1. A method for elevator destination prediction, the method comprising:

responding to an external calling instruction, and acquiring external calling time and an external calling floor corresponding to the external calling instruction;

inputting the external calling time and the external calling floor into a target layer prediction model to obtain an internal calling floor output by the target layer prediction model;

according to the internal calling floor, a target layer predicted value corresponding to the external calling instruction is obtained; and the target layer predicted value is used for planning the path of the elevator.

2. The method of claim 1, further comprising:

inputting the external calling time and the external calling floor into the target layer prediction model to obtain the internal calling number output by the target layer prediction model;

and obtaining a target floor number predicted value corresponding to the target floor number predicted value according to the calling number, and planning a path of the elevator according to the target floor number predicted value and the target floor number predicted value.

3. The method of claim 2, wherein the step of obtaining an external call time and an external call floor corresponding to the external call instruction in response to the external call instruction is preceded by the step of:

acquiring a training sample and a sample label; the training samples comprise an external calling time sample and an external calling floor sample, and the sample labels comprise an internal calling floor label and an internal calling number label;

inputting the training sample into a target layer prediction model to be trained to obtain an internal calling floor and the number of internal calling persons output by the target layer prediction model to be trained;

and adjusting parameters of the target layer prediction model to be trained by comparing the internal calling floor with the internal calling floor labels and comparing the number of the internal calling persons with the internal calling person number labels to obtain the target layer prediction model.

4. The method of claim 3, wherein the obtaining training samples and sample labels comprises:

acquiring a video image in an elevator car;

obtaining the label of the number of people calling in the house by identifying the video image;

or the like, or, alternatively,

acquiring the reduced weight of the elevator car when a passenger goes out of the elevator;

determining the number of people calling in tag according to the reduced weight of the elevator car.

5. The method of claim 3, wherein the obtaining of the training sample and the sample label further comprises:

when an internal calling instruction is received, acquiring a floor corresponding to the internal calling instruction;

and obtaining the internal calling floor label according to the floor corresponding to the internal calling instruction.

6. The method according to claim 1, wherein after the step of obtaining the target floor prediction value corresponding to the outbound call according to the inbound call floor, the method further comprises:

acquiring a target layer actual value;

comparing the target layer predicted value with the target layer actual value to obtain a target layer deviation;

if the target layer deviation exceeds a preset deviation threshold, updating the target layer prediction model to obtain an updated target layer prediction model;

and determining a target layer predicted value corresponding to the next external calling instruction according to the updated target layer prediction model.

7. The method according to claim 1, wherein after the step of obtaining the target floor prediction value corresponding to the outbound call according to the inbound call floor, the method further comprises:

when a stair dispatching request is received, obtaining a stair to be dispatched floor corresponding to the stair dispatching request;

if the to-be-dispatched elevator floor is matched with the target floor predicted value, obtaining an elevator identification corresponding to the target floor predicted value;

and determining a target elevator according to the elevator identification, and controlling the target elevator to respond to the elevator dispatching request.

8. An elevator destination floor prediction apparatus, comprising:

the acquisition module is used for responding to an external calling instruction and acquiring external calling time and an external calling floor corresponding to the external calling instruction;

the model prediction module is used for inputting the external calling time and the external calling floor into a target layer prediction model to obtain an internal calling floor output by the target layer prediction model;

the destination layer determining module is used for obtaining a destination layer predicted value corresponding to the external calling instruction according to the internal calling floor; and the target layer predicted value is used for planning the path of the elevator.

9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 7.

10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.

Background

Passengers currently board a vertical elevator, usually via a control panel in the elevator car, give the desired destination floor. The destination floor is obtained in advance, namely the elevator control system can obtain the destination floor which the passenger expects to reach before the passenger triggers the control panel. Through obtaining the destination floor in advance, can plan elevator traffic route in advance according to the destination floor, shorten passenger latency, improve elevator operating efficiency.

In the prior art, a passenger usually swipes a card when calling outside, or a destination floor selector is used to acquire a destination floor in advance, so that although the destination floor can be accurately guided, the cost of using an elevator by a user is increased.

Therefore, the current technology for acquiring the elevator destination floor in advance has the problem of higher cost.

