Server and method for determining candidate device

Abstract

The embodiment of the invention provides a server and a method for determining candidate equipment. The method comprises the following steps: receiving a track parameter corresponding to a real-time moving track of a mobile terminal sent by the mobile terminal; processing the trajectory parameters based on a preset size model to obtain trajectory features; processing the trajectory features based on a preset size model, and outputting a predicted size; calculating a prediction size to obtain a selection probability of each binding device; selecting a candidate device corresponding to at least one highest selection probability from the plurality of binding devices; and sending the indication information of each candidate device to the mobile terminal. According to the embodiment of the invention, few candidate devices can be recommended to the user, so that the user can select the target device from the few candidate devices, the user operation is simplified, and the user experience is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of servers, and in particular to a server and a method for determining candidate devices. BACKGROUND

[0002] In the related art, with the continuous development of the control mode of smart home binding devices, virtual remote controllers gradually become mainstream interactive interfaces. Through a single mobile terminal, such as a smart phone, integrated whole-house binding device control is provided to support value-added services such as visual direct operation, intuitive interaction expectations of "what you see is what you get", scene mode configuration, and energy consumption monitoring, and meet the needs of users for intelligent management.

[0003] However, users need to scroll through dozens of icons on the operation interface of the mobile terminal to find them, and the operation path is long. In the case of multiple binding devices and multiple rooms, misoperation is easy to occur, for example, misselecting adjacent room binding devices, which brings great inconvenience to users and reduces user experience. SUMMARY

[0004] To solve the above problems, the embodiments of the present application provide a server and a method for determining candidate devices, which can recommend a few candidate devices to the user, so that the user can select the target device from the few candidate devices, thereby simplifying the user operation and improving the user experience. Specifically:

[0005] The first aspect of the embodiments of the present application provides a method for determining candidate devices, comprising:

[0006] receiving trajectory parameters corresponding to a real-time movement trajectory of a mobile terminal sent by the mobile terminal;

[0007] processing the trajectory parameters based on a preset size model to obtain trajectory features;

[0008] processing the trajectory features based on the preset size model to output predicted sizes;

[0009] calculating the predicted sizes to obtain selection probabilities of each binding device;

[0010] selecting candidate devices corresponding to at least one of the highest selection probabilities from a plurality of binding devices;

[0011] sending indication information of each candidate device to the mobile terminal.

[0012] The above technical scheme has the following beneficial effects or advantages: the trajectory parameters corresponding to the mobile terminal are received by the server to obtain trajectory features; the trajectory features are processed by the server based on a preset size model to output predicted sizes; the predicted sizes are calculated to obtain selection probabilities of each binding device; candidate devices corresponding to the highest selection probability are selected from the plurality of binding devices; and the indication information of each candidate device is sent to the mobile terminal. Compared with the related art, the user is recommended with fewer candidate devices instead of needing to select a target device from all binding devices, which greatly simplifies the user operation and improves the user experience.

[0013] In some embodiments, the trajectory parameters include a plurality of coordinate points;

[0014] The trajectory features include a standard trajectory length feature, a length feature, and a shape feature;

[0015] The trajectory features are used to represent the movement trajectory of the mobile terminal;

[0016] The trajectory parameters are processed based on the preset size model to obtain trajectory features, including:

[0017] Each of the coordinate points is processed based on the preset size model to obtain a trajectory length parameter, a spatial range parameter, and a shape parameter;

[0018] The trajectory length parameter and a size factor are multiplied based on the preset size model to obtain a standard trajectory length feature;

[0019] The spatial range parameter is divided based on the preset size model to obtain a length-width feature;

[0020] The shape parameter is divided based on the preset size model to obtain a shape feature.

[0021] The above technical scheme has the following beneficial effects or advantages: the trajectory features are obtained by calculating the standard trajectory length feature, the length feature, and the shape feature.

[0022] In some embodiments, before the predicted sizes are calculated to obtain the selection probabilities of each binding device, the method further includes:

[0023] Real-time pitch angles, real-time horizontal angles, position parameters, position weights, angle weights, size weights, historical behavior parameters, and historical behavior weights of the mobile terminal are obtained;

[0024] The historical behavior parameters are obtained based on the historical behavior of the user;

[0025] The position parameters are obtained based on the position of the mobile terminal and the position of each binding device.

[0026] The technical scheme has the following beneficial effects or advantages: the parameters for calculating the selection probability are obtained, so as to facilitate the calculation of the selection probability.

[0027] In some embodiments, the predicted size is calculated to obtain the selection probability of each binding device, comprising:

[0028] For each binding device, the predicted size and the size weight are multiplied to obtain a size weight term;

[0029] The real-time pitch angle and real-time horizontal angle of the mobile terminal and the angle weight are multiplied to obtain an angle weight term;

[0030] The position parameter and the position weight are multiplied to obtain a position weight term;

[0031] The historical behavior parameter and the historical behavior weight are multiplied to obtain a historical behavior weight term;

[0032] The size weight term, the angle weight term, the position weight term and the historical behavior weight term are added to obtain the selection probability of the binding device.

[0033] The technical scheme has the following beneficial effects or advantages: the highest at least one selection probability is selected from the selection probabilities of the binding devices, and the binding device corresponding to each selection probability in the highest at least one selection probability is taken as a candidate device.

[0034] In some embodiments, the predicted size comprises a predicted length parameter and a predicted width parameter;

[0035] The predicted size and the size weight are multiplied to obtain a size weight term, comprising:

[0036] The predicted length parameter and the actual length parameter are subtracted to obtain a length difference;

[0037] The predicted width parameter and the actual width parameter are subtracted to obtain a width difference;

[0038] The length difference and the width difference are exponentially calculated to obtain a size matching degree;

[0039] The size matching degree and the size weight are multiplied to obtain the size weight term.

