a sample data set;

sample extraction is carried out on the sample data set, and a training subset of a plurality of decision trees is obtained;

and (3) constructing a decision tree: based on a minimization principle, each training subset is correspondingly trained to obtain a decision tree, in the decision tree training process, a set number of characteristic variables with larger relativity are selected to participate in decision tree node splitting, and a random forest regression model is obtained through training of a plurality of training subsets;

random forest regression model prediction results: and acquiring the bicycle travel data and the corresponding characteristic variable data of the bicycle pick-and-place point in real time, inputting the bicycle travel data and the corresponding characteristic variable data into a random forest regression model, obtaining voting results of each decision tree, and weighting to obtain a random forest regression prediction result, namely the third bicycle demand of the bicycle pick-and-place point.

8. The automatic scheduling method based on subdivision scheme as claimed in claim 5, wherein:

the control of the dispatch performed according to the dispatch scheme includes a conveyance control between the storage points of the bicycles, an access control of the access vehicles, and a storage control of the bicycles.

9. The automatic scheduling method based on subdivision scheme as claimed in claim 5, wherein:

the method for generating the scheduling scheme according to the division result, namely the dynamic subarea division scheduling method, comprises the following steps:

acquiring bicycle access data and bicycle track information of each bicycle access point;

classifying the bicycle picking and placing points according to different characteristics according to the acquired data;

according to the classification result, dynamically dividing adjacent bicycle picking and placing points according to the dynamic change of the demand quantity to form a plurality of subareas;

scheduling according to the subarea division result: and if the scheduling cannot meet the bicycle requirement of the subarea, executing the steps to re-divide the subarea.

10. The automatic scheduling method based on subdivision scheme as claimed in claim 5, wherein: the dynamic division of the subareas can comprise classifying the picking and placing points of the bicycle according to different characteristics, namely dividing according to time characteristics and grades according to requirements, and obtaining a time-division station, a full-time-division station and a common station.

11. The automatic scheduling method based on subdivision scheme as claimed in claim 9, wherein: according to the classification result, dynamically dividing adjacent bicycle picking and placing points according to the dynamic change of the requirements to form a plurality of dynamic subareas, wherein the method specifically comprises the following steps:

Aiming at the time characteristics of the time-division stations and the full-time-division stations, a subarea is formed with the surrounding stations by combining the dynamic demand conditions of the surrounding stations;

combining sites with high complementarity in common sites to serve as a picking and placing point;

dynamically selecting a pick-and-place point with the largest real-time demand and stable demand in a set area range as a main station;

division of the sub-regions is performed according to the tree branch principle: and selecting the most appropriate surrounding sites to be combined by taking the main site as a center and adopting a complementarity principle and a shared trip attraction principle to form a subarea.

12. The automatic scheduling method based on subdivision scheme as claimed in claim 9, wherein: scheduling according to the subarea division result, specifically: and aiming at the obtained subareas, carrying out bicycle adjustment among all taking and placing points in the subareas, when the internal adjustment cannot meet the requirement of the required quantity, carrying out bicycle adjustment on the subarea level, namely, the subarea level cannot meet the requirement of the required quantity, and carrying out subarea division again.

13. An automatic scheduling system based on subarea division is characterized in that: a control platform comprising the shared bicycle flow system of any one of claims 1-4, and sending scheduling instructions to the shared bicycle flow system; the control platform comprises a shared bicycle demand prediction system and a dynamic subarea division scheduling system;

The shared bicycle demand prediction system is configured to perform the shared bicycle demand prediction method in the automatic scheduling method based on subdivision according to any one of claims 5-12;

alternatively, the dynamic subdivision scheduling system is configured for performing a dynamic subdivision scheduling method of the subdivision-based automatic scheduling method as claimed in any one of claims 5-12.