Disclosure of Invention

In view of the above, it is desirable to provide a method, an apparatus, a computer device, and a storage medium for predicting a destination floor of an elevator, which can reduce costs.

A method of elevator destination prediction, the method comprising:

responding to an external calling instruction, and acquiring external calling time and an external calling floor corresponding to the external calling instruction;

inputting the external calling time and the external calling floor into a target layer prediction model to obtain an internal calling floor output by the target layer prediction model;

according to the internal calling floor, a target layer predicted value corresponding to the external calling instruction is obtained; and the target layer predicted value is used for planning the path of the elevator.

In one embodiment, the method further comprises:

inputting the external calling time and the external calling floor into the target layer prediction model to obtain the internal calling number output by the target layer prediction model;

and obtaining a target floor number predicted value corresponding to the target floor number predicted value according to the calling number, and planning a path of the elevator according to the target floor number predicted value and the target floor number predicted value.

In one embodiment, before the step of acquiring an external call time and an external call floor corresponding to an external call instruction in response to the external call instruction, the method further includes:

acquiring a training sample and a sample label; the training samples comprise an external calling time sample and an external calling floor sample, and the sample labels comprise an internal calling floor label and an internal calling number label;

inputting the training sample into a target layer prediction model to be trained to obtain an internal calling floor and the number of internal calling persons output by the target layer prediction model to be trained;

and adjusting parameters of the target layer prediction model to be trained by comparing the internal calling floor with the internal calling floor labels and comparing the number of the internal calling persons with the internal calling person number labels to obtain the target layer prediction model.

In one embodiment, the obtaining the training sample and the sample label includes:

acquiring a video image in an elevator car;

obtaining the label of the number of people calling in the house by identifying the video image;

or the like, or, alternatively,

acquiring the reduced weight of the elevator car when a passenger goes out of the elevator;

determining the number of people calling in tag according to the reduced weight of the elevator car.

In one embodiment, the obtaining the training sample and the sample label further includes:

when an internal calling instruction is received, acquiring a floor corresponding to the internal calling instruction;

and obtaining the internal calling floor label according to the floor corresponding to the internal calling instruction.

In one embodiment, after the step of obtaining the target floor prediction value corresponding to the external call instruction according to the internal call floor, the method further includes:

acquiring a target layer actual value;

comparing the target layer predicted value with the target layer actual value to obtain a target layer deviation;

if the target layer deviation exceeds a preset deviation threshold, updating the target layer prediction model to obtain an updated target layer prediction model;

and determining a target layer predicted value corresponding to the next external calling instruction according to the updated target layer prediction model.

In one embodiment, after the step of obtaining the target floor prediction value corresponding to the external call instruction according to the internal call floor, the method further includes:

when a stair dispatching request is received, obtaining a stair to be dispatched floor corresponding to the stair dispatching request;

if the to-be-dispatched elevator floor is matched with the target floor predicted value, obtaining an elevator identification corresponding to the target floor predicted value;

and determining a target elevator according to the elevator identification, and controlling the target elevator to respond to the elevator dispatching request.

An elevator destination floor prediction device, the device comprising:

the acquisition module is used for responding to an external calling instruction and acquiring external calling time and an external calling floor corresponding to the external calling instruction;

the model prediction module is used for inputting the external calling time and the external calling floor into a target layer prediction model to obtain an internal calling floor output by the target layer prediction model;

the destination layer determining module is used for obtaining a destination layer predicted value corresponding to the external calling instruction according to the internal calling floor; and the target layer predicted value is used for planning the path of the elevator.

A computer device comprising a memory and a processor, the memory storing a computer program, the processor implementing the following steps when executing the computer program:

responding to an external calling instruction, and acquiring external calling time and an external calling floor corresponding to the external calling instruction;

inputting the external calling time and the external calling floor into a target layer prediction model to obtain an internal calling floor output by the target layer prediction model;

according to the internal calling floor, a target layer predicted value corresponding to the external calling instruction is obtained; and the target layer predicted value is used for planning the path of the elevator.