[0040] The technical scheme has the following beneficial effects or advantages: the selection probability of each binding device is selected, at least one selection probability with the highest value is selected, and the binding device corresponding to each selection probability with the highest value is taken as a candidate device.

[0041] In some embodiments, the angle weight includes a horizontal angle weight and a pitch angle weight; and the angle weight term includes a horizontal angle weight term and a pitch angle weight term.

[0042] The real-time pitch angle, the real-time horizontal angle and the angle weight of the mobile terminal are multiplied to obtain an angle weight term, including:

[0043] The real-time horizontal angle and the historical horizontal angle are matched to obtain a horizontal angle matching degree, and the horizontal angle matching degree and the horizontal angle weight are multiplied to obtain a horizontal angle weight term.

[0044] The real-time pitch angle and the historical pitch angle are matched to obtain a pitch angle matching degree, and the pitch angle matching degree and the pitch angle weight are multiplied to obtain a pitch angle weight term.

[0045] The pitch angle weight term and the horizontal angle weight term are added to obtain the angle weight term.

[0046] The technical scheme has the following beneficial effects or advantages: the pitch angle of the mobile terminal is obtained by a gyroscope, the horizontal angle of the mobile terminal is obtained by a compass, a sensor is not needed, calculation power is greatly saved, and delay is reduced. In addition, the angle weight term is used as other dimension data, which is conducive to eliminating errors caused by size factors, thereby improving the accuracy of the selection probability of each binding device.

[0047] In some embodiments, the historical behavior parameter includes a historical moving track time parameter and a historical moving track frequency parameter; the historical moving track time parameter includes a historical moving track time coefficient and a historical moving track time median; the historical moving track frequency parameter includes a historical moving track frequency and a historical moving track frequency coefficient; the historical behavior weight includes a historical moving track time weight and a historical moving track frequency weight; the historical moving track time coefficient and the historical moving track time median are in one-to-one correspondence; and the historical moving track frequency and the historical moving track frequency coefficient are in one-to-one correspondence.

[0048] The historical behavior parameter and the historical behavior weight are multiplied to obtain a historical behavior weight term, including:

[0049] multiply each group of the historical movement trajectory time coefficient and the historical movement trajectory time median to obtain the historical movement trajectory time parameter;

[0050] multiply each group of the historical movement trajectory frequency and the historical movement trajectory frequency coefficient to obtain the historical movement trajectory frequency parameter;

[0051] multiply the historical movement trajectory time parameter and the historical movement trajectory time weight to obtain a historical movement trajectory time weight term;

[0052] multiply the historical movement trajectory frequency weight and the historical movement trajectory frequency parameter to obtain a historical movement trajectory frequency weight term;

[0053] add the historical movement trajectory frequency weight term and the historical movement trajectory time weight term to obtain the historical behavior weight term.

[0054] The technical scheme has the following beneficial effects or advantages: the historical behavior weight term, as other dimension data, is beneficial to eliminate errors caused by size factors, thereby improving the accuracy of the selection probability of each bound device.

[0055] The second aspect of the embodiment of the application provides a method for determining a candidate device, comprising:

[0056] receiving real-time movement trajectory of the mobile terminal sent by the mobile terminal;

[0057] calculating the matching degree of the real-time movement trajectory of the mobile terminal and each bound device;

[0058] selecting a candidate device corresponding to the highest at least one matching degree from a plurality of bound devices;

[0059] sending the indication information of each candidate device to the mobile terminal.

[0060] The technical scheme has the following beneficial effects or advantages: another method for selecting at least one candidate device from a plurality of bound devices is provided, the method is simple, thereby saving computing power, reducing delay, and improving user experience.

[0061] The third aspect of the embodiment of the application provides a server, comprising:

[0062] a communicator configured to communicate with a mobile terminal;

[0063] a controller configured to:

[0064] control the communicator to receive trajectory parameters corresponding to the real-time movement trajectory of the mobile terminal sent by the mobile terminal.

[0065] processing the trajectory parameters based on a preset size model to obtain trajectory features;

[0066] processing the trajectory features based on the preset size model to output predicted sizes;

[0067] calculating the predicted sizes to obtain selection probabilities of each binding device;

[0068] selecting, from a plurality of the binding devices, candidate devices corresponding to at least one highest selection probability;

[0069] controlling the communicator to send indication information of each of the candidate devices to the mobile terminal.

[0070] The above technical solution has the following beneficial effects or advantages: the trajectory features are obtained by the server receiving trajectory parameters corresponding to the mobile terminal; the predicted sizes are output by the server processing the trajectory features based on a preset size model; the selection probabilities of each binding device are obtained by calculating the predicted sizes; the candidate devices corresponding to at least one highest selection probability are selected from a plurality of binding devices; and the indication information of each candidate device is sent to the mobile terminal. Compared with the related art, the user is recommended with fewer candidate devices instead of needing to select a target device from all binding devices, which greatly simplifies the user operation and improves the user experience.

[0071] A fourth aspect of an embodiment of the present application provides a server, comprising:

[0072] a communicator configured to communicate with a mobile terminal;

[0073] a controller configured to:

[0074] control the communicator to receive a real-time movement trajectory of the mobile terminal sent by the mobile terminal;

[0075] calculate a matching degree of the real-time movement trajectory of the mobile terminal and each binding device;

[0076] select, from a plurality of the binding devices, candidate devices corresponding to at least one highest matching degree;

[0077] control the communicator to send indication information of each of the candidate devices to the mobile terminal.