A computer-readable storage medium, on which a computer program is stored which, when executed by a processor, carries out the steps of:

responding to an external calling instruction, and acquiring external calling time and an external calling floor corresponding to the external calling instruction;

inputting the external calling time and the external calling floor into a target layer prediction model to obtain an internal calling floor output by the target layer prediction model;

according to the internal calling floor, a target layer predicted value corresponding to the external calling instruction is obtained; and the target layer predicted value is used for planning the path of the elevator.

According to the elevator target layer prediction method, the device, the computer equipment and the storage medium, the external calling time and the external calling floor corresponding to the external calling instruction are obtained by responding to the external calling instruction, when a user calls an elevator outside an elevator car, the time of calling the elevator by the user and the floor where the user is located are obtained, the external calling time and the external calling floor are input into the target layer prediction model to obtain the internal calling floor output by the target layer prediction model, the target layer prediction value corresponding to the external calling instruction is obtained according to the internal calling floor, and the target floor of the user can be predicted according to the time of calling the elevator by the user and the floor where the user is located, so that the target layer can be obtained without swiping a card or a target layer selector, and the cost of obtaining the elevator target layer is reduced.

Drawings

Fig. 1 is a diagram of an application environment of a method for predicting a destination floor of an elevator in one embodiment;

fig. 2 is a schematic flow chart of a method for predicting a destination floor of an elevator according to an embodiment;

fig. 3 is a schematic flow diagram of elevator path planning in one embodiment;

fig. 4 is a schematic flow chart of a method for predicting a destination floor of an elevator in another embodiment;

fig. 5 is a block diagram showing a construction of an elevator destination prediction apparatus in one embodiment;

FIG. 6 is a diagram illustrating an internal structure of a computer device according to an embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

The method for predicting the elevator destination floor can be applied to the application environment shown in fig. 1. The elevator group control system 102 communicates with the cloud 106 through the communication module 104, and the control system 108 of the elevator N (N is 1, 2, …, N) is connected to the group control system 102 and transmits an internal call signal or an external call signal to the group control system 102. The group control system 102 may include a prediction model, a real-time prediction module, and a calling data acquisition module, where the prediction model is used to predict a target layer through a target layer prediction model, the real-time prediction module is used to update the target layer prediction model in real time, and the calling data acquisition module is used to acquire a training sample of the target layer prediction model.

In one embodiment, as shown in fig. 2, there is provided a method for predicting a destination floor of an elevator, which is illustrated by applying the method to the group control system 102 in fig. 1, and includes the following steps:

step S210, responding to the external calling instruction, and acquiring external calling time and external calling floor corresponding to the external calling instruction.

The external calling command can be a command generated when a user calls an elevator outside an elevator car.

The external calling time can be the time when the user sends the external calling instruction. The external calling floor can be a floor where the user sends the external calling instruction.

In the specific implementation, when a user calls an elevator outside an elevator car, an external calling instruction can be generated and sent to the control system of each elevator, the control system transmits the external calling instruction to the group control system, and when the group control system receives the external calling instruction, the current time can be collected and used as the external calling time. When the control system receives the external calling instruction, the floor where the user sends the external calling instruction can be obtained and used as the external calling floor.

For example, a building has 6 elevators, a user can press an up button outside an elevator car of a 1-floor elevator to generate an external call command, and control systems of the 6 elevators can receive the external call command and transmit the external call command to a group control system. The cluster control system can be configured with a clock, and when an external calling instruction is received, the current time can be collected as the external calling time. The control system of 6 elevators can also acquire the floor for sending the external call instruction, and transmit the floor to the group control system as the external call floor.

And step S220, inputting the external calling time and the external calling floor into the target layer prediction model to obtain an internal calling floor output by the target layer prediction model.

The target layer prediction model may be a model obtained by training through a machine learning algorithm, and is used for predicting the target floor of the user.

The internal calling floor can be a floor pressed by a user when the user enters the elevator car.

In specific implementation, a target layer prediction model can be preset in the group control system, after the external calling time and the external calling floor are obtained, the external calling time and the external calling floor can be input into the target layer prediction model, and the target layer prediction model can output the internal calling floor.

For example, the external call time 7:01 and the external call floor 1 are input into a destination floor prediction model, and the destination floor prediction model can predict that the user will press the floor 5 on the control panel in the elevator car after entering the elevator car, namely, the internal call floor is 5.