[0078] The above technical solution has the following beneficial effects or advantages: another method for selecting at least one candidate device from a plurality of binding devices is provided, the method is simple, thereby saving computing power, reducing delay, and further improving user experience. BRIEF DESCRIPTION OF DRAWINGS

[0079] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the accompanying drawings needed to be used in the embodiments or prior art description will be briefly introduced as follows. Obviously, the accompanying drawings in the following description only only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained without creative labor on the basis of these drawings.

[0080] Figure 1 A schematic diagram of an operating scenario between a mobile terminal and a server is provided for some embodiments of the present application.

[0081] Figure 2 A structural schematic diagram of a server is provided for some embodiments of the present application.

[0082] Figure 3 An information interaction timing diagram of a first method for determining a candidate device is provided for some embodiments of the present application.

[0083] Figure 4 A schematic diagram of a remote control button is provided for some embodiments of the present application.

[0084] Figure 5 A schematic diagram of a real-time moving track is provided for some embodiments of the present application.

[0085] Figure 6 A schematic diagram of an icon is provided for some embodiments of the present application.

[0086] Figure 7 A schematic diagram of a remote control is provided for some embodiments of the present application.

[0087] Figure 8 An information interaction timing diagram of a second method for determining a candidate device is provided for some embodiments of the present application.

[0088] Figure 9 A flow schematic diagram of step 803 is provided for some embodiments of the present application.

[0089] Figure 10 A flow schematic diagram of step S805 is provided for some embodiments of the present application.

[0090] Figure 11 A flow schematic diagram of step S8051 is provided for some embodiments of the present application.

[0091] Figure 12 A flow schematic diagram of step S8052 is provided for some embodiments of the present application.

[0092] Figure 13 A flow schematic diagram of step S8054 is provided for some embodiments of the present application. DETAILED DESCRIPTION

[0093] The embodiments will be described in detail below with reference to examples thereof as illustrated in the accompanying drawings. In the following description, same drawing reference numerals are used to denote elements having substantially similar or analogous functions to each other. The embodiments described in the following examples are not representative of all embodiments consistent with the present application. Rather, they are merely examples of systems and methods consistent with some aspects of the present application as detailed in the claims.

[0094] It is to be noted that the brief explanation of the terms in the present application is only for the convenience of understanding the embodiments described below, and is not intended to limit the embodiments of the present application. Unless otherwise specified, these terms should be understood according to their ordinary and customary meanings.

[0095] The terms "first", "second", "third", and the like in the specification and claims of the present application and the above-described drawings are used to distinguish similar or like objects or entities, and do not necessarily mean a specific order or sequence, unless otherwise noted. It should be understood that the terms used in this way can be interchanged as appropriate.

[0096] The terms "include" and "have" and any variations thereof are intended to cover but not exclusive of inclusion, for example, a product or device that includes a list of components does not necessarily have to include all of the components clearly listed, but can include other components not clearly listed or inherent to such products or devices.

[0097] The term "module" refers to any known or later developed hardware, software, firmware, artificial intelligence, fuzzy logic, or combination of hardware and / or software code that can perform the functions associated with that element.

[0098] Figure 1 The schematic diagram of the operation scenario between the mobile terminal and the server is provided for some embodiments of the present application.

[0099] As Figure 1 It is also shown in the middle that the mobile terminal 200 also communicates data with the server 100 through various communication modes. For example, the mobile terminal 200 can be connected in communication through a local area network (LAN), a wireless local area network (WLAN), and other networks.

[0100] In the related art, with the continuous development of the control mode of smart home devices, virtual remote controllers gradually become the mainstream of the interactive interface. Through a single mobile terminal, such as a smart phone, the whole house device control is integrated, visual direct operation is provided, the intuitive interaction expectation of "what you see is what you get" is supported, the value-added services such as scene mode configuration and energy consumption monitoring are supported, and the demand of users for intelligent management is met.

[0101] However, the related art still has the following problems:

[0102] 1) The user needs to scroll through dozens of icons on the operation interface of the mobile terminal to find, and the operation path is long. In a multi-device, multi-room scenario, mis-touching easily occurs, for example, mis-selecting a device in an adjacent room, which brings great inconvenience to the user and reduces the user experience.

[0103] 2) The room or type classification is used, but this requires the user to manually annotate the device attributes in advance, which not only has high maintenance cost but also is difficult to cover cross-area devices, for example, whole-house lighting. Moreover, the classification label switching operation breaks the overall interaction, thereby greatly reducing the user experience.

[0104] 3) When a new device needs to be accessed or a low-frequency device needs to be quickly located, using sorting requires a large amount of historical data to be accumulated, and there is a significant cold start period experience fault.

[0105] To solve the above technical problems, the embodiments of the present application provide a method for determining a candidate device, applied to a server, receiving a trajectory parameter corresponding to a real-time movement trajectory of a mobile terminal sent by the mobile terminal; processing the trajectory parameter based on a preset size model to obtain a trajectory feature; processing the trajectory feature based on the preset size model to output a predicted size; calculating the predicted size to obtain a selection probability of each bound device; selecting a candidate device corresponding to the highest selection probability from a plurality of bound devices; and sending indication information of each candidate device to the mobile terminal. The technical solution of the present application can quickly recommend a few candidate devices to the user, the server sends the indication information of the few candidate devices to the mobile terminal, the mobile terminal displays an icon corresponding to the indication information of each candidate device on the operation interface, and after the user selects an icon, the remote controller corresponding to the icon is displayed on the operation interface. In response to the remote controller operation of the user, the target device is started, therefore, the embodiments of the present application do not need the user to select the indication information of the target device from a list including indication information of many bound devices, can help the user to quickly select the indication information of the target device, and start and control the target device through the remote controller, greatly improving the user experience. In addition, the embodiments of the present application do not need pre-annotation and cold start, and can recommend at least one candidate device to the user when the user uses it for the first time, thereby solving the problems of pre-annotation and cold start period in the related art.

[0106] Figure 2 A structural schematic diagram of a server provided by some embodiments of the present application is provided. The server provided by the embodiments of the present application will be described in detail below. Figure 2 The server provided by the embodiments of the present application will be described in detail. The server 100 includes:

[0107] a controller 101 and a communicator 102.

[0108] The communicator 102 is configured to communicate with the mobile terminal 200.

[0109] In some examples, the controller 101 is configured to control the communicator 102 to receive trajectory parameters corresponding to a real-time moving trajectory of the mobile terminal 200 sent by an application program running on the mobile terminal; process the trajectory parameters based on a preset size model to obtain trajectory features; process the trajectory features based on the preset size model to output predicted sizes; calculate the predicted sizes to obtain selection probabilities of each binding device; select candidate devices corresponding to the highest at least one selection probability from the plurality of binding devices; and control the communicator 102 to send indication information of each candidate device to the mobile terminal 200.

[0110] In some other examples, the controller 101 is further configured to control the communicator 102 to receive a real-time moving trajectory of the mobile terminal 200 sent by an application program running on the mobile terminal 200; calculate a matching degree of the real-time moving trajectory of the mobile terminal 200 and each binding device; select candidate devices corresponding to the highest at least one matching degree from the plurality of binding devices; and control the communicator 102 to send indication information of each candidate device to the mobile terminal 200.

[0111] Compared with the related art, the embodiments of the present application need the user to select a target device from all binding devices, recommend fewer candidate devices to the user, and the user only needs to select the target device from the candidate devices, which greatly simplifies the user operation and improves the user experience. In addition, the embodiments of the present application do not need pre-annotation, and solve the problem of needing pre-annotation in the related art.

[0112] Figure 3 The first information interaction timing diagram for determining the candidate device is provided for some embodiments of the present application. The following will be described in detail in combination with Figure 3 The information interaction for determining the candidate device provided by the embodiments of the present application will be described in detail. The information interaction includes the following steps S301 to S310.

[0113] In step S301, the user long-presses a remote controller button on the operation interface of the mobile terminal, points the mobile terminal to the target device, and moves the mobile terminal.

[0114] Figure 4 A remote controller button schematic diagram is provided for some embodiments of the present application. The following will be described in detail in combination with Figure 4 The remote controller button provided by the embodiments of the present application will be described in detail.

[0115] In some examples, an application program for the remote controller is running on the mobile terminal. It should be noted that the remote controller involved in the present application is a virtual remote controller. The application program is controlled by the controller on the mobile terminal.

[0116] In some examples, in response to the user clicking the operation of the application, an operation interface is displayed, and the remote controller button and the prompt of binding a new binding device are included in the operation interface.

[0117] Exemplarily, when the user needs to bind a new binding device, the user clicks "+" under the prompt of binding a new binding device, jumps to a binding page, and in the binding page, the user can input the model and name of the binding device to be bound and the like, and send the model and name of the binding device input by the user to the server. After the server obtains the model and name of the binding device, the server can obtain the size and unique identifier of the binding device and the like, so as to realize binding of the new binding device. The unique identifier is a string.

[0118] Exemplarily, for each binding device that has completed binding, the server stores the unique identifier, the name of the binding device and the size parameter of the binding device of the binding device that has completed binding. For example, the size parameter of the binding device includes the actual length parameter and the actual width parameter of the binding device.

[0119] Exemplarily, after the user long-presses the remote controller button in the operation interface, the user points the mobile terminal to the target device in the space.

[0120] Exemplarily, the remote controller button is displayed at the bottom of the operation interface. "Long-press" can be understood as pressing the remote controller button for a period of time. For example, the user keeps pressing the remote controller button for a few seconds.

[0121] For example, the mobile terminal is a smart phone.

[0122] For example, the space can be a room where the user is located.

[0123] For example, the server is a cloud server.

[0124] In step S302, the application sends the real-time movement track of the mobile terminal to the server.

[0125] Figure 5 A real-time movement track schematic diagram is provided for some embodiments of the present application. The real-time movement track provided by the embodiments of the present application is described in detail below. Figure 5 The real-time movement track provided by the embodiments of the present application is described in detail.

[0126] In some examples, the real-time movement track includes drawing the contour of the target device with the mobile terminal in the room.

[0127] Exemplarily, the user can stand at a position far or close to the target device and draw the contour of the target device with the mobile terminal.

[0128] For example, the user can face the target device and use the mobile terminal to delineate the contour of the target device.

[0129] For example, the user can face the target device and use the mobile terminal to delineate the contour of the target device.

[0130] For example, the contour can be "rectangular", "square", "circular" or "triangular".

[0131] For example, the target device is a living room air conditioner.

[0132] In step S303, the server calculates the matching degree of the real-time moving track of the mobile terminal and each bound device.

[0133] In some examples, for each bound device, the matching degree of the contour of the target device and the actual shape of the bound device is calculated. For example, the actual shape of the bound device is "rectangular", "square", "circular" or "triangular".

[0134] For example, if the contour of the target device and the actual shape of the bound device are both quadrilaterals, the matching degree is higher when the area ratio of the two quadrilaterals is equal to the square of the length ratio of the sides.

[0135] For example, if the contour of the target device and the actual shape of the bound device are both quadrilaterals, the matching degree is higher when the length ratio of the corresponding sides of the two quadrilaterals is consistent.