The target layer prediction model can be obtained by training in the cloud. Can call the ladder data acquisition module through crowd control system and gather the time of calling outside, call the floor in, call the number of people in, and call the time outside and call the floor outside as the training sample, call the floor in and call the number of people in as the sample label in, send for the high in the clouds through communication module, the high in the clouds can utilize training sample and sample label after receiving training sample and sample label, train through the machine learning algorithm, obtain the target layer prediction model that trains, and return it to crowd control system, the storage is in the prediction model module of crowd control system.

The number of passengers who enter the car at the same floor at the same time and who exit the car at the same floor may be the number of passengers.

The Machine learning algorithm for training the target layer prediction model may be LSTM (Long-Short Term Memory) or LightGBM (Light Gradient Boosting Machine).

External calling time	Floor calling outside	Floor calling in	Number of persons calling in
				7:01	1	5	3
7:05	3	6	2
				7:05	8	3	1
7:07	2	4	1
				……	……	……	……
……	……	……	……

TABLE 1

Table 1 provides examples of training samples and sample labels. According to the table 1, at 7:01, a floor 1 receives an external calling instruction, the group control system dispatches a ladder to the floor 1, and calls the floor 5 after entering the lift car, and the passengers leaving the floor 5 have 3 persons; at 7:05, the floor 3 receives an external calling instruction, the group control system dispatches a ladder for the floor 3, the 6 floors are called after the group control system enters the elevator car, and 2 passengers leaving the 6 floors are present; meanwhile, at 7:05, the floor 8 receives an external calling instruction, the group control system dispatches a ladder for the 8 th floor, calls the 3 rd floor after entering the lift car, and the passengers leaving the 3 rd floor have 1 person; and at 7:07, the floor 2 receives an external calling instruction, the group control system dispatches a ladder for the floor 2, calls the floor 4 after entering the elevator car, and calls 1 person for passengers going out of the floor 4, and so on.

Step S230, according to the internal calling floor, obtaining a target floor predicted value corresponding to the external calling instruction; the predicted value of the destination floor is used for planning the path of the elevator.

The target floor prediction value can be a predicted floor which a user desires to reach.

In specific implementation, after the group control system obtains the internal calling floor through the target floor prediction model, the internal calling floor can be used as a target floor prediction value corresponding to the current external calling instruction, and path planning is carried out on each elevator according to the target floor prediction value.

For example, after the external calling time and the external calling floor are input into the target floor prediction model and the internal calling floor 5 is output, the group control system can use 5 as a target floor prediction value, namely, the floor which the user desires to reach is predicted to be 5, and the group control system can plan the path of each elevator according to the target floor prediction value before the user enters the elevator car for internal calling, even before the user connects the elevator for the external calling and reaches the floor where the user is located for the external calling.

It should be noted that the group control system may further obtain an actual destination floor of the user, for example, the user may call in after entering the elevator, the elevator generates an inner call signal and sends the inner call signal to the group control system, the group control system obtains the actual destination floor of the user according to the inner call signal, and may further acquire the actual inner call floor of the passenger through the camera to obtain the actual destination floor, the group control system may further compare the predicted destination floor with the actual destination floor, and if a difference between the predicted destination floor and the actual destination floor exceeds a preset threshold, the current outer call time, the current outer call floor, the actual inner call floor, and the actual inner call number may be used as a set of training data, and the target floor prediction model is updated through the real-time prediction module to improve the accuracy of the target floor prediction model.

According to the elevator target layer prediction method, the external calling time and the external calling floor corresponding to the external calling instruction are obtained by responding to the external calling instruction, when a user calls an elevator outside an elevator car, the time and the floor where the user calls the elevator are obtained, the external calling time and the external calling floor are input into the target layer prediction model to obtain the internal calling floor output by the target layer prediction model, the target layer prediction value corresponding to the external calling instruction is obtained according to the internal calling floor, and the target floor of the user can be predicted according to the time and the floor where the user calls the elevator, so that the target layer can be obtained without swiping a card or a target layer selector, and the elevator target layer acquisition cost is reduced.

And by carrying out destination layer prediction, the elevator can be subjected to path planning according to the destination layer prediction value before the user enters the elevator car for calling in, even before the user for calling out connects the elevator and arrives at the floor where the user for calling out is located, so that the operation efficiency of the elevator is improved.

In one embodiment, the method for predicting a destination floor of an elevator further includes: inputting the external calling time and the external calling floor into a target layer prediction model to obtain the internal calling number output by the target layer prediction model; and obtaining a target floor number predicted value corresponding to the target floor number predicted value according to the number of calling persons, and planning a path of the elevator according to the target floor number predicted value and the target floor number predicted value.

The number of passengers who enter the car at the same floor at the same time and who exit the car at the same floor may be the number of passengers.

The predicted number of people at the destination floor can be the predicted number of people expected to reach the specified floor.

In specific implementation, the target layer prediction model can also predict the number of people calling in, and the number of people calling in is used as a target layer people number prediction value. For example, when the calling-out time is 7:01 and the calling-out floor is 1, the target floor prediction model can predict that the calling-out floor in the user is 5, and also can predict that the number of passengers on the calling-in floor is 3, namely the number of calling-in passengers is 3, and 3 is taken as a target floor number prediction value, so that 3 people will exit from the 5 th floor. The group control system can plan the elevator path in advance according to the target floor predicted value and the target floor number predicted value, specifically, when receiving the elevator dispatching request, the group control system can obtain the elevator dispatching floor to be dispatched corresponding to the elevator dispatching request, if the elevator dispatching floor to be dispatched is the target floor predicted value of the elevator 1, whether the target floor number predicted value exceeds a preset number threshold value can be further judged, and if the target floor number predicted value exceeds the preset number threshold value, the elevator 1 can be assigned to respond to the elevator dispatching request.

In the embodiment, the external calling time and the external calling floor are input into the target floor prediction model to obtain the internal calling number output by the target floor prediction model, the internal calling number can be predicted, the target floor number prediction value corresponding to the target floor prediction value is obtained according to the internal calling number, the number of users expecting to reach the target floor can be predicted, so that the elevator is subjected to path planning according to the target floor prediction value and the target floor number prediction value, and the elevator operation efficiency is improved.

In an embodiment, before the step S210, the method further includes:

step S201, obtaining a training sample and a sample label; the training samples comprise external calling time samples and external calling floor samples, and the sample labels comprise internal calling floor labels and internal calling number labels;

step S202, inputting a training sample into a target layer prediction model to be trained to obtain an internal calling floor and the number of internal calling persons output by the target layer prediction model to be trained;

step S203, comparing the internal calling floor with the internal calling floor label, and comparing the number of internal calling persons with the internal calling person number label, and adjusting the parameters of the target layer prediction model to be trained to obtain the target layer prediction model.

In the specific implementation, the group control system can collect external calling time, external calling floors, internal calling floors and internal calling number, the external calling time and the external calling floors are used as training samples, the internal calling floors and the internal calling number are used as sample labels, the training samples are input into a target layer prediction model to be trained, the internal calling floors and the internal calling number output by the target layer prediction model can be obtained, the internal calling floors and the internal calling number labels are compared, the internal calling number labels and the internal calling number labels are compared to obtain a comparison result, how to adjust parameters of the target layer prediction model to be trained can be determined according to the comparison result, and the target layer prediction model can be obtained by adjusting the parameters of the target layer prediction model to be trained.

The target layer prediction model may be trained by using a machine learning algorithm, which may be LSTM or LightGBM.

In this embodiment, by obtaining a training sample and a sample label, the training sample is input into a target layer prediction model to be trained, an internal calling floor and an internal calling number output by the target layer prediction model to be trained are obtained, the internal calling floor is compared with the internal calling floor label, the internal calling number is compared with the internal calling number label, a parameter of the target layer prediction model to be trained is adjusted, the target layer prediction model is obtained, and model parameters can be adjusted by comparing a predicted value and an actual value of the target layer prediction model, so that accuracy of model prediction is improved.

In one embodiment, the step S201 includes: acquiring a video image in an elevator car, and identifying the video image to obtain a tag of the number of people calling in the elevator car; or when the passenger goes out of the elevator, the weight reduced by the elevator car is obtained, and the number of people calling in the elevator is determined according to the weight reduced by the elevator car.