[0136] In step S304, the server selects candidate devices corresponding to the highest at least one matching degree from the plurality of bound devices.

[0137] For example, the matching degrees of each bound device are sorted to obtain a matching degree queue.

[0138] For example, the sorting is from large to small, and the highest at least one matching degree is selected from the head of the matching degree queue.

[0139] For example, the bound devices corresponding to the highest preset number of matching degrees are selected as candidate devices. For example, the preset number is 3.

[0140] In step S305, the server sends the indication information of each candidate device to the mobile terminal.

[0141] For example, the indication information can be a name; for example, "living room air conditioner".

[0142] In step S306, the application displays an icon corresponding to the indication information of each candidate device on the operation interface.

[0143] Figure 6 An icon schematic diagram is provided for some embodiments of the present application. The following will be described in combination withFigure 6 The icon provided by the embodiments of the present application is described in detail.

[0144] In some examples, the server selects a preset number of candidate devices from the plurality of binding devices based on the selection probability of each candidate device. For example, the preset number is 3.

[0145] For example, when the candidate devices are "living room air conditioner", "dining room air conditioner" and "living room fan", the icons of the three candidate devices are displayed on the operation interface.

[0146] In some examples, the target device is included in the preset number of candidate devices, but it can still be inaccurate, in which case the preset size model needs to be corrected, or in other words, the preset size model needs to be further trained.

[0147] For example, after the trajectory parameters corresponding to the real-time movement trajectory and the target device are stored, the trajectory parameters corresponding to the real-time movement trajectory and the target device are taken as sample data, and the preset size model is further trained based on the sample data to improve the accuracy of the preset size model.

[0148] For example, when the target device is "living room air conditioner", but the top three selection probabilities are 0.5, 0.48 and 0.45, and the corresponding candidate devices are "dining room air conditioner", "living room air conditioner" and "living room fan", respectively, that is, the selection probability of the target device among the three candidate devices is not the largest, that is, the preset size model is inaccurate and needs to be further corrected.

[0149] In step S307, the user performs an icon selection operation.

[0150] In step S308, the application displays a remote controller corresponding to the icon selection operation on the operation interface in response to the icon selection operation.

[0151] Figure 7 A schematic diagram of a remote controller provided by some embodiments of the present application is provided. The remote controller provided by the embodiments of the present application is described in detail below. Figure 7 The remote controller provided by the embodiments of the present application is described in detail.

[0152] For example, when the target device is "living room air conditioner", the remote controller corresponding to "living room air conditioner" is displayed on the operation interface.

[0153] For example, the remote controller includes a temperature button and a wind speed button, etc., so as to control the temperature and wind speed of the "living room air conditioner" when the "living room air conditioner" is working.

[0154] In step S309, the user performs a remote controller operation.

[0155] In step S310, the application program starts the target device corresponding to the remote controller in response to the remote controller operation.

[0156] In some examples, for a new binding device, the server has not stored the user historical behavior information related to the new binding device, in which case, the algorithm of the present application can still be used to determine the target device, but the target device can not be included in the preset number of candidate devices, but the server can store the trajectory parameters corresponding to the real-time movement trajectory and the target device as sample data, and continue to train the preset size model, and after multiple training, the server can also accurately determine the selection probability of the new binding device, thereby greatly improving the user experience.

[0157] In some other embodiments, the target device can also be started through the millimeter wave module provided on the mobile terminal.

[0158] In the embodiments of the present application, the server receives the real-time movement trajectory of the mobile terminal sent by the application program running on the mobile terminal; calculates the matching degree of the real-time movement trajectory of the mobile terminal and each binding device; selects at least one candidate device corresponding to the highest matching degree from a plurality of binding devices; and controls the communicator to send the indication information of each candidate device to the mobile terminal, thereby realizing the selection of at least one candidate device from a plurality of binding devices, and then the user only needs to select the target device from at least one candidate device, greatly simplifying the user operation and improving the user experience.

[0159] Figure 8 The second method for determining the candidate device provided by some embodiments of the present application is provided. The information interaction timing diagram of the method is as follows. Figure 8 The information interaction of the method for determining the candidate device provided by the embodiments of the present application is described in detail. The information interaction includes the following steps S801 to S812.

[0160] In step S801, the user long-presses the remote controller button on the operation interface of the mobile terminal, points the mobile terminal to the target device, and moves the mobile terminal.

[0161] In step S802, the application program sends the trajectory parameters corresponding to the real-time movement trajectory of the mobile terminal to the server.

[0162] In step S803, the server processes the trajectory parameters based on the preset size model to obtain the trajectory features.

[0163] Figure 9 A flowchart of step 803 is provided for some embodiments of the present application. The flowchart is as follows. Figure 9The step S803 provided by the embodiment of the present application is described in detail. The step S803 includes the following steps S8031 to S8034.

[0164] In some examples, the trajectory parameter includes a plurality of coordinate points; the trajectory feature includes a standard trajectory length feature, a length feature, and a shape feature. The trajectory feature is used to represent the moving trajectory of the mobile terminal;

[0165] In step S8031, the server processes each coordinate point based on a preset size model to obtain a trajectory length parameter, a space range parameter, and a shape parameter.

[0166] For example, after obtaining the plurality of coordinate points, the server pre-processes the plurality of coordinate points based on a preset size model to increase the accuracy of the coordinate points. Specifically, the pre-processing includes smoothing processing, outlier filtering, and sampling rate standardization of the plurality of coordinate points. The methods of smoothing processing, outlier filtering, and sampling rate standardization can be directly obtained from related technologies.