In concrete the realization, can install the camera in the elevator car, gather the video image in the car through the camera, give the crowd control system, the crowd control system can confirm the number of going out the terraced people on the destination floor through discerning the video image who receives to call the number label in with the number of going out the terraced people on destination floor as. The elevator car can reach a destination floor, when the car door is opened, the passenger goes out of the elevator, the weight reduced by the elevator car is collected through the control system, the weight is divided by the preset weight of the passenger, the number of people going out of the elevator is obtained, and the number of people going out of the elevator is used as an in-call number label.

In the embodiment, the number of people calling in can be accurately determined by acquiring the video image in the elevator car and identifying the video image to obtain the number of people calling in tag, so that the accuracy of the number of people calling in tag is improved; when the passenger goes out of the elevator, the weight reduced by the elevator car is obtained, the number of people calling in is determined according to the weight reduced by the elevator car, the number of people calling in can be rapidly determined, and the efficiency of obtaining the number of people calling in is improved.

In an embodiment, the step S201 further includes: when an internal calling instruction is received, a floor corresponding to the internal calling instruction is obtained; and obtaining the internal calling floor label according to the floor corresponding to the internal calling instruction.

The internal calling instruction can be an instruction sent by triggering a floor key on the control panel after a user enters the elevator car.

In the specific implementation, when a user enters an elevator car and sends an internal calling instruction by triggering a floor key on a control panel, the internal calling instruction can be transmitted to a group control system through the control system, the group control system can determine the floor key triggered by the user according to the received internal calling instruction, a target floor expected to be reached by the user is obtained, and the target floor is used as an internal calling floor label.

In this embodiment, through when receiving the internal calling instruction, obtain the floor that the internal calling instruction corresponds, obtain the internal calling floor label according to the floor that the internal calling instruction corresponds, can gather the internal calling floor label high-efficiently accurately.

In an embodiment, after the step S230, the method further includes: acquiring a target layer actual value; comparing the target layer predicted value with the target layer actual value to obtain a target layer deviation; if the target layer deviation exceeds a preset deviation threshold, updating the target layer prediction model to obtain an updated target layer prediction model; and determining a target layer predicted value corresponding to the next external calling instruction according to the updated target layer prediction model.

The actual value of the destination floor may be a floor actually reached by the user.

In the specific implementation, a target layer actual value can be acquired through a camera installed in an elevator car, an absolute difference is calculated between a target layer predicted value and the target layer actual value to obtain a target layer deviation, the target layer deviation is compared with a preset deviation threshold, if the target layer deviation does not exceed the deviation threshold, the target layer prediction model is not required to be processed, otherwise, if the target layer deviation exceeds the deviation threshold, the target layer prediction model can be updated to obtain an updated target layer prediction model, and the updated target layer prediction model is used for subsequent target layer prediction.

In this embodiment, a target layer actual value is obtained, a target layer deviation is obtained by comparing a target layer predicted value with the target layer actual value, if the target layer deviation exceeds a preset deviation threshold, a target layer prediction model is updated to obtain an updated target layer prediction model, a target layer predicted value corresponding to a next external call instruction is determined according to the updated target layer prediction model, the target layer prediction model can be updated in real time in a prediction process, and accuracy of the target layer prediction model is improved.

In an embodiment, after the step S201, the method further includes: when a stair dispatching request is received, obtaining a stair to be dispatched floor corresponding to the stair dispatching request; if the to-be-dispatched elevator floor is matched with the target floor predicted value, obtaining an elevator identifier corresponding to the target floor predicted value; and determining a target elevator according to the elevator identification, and controlling the target elevator to respond to the elevator dispatching request.