[0167] It should be noted that the preset size model involved in the present application is a trained model.

[0168] For example, the plurality of coordinate points are smoothed by a filter, and the outlier coordinate points are filtered by clustering.

[0169] For example, each coordinate point can be represented as (x n , y n ), and n is an integer greater than or equal to 0. The x and y in (x n , y n ) are determined by a preset coordinate system in which the room is located, and the embodiments of the present application do not limit the preset coordinate system.

[0170] For example, the trajectory length parameter is the perimeter of the contour obtained based on the pre-processed plurality of coordinate points, which can also be expressed as an initial trajectory length feature. The length parameter is obtained based on the x max in the pre-processed plurality of coordinate points and the x min in the pre-processed plurality of coordinate points. The width parameter is obtained based on the y max in the pre-processed plurality of coordinate points and the y min in the pre-processed plurality of coordinate points. The shape parameter includes the convex hull area and the convex hull perimeter. y max is the maximum vertical coordinate in the pre-processed plurality of coordinate points; y min is the minimum vertical coordinate in the pre-processed plurality of coordinate points; x min is the minimum horizontal coordinate in the pre-processed plurality of coordinate points; and x max is the maximum horizontal coordinate in the pre-processed plurality of coordinate points.

[0171] In step S8032, the server performs multiplication calculation on the trajectory length parameter and the size factor based on the preset size model to obtain a standard trajectory length feature.

[0172] In some examples, for the size factor, when the user uses the application for the first time, the trajectory parameters corresponding to the first real-time moving trajectory are processed by the preset size model to obtain a first predicted size, the predicted length parameter and the actual length parameter are calculated to obtain a first ratio. In order to improve the accuracy of the size factor, after obtaining the first ratio, the user is prompted by the application to perform a second real-time moving trajectory, the trajectory parameters corresponding to the second real-time moving trajectory are processed by the preset size model to obtain a second predicted size, the predicted length parameter and the actual length parameter are calculated to obtain a second ratio. The first ratio and the second ratio are calculated to obtain the size factor.

[0173] Embodiments of the present application do not have a cold start period, which solves the problem of the cold start period in the related art.

[0174] In step S8033, the server performs division calculation on the space range parameter based on the preset size model to obtain a length-width feature.

[0175] Exemplarily, the predicted size includes a predicted length parameter and a predicted width parameter.

[0176] For example, the size factor=(first ratio+second ratio) / 2.

[0177] For example, the standard trajectory length feature=initial trajectory length feature / α. α is the size factor.

[0178] For example, the space range parameter includes a length parameter and a width parameter.

[0179] Length parameter=x max -x min Width parameter=y max -y min .

[0180] Length-width feature=length parameter / width parameter.

[0181] In step S8034, the server performs division calculation on the shape parameter based on the preset size model to obtain a shape feature.

[0182] For example, the shape feature=4π×convex hull area / convex hull perimeter 2 .

[0183] It should be noted that the calculation method of the convex hull area and the convex hull perimeter can be directly obtained from the related art, and embodiments of the present application will not be described again.

[0184] For example, the trajectory feature can be represented as:

[0185] l (standard trajectory length feature, length-width feature, shape feature).

[0186] It should be noted that the size factor is generally not changed after being obtained, although this may cause some errors, but in the embodiment of the present application, when determining the selection probability of each binding device, not only the trajectory parameters of the real-time moving trajectory of the mobile terminal are used, but also the real-time horizontal angle and real-time pitch angle of the mobile terminal and the position parameters of the mobile terminal are used, so as to avoid determining the selection probability of each binding device only according to one dimension (size factor), that is, to reduce the error influence caused by the fixed size factor.

[0187] In the embodiment of the present application, first, the average of the first ratio and the second ratio is used as the size factor, rather than directly using the first ratio as the size factor, in order to improve the accuracy of the size factor. Secondly, through the size factor, it is ensured that the contours drawn by different users correspond to the same trajectory length feature, so as to eliminate the influence of distance and individual real-time moving trajectory difference.

[0188] In step S804, the trajectory feature is processed based on the preset size model, and a predicted size is output.

[0189] In some examples, the preset size model includes a regression model, and the trajectory feature is processed by the regression model to output the predicted size.

[0190] For example, the regression model is a random forest model. Specifically, the random forest model includes a plurality of decision trees, each decision tree processes the trajectory feature to obtain an intermediate predicted size, and the average of each intermediate predicted size is calculated. The average of each intermediate predicted size is the predicted size.

[0191] In step S805, the predicted size is calculated to obtain the selection probability of each binding device.

[0192] In some examples, before step S805, the real-time pitch angle, the real-time horizontal angle, the position parameters, the position weight, the angle weight, the size weight, the historical behavior parameters and the historical behavior weight of the mobile terminal are obtained.

[0193] The historical behavior parameters are obtained based on the historical behavior of the user.

[0194] The position parameters are obtained based on the position of the mobile terminal and the position of each binding device.

[0195] In some examples, the sum of the position weight, the angle weight, the historical behavior weight, and the size weight is 1. The position weight is determined according to a range of the position weight, the angle weight is determined according to a range of the angle weight, the historical behavior weight is determined according to a range of the historical behavior weight, and the size weight is determined according to a range of the size weight.

[0196] Exemplarily, the higher the positioning accuracy of the position where the user is located, the greater the position weight.

[0197] The more the historical behavior parameters, the greater the historical behavior weight.

[0198] The greater the number of historical horizontal angles and the number of historical pitch angles, the greater the angle weight.

[0199] For example, the range of the size weight is 0.4-0.6. The range of the position weight is 0.3-1. The range of the angle weight is 0.1-0.3. The range of the historical behavior weight is 0-0.2.

[0200] When the number of historical horizontal angles and the number of historical pitch angles are both greater than or equal to 10, the angle weight is 0.3. When the number of historical behavior parameters is less than 50, the historical behavior weight is 0. When the positioning accuracy of the position where the user is located is 100%, the position weight is 0.4; or when the positioning accuracy of the position where the user is located is 50%, the position weight is 0.3.

[0201] Exemplarily, the position weight, the angle weight, the historical behavior weight, and the size weight are distributed using a distributor.

[0202] For example, the distributor is a weight balancer.

[0203] Embodiments of the present application can determine the distance between the position where the user is located and each binding device by calculating the position matching degree. The greater the position matching degree, the higher the possibility that the binding device corresponding to the position matching degree is the target device, thereby facilitating the selection of at least one candidate device from multiple binding devices.

[0204] Embodiments of the present application obtain various parameters for calculating the selection probability, thereby facilitating the calculation of the selection probability.

[0205] Figure 10 A flowchart of step S805 is provided for some embodiments of the present application. The following will be described in combination with Figure 10 The step S805 provided by embodiments of the present application will be described in detail. The step S805 is as follows: steps S8051 to S8055.

[0206] In some examples, in step S8051, the server performs a multiplication calculation on the predicted size and the size weight to obtain a size weight term.

[0207] Figure 11 A flowchart of step S8051 provided by some embodiments of the present application is shown. The following will be described in detail in combination with Figure 11 The step S8051 provided by embodiments of the present application will be described in detail. The step S8051 includes the following steps S80511 to S80514.

[0208] In step S80511, the server performs subtraction calculation on the predicted length parameter and the actual length parameter to obtain a length difference.

[0209] In some examples, the length difference is obtained based on the absolute value of the difference between the preset length parameter and the actual length parameter.

[0210] In step S80512, the server performs subtraction calculation on the predicted width parameter and the actual width parameter to obtain a width difference.

[0211] In some examples, the width difference is obtained based on the absolute value of the difference between the preset width parameter and the actual width parameter.

[0212] In step S80513, the server performs exponential calculation on the length difference and the width difference to obtain a size matching degree.

[0213] In some examples, the size matching degree is obtained based on exponential function calculation on the length difference and the width difference.

[0214] Exemplarily, the size matching degree = e -(ΔL+ΔW) / β ;

[0215] In the above formula, ΔL represents the length difference, ΔW represents the width difference, and β represents a preset size coefficient.

[0216] For example, β is 10.

[0217] In step S80514, the server performs multiplication calculation on the size matching degree and the size weight to obtain a size weight term.

[0218] In embodiments of the present application, the greater the sum of the length difference and the width difference, the smaller the size matching degree, that is, the smaller the matching degree of the binding device corresponding to the size matching degree and the target device, and the smaller the sum of the length difference and the width difference, the greater the size matching degree, that is, the greater the matching degree of the binding device corresponding to the size matching degree and the target device, thereby facilitating selection of at least one candidate device from multiple binding devices.

[0219] In step S8052, the server performs multiplication calculation on the real-time pitch angle, the real-time horizontal angle of the mobile terminal and the angle weight to obtain an angle weight term.

[0220] Figure 12 A flowchart of step S8052 is provided for some embodiments of the present application. The following will be described in detail in combination with Figure 12 The step S8052 provided by the embodiments of the present application will be described in detail. The step S8052 includes the following steps S80521 to S80523.

[0221] In some examples, the angle weight includes a horizontal angle weight and a pitch angle weight; and the angle weight term includes a horizontal angle weight term and a pitch angle weight term.

[0222] In some examples, in step 80521, the server performs matching degree calculation on the real-time horizontal angle and the historical horizontal angle to obtain a horizontal angle matching degree, and performs multiplication calculation on the horizontal angle matching degree and the horizontal angle weight to obtain a horizontal angle weight term.

[0223] In some examples, the real-time horizontal angle and the real-time pitch angle of the mobile terminal further enrich the position information of each binding device in the user's room, thereby increasing the accuracy of the selection probability of the binding device. For example, when the last historical horizontal angle is 90° (pointing to the east wall) and the last historical pitch angle is 50°, and the target device is the living room air conditioner, then when the user points to the east wall next time, the selection probability of the living room air conditioner will be greater than that of the binding device corresponding to the south wall.

[0224] In some examples, each movement of the mobile terminal corresponds to a real-time horizontal angle and a real-time pitch angle. Illustratively, the real-time pitch angle is obtained through the gyroscope of the mobile terminal, and the real-time horizontal angle is obtained through the compass of the mobile terminal. Compared with the related art which uses sensors to obtain the real-time pitch angle and the real-time horizontal angle, the embodiments of the present application are simple in calculation, save computing power, and reduce delay.

[0225] In some examples, a plurality of historical horizontal angles are obtained, the average of the plurality of historical horizontal angles is calculated; the average of the plurality of historical horizontal angles is subtracted from the real-time horizontal angle to obtain a first horizontal angle difference value; the first horizontal angle difference value is subtracted from a preset angle threshold value to obtain a second horizontal angle difference value; and the minimum value of the first horizontal angle difference value and the second horizontal angle difference value is taken as a target horizontal angle difference value. For example, the preset angle threshold value is 360°.

[0226] Illustratively, the target horizontal angle difference value = min (the first horizontal angle difference value, the second horizontal angle difference value).

[0227] The plurality of historical The historical horizontal angle. χ is the number of historical horizontal angles.

[0228] The first horizontal angle difference value = |real-time horizontal angle - average of the plurality of historical horizontal angles|.