In the concrete implementation, when the user calls outside, the control system can generate a calling request and send the calling request to the group control system, the calling request can carry a floor where the user is called outside, and the group control system can acquire the floor where the user is called outside in the calling request and serve as the floor to be called. The group control system can also compare the predicted value of the target layer of each elevator with the floor to be dispatched, when the predicted value of the target layer of one elevator is the same as that of the floor to be dispatched, the elevator identification of the elevator can be obtained, the elevator is taken as the target elevator, and the elevator is controlled to respond to the elevator dispatching request of the user. Further, the group control system can also determine a target elevator according to the predicted value of the number of people at the destination floor, if the predicted value of the number of people at the destination floor exceeds a preset threshold value of the number of people, the predicted destination floor is indicated, the number of people going out of the elevator exceeds a certain number, new passengers can be allowed to enter the elevator car, the elevator can respond to an elevator dispatching request, otherwise, if the predicted value of the number of people at the destination floor does not exceed the preset threshold value of the number of people, the predicted destination floor is indicated, the number of people going out of the elevator does not exceed the certain number, the elevator is limited by the load of the elevator, new passengers cannot be allowed to enter the elevator, and the elevator can not respond to the elevator dispatching request.

Fig. 3 is a flow diagram of elevator path planning. According to fig. 3, the elevator path planning may comprise the following steps:

step S310, generating an external call;

step S320, predicting the number of people calling in and the floor N calling in, and dispatching an elevator M to receive the elevator;

in step S330, for the external call of the floor N, the elevator M is preferentially sent to respond.

In the concrete implementation, when a user A calls a floor L externally, the group control system can respond to the external calling instruction, obtain the external calling time T and the external calling floor L corresponding to the external calling instruction, input the external calling time T and the external calling floor L into the target layer prediction model to obtain the internal calling number K and the internal calling floor N output by the target layer prediction model, and can dispatch the elevator M which is closest to the internal calling floor N and can accommodate K persons to receive the elevator. If the user B generates an external call on the floor N within a period of time after the user A sends the external call, the elevator M can be preferentially dispatched to respond to the external call of the user B on the floor N because the destination floor of the elevator M is N, the elevator M can convey the user A to the floor N, and the user B is connected to the elevator on the floor N.

In this embodiment, through when receiving the group ladder request, obtain the group ladder floor of waiting that the group ladder request corresponds, if wait to send the ladder floor and target floor predicted value phase-match, then obtain the elevator sign that the target floor predicted value corresponds, confirm the target elevator according to the elevator sign to control target elevator response and send the ladder request, can carry out route planning and dispatch to the elevator based on the target floor prediction result, improve elevator operating efficiency.

Fig. 4 is a flow chart of a method for predicting a destination floor of an elevator. According to fig. 4, the elevator destination prediction method may include the steps of:

step S410, obtaining a training sample and a sample label; the training samples comprise an external calling time sample and an external calling floor sample, and the sample labels comprise an internal calling floor label and an internal calling number label;

step S420, inputting the training sample into a target layer prediction model to be trained to obtain an internal calling floor and the number of internal calling persons output by the target layer prediction model to be trained;

step S430, comparing the calling-in floors with the calling-in floor labels, and comparing the calling-in number with the calling-in number labels, and adjusting parameters of the target layer prediction model to be trained to obtain a target layer prediction model;

step S440, responding to an external calling instruction, and acquiring external calling time and an external calling floor corresponding to the external calling instruction;

step S450, inputting the external calling time and the external calling floor into the target layer prediction model to obtain an internal calling floor and the number of people to be called;

step S460, obtaining a target floor predicted value and a target floor number predicted value corresponding to the external calling instruction according to the internal calling floor and the internal calling number;

step S470, when a stair dispatching request is received, obtaining a to-be-dispatched stair floor corresponding to the stair dispatching request;

step S480, if the to-be-dispatched elevator floor is matched with the target floor predicted value and the target floor number predicted value is within a preset interval, obtaining an elevator identification corresponding to the target floor predicted value;

and step S490, determining a target elevator according to the elevator identification, and controlling the target elevator to respond to the elevator dispatching request.

It should be understood that although the various steps in the flow charts of fig. 2-4 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 2-4 may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, which are not necessarily performed in sequence, but may be performed in turn or alternately with other steps or at least some of the other steps.

In one embodiment, as shown in fig. 5, there is provided an elevator destination prediction apparatus including: an obtaining module 510, a model prediction module 520, and a destination layer determination module 530, wherein:

an obtaining module 510, configured to, in response to an external call instruction, obtain an external call time and an external call floor corresponding to the external call instruction;

the model prediction module 520 is configured to input the external calling time and the external calling floor into a target layer prediction model to obtain an internal calling floor output by the target layer prediction model;

a destination layer determining module 530, configured to obtain a destination layer predicted value corresponding to the external calling instruction according to the internal calling floor; and the target layer predicted value is used for planning the path of the elevator.