[0229] The second horizontal angle difference value = |360°-the first horizontal angle difference value|.

[0230] Exemplarily, the actual size of χ can be determined according to actual needs. For example, χ is 10.

[0231] Exemplarily, when the target horizontal angle difference value is not greater than 90°, the cosine of the target horizontal angle difference is calculated to obtain the horizontal angle matching degree, and when the target horizontal angle difference value is less than 90°, the horizontal angle matching degree is 0.

[0232] Exemplarily,

[0233]

[0234] When the target horizontal angle difference value is greater than 90°, the horizontal angle matching degree is 0, avoiding the influence of negative values.

[0235] In step 80522, the server calculates the matching degree of the real-time pitch angle and the historical pitch angle to obtain the pitch angle matching degree, and performs multiplication calculation on the pitch angle matching degree and the pitch angle weight to obtain the pitch angle weight term.

[0236] In some examples, a plurality of historical pitch angles is obtained, and the average value of the plurality of historical pitch angles is calculated. The difference between the average value of the plurality of historical pitch angles and the real-time pitch angle is taken as the target pitch angle difference value.

[0237] Exemplarily, the plurality of historical The historical pitch angles. δ is the number of historical horizontal angles.

[0238] The target pitch angle difference value = |the real-time pitch angle-the average value of the plurality of historical pitch angles|.

[0239] Exemplarily, the actual size of δ can be determined according to actual needs. For example, δ is 10.

[0240] In some examples, the target pitch angle difference value is subjected to linear decay calculation to obtain the pitch angle matching degree.

[0241] Exemplarily, when the target pitch angle difference value is not greater than 90°, the target pitch angle difference value is subjected to linear decay calculation to obtain the pitch angle matching degree, and when the target pitch angle difference value is less than 90°, the pitch angle matching degree is 0.

[0242] Exemplarily,

[0243] When the target pitch angle difference value is greater than 90°, the pitch angle matching degree is 0, avoiding the influence of negative values.

[0244] In step 80523, the server performs an addition operation on the pitch angle weight and the horizontal angle weight to obtain an angle weight.

[0245] In some examples, the horizontal angle weight is obtained by performing a multiplication operation on the horizontal angle matching degree and the horizontal coefficient, and the pitch angle weight is obtained by performing a multiplication operation on the pitch angle matching degree and the pitch coefficient.

[0246] For example, the horizontal angle weight = horizontal angle matching degree x horizontal coefficient; the pitch angle weight = pitch angle matching degree x pitch coefficient; and the angle weight = horizontal angle weight + pitch angle weight.

[0247] For example, the sum of the horizontal coefficient and the pitch coefficient is 1. The specific values of the horizontal coefficient and the pitch coefficient are determined according to actual needs.

[0248] For example, the horizontal coefficient is 0.3 and the pitch coefficient is 0.7.

[0249] In step S8053, the server performs a multiplication operation on the position parameter and the position weight to obtain a position weight.

[0250] In some examples, for each binding device, the position parameter is determined according to the position bound by the binding device and the position currently occupied by the user.

[0251] For example, the application program can bind multiple binding devices, and the position bound by each binding device and the position currently occupied by the user can be different. In order to distinguish these cases, embodiments of the present application determine different position weights for different position parameters, thereby obtaining a position weight.

[0252] For example, when the position parameter is that the position bound by the binding device and the position currently occupied by the user belong to the same city, a first preset position value is taken as the position matching degree.

[0253] For example, when the position parameter is that the position bound by the binding device and the position currently occupied by the user belong to the same province but not to the same city, a second preset position value is taken as the position matching degree.

[0254] For example, when the position parameter is that the position bound by the binding device and the position currently occupied by the user do not belong to the same province, a third preset position value is taken as the position matching degree.

[0255] For example, the first preset position value is 1, the second preset position value is 0.7, and the third preset position value is 0.3.

[0256] In step S8054, the server performs multiplication calculation on the historical behavior parameter and the historical behavior weight to obtain a historical behavior weight item.

[0257] In step S8055, the server performs addition calculation on the size weight item, the angle weight item, the position weight item and the historical behavior weight item to obtain the selection probability of each binding device.

[0258] In some examples, the server calculates the sum of the server size weight item, the angle weight item, the position weight item and the historical behavior weight item to obtain the selection probability of each binding device.

[0259] Embodiments of the present application determine the actual intention of the user from multiple dimensions through the size weight item, the angle weight item, the position weight item and the historical behavior weight item, thereby increasing the selection probability of each binding device.

[0260] Figure 13 A flowchart of step S8054 is provided for some embodiments of the present application. The following will be described in combination with Figure 13 The step S8054 provided by the embodiments of the present application will be described in detail. The step S8054 includes the following steps S80541 to S80545.

[0261] In some examples, for each historical movement trajectory, the server stores the historical horizontal angle, the historical trajectory parameter, the historical pitch angle, the target device and the historical movement trajectory time and other user historical behavior information corresponding to the historical movement trajectory.

[0262] In some examples, the server inputs the historical trajectory parameter into the preset size model for further training of the preset size model, thereby improving the accuracy of the preset size model.

[0263] In some examples, the server also stores the historical movement trajectory frequency.

[0264] In some examples, the historical behavior parameter includes historical movement trajectory time parameter and historical movement trajectory frequency parameter of each group; the historical movement trajectory time parameter includes historical movement trajectory time coefficient and historical movement trajectory time median of each group; the historical movement trajectory frequency parameter includes historical movement trajectory frequency and historical movement trajectory frequency coefficient of each group; the historical behavior weight includes historical movement trajectory time weight and historical movement trajectory frequency weight; the historical movement trajectory time