In one embodiment, the device for predicting a destination floor of an elevator further includes:

the people number prediction module is used for inputting the external calling time and the external calling floor into the target layer prediction model to obtain the internal calling number output by the target layer prediction model;

and the destination floor number determining module is used for obtaining a destination floor number predicted value corresponding to the destination floor predicted value according to the calling number, and planning a path of the elevator according to the destination floor predicted value and the destination floor number predicted value.

In one embodiment, the device for predicting a destination floor of an elevator further includes:

the sample acquisition module is used for acquiring a training sample and a sample label; the training samples comprise an external calling time sample and an external calling floor sample, and the sample labels comprise an internal calling floor label and an internal calling number label;

the training module is used for inputting the training samples into a target layer prediction model to be trained to obtain internal calling floors and internal calling people number output by the target layer prediction model to be trained;

and the parameter adjusting module is used for adjusting the parameters of the target layer prediction model to be trained by comparing the calling-in floors with the calling-in floor labels and comparing the calling-in number with the calling-in number labels to obtain the target layer prediction model.

In one embodiment, the sample acquiring module is further configured to acquire a video image of the elevator car; obtaining the label of the number of people calling in the house by identifying the video image; or, when the passenger goes out of the elevator, acquiring the reduced weight of the elevator car; determining the number of people calling in tag according to the reduced weight of the elevator car.

In an embodiment, the sample obtaining module is further configured to, when an internal call instruction is received, obtain a floor corresponding to the internal call instruction; and obtaining the internal calling floor label according to the floor corresponding to the internal calling instruction.

In one embodiment, the device for predicting a destination floor of an elevator further includes:

the actual value acquisition module is used for acquiring the actual value of the target layer;

the comparison module is used for comparing the target layer predicted value with the target layer actual value to obtain a target layer deviation;

the model updating module is used for updating the target layer prediction model if the target layer deviation exceeds a preset deviation threshold to obtain an updated target layer prediction model;

and the real-time prediction module is used for determining a target layer prediction value corresponding to the next external calling instruction according to the updated target layer prediction model.

In one embodiment, the device for predicting a destination floor of an elevator further includes:

the device comprises a to-be-dispatched ladder floor acquisition module, a to-be-dispatched ladder floor acquisition module and a to-be-dispatched ladder floor acquisition module, wherein the to-be-dispatched ladder floor acquisition module is used for acquiring a to-be-dispatched ladder floor corresponding to a ladder dispatching request when the ladder dispatching request is received;

the elevator identification obtaining module is used for obtaining an elevator identification corresponding to the target floor predicted value if the to-be-dispatched elevator floor is matched with the target floor predicted value;

and the elevator dispatching request response module is used for determining a target elevator according to the elevator identification and controlling the target elevator to respond to the elevator dispatching request.

For the specific definition of the elevator destination floor prediction device, reference may be made to the above definition of the elevator destination floor prediction method, which is not described herein again. The modules in the elevator destination layer prediction device can be wholly or partially realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.

In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in fig. 6. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used for storing elevator destination floor prediction data. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method for elevator destination prediction.

Those skilled in the art will appreciate that the architecture shown in fig. 6 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, a computer arrangement is provided comprising a memory and a processor, the memory storing a computer program which, when executed by the processor, causes the processor to carry out the steps of the above-mentioned elevator destination prediction method. Here, the steps of an elevator destination prediction method may be the steps in an elevator destination prediction method of the above embodiments.

In one embodiment, a computer-readable storage medium is provided, in which a computer program is stored, which, when being executed by a processor, causes the processor to carry out the steps of the above-mentioned elevator destination prediction method. Here, the steps of an elevator destination prediction method may be the steps in an elevator destination prediction method of the above embodiments.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database or other medium used in the embodiments provided herein can include at least one of non-volatile and volatile memory. Non-volatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical storage, or the like. Volatile Memory can include Random Access Memory (RAM) or external cache Memory. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others.

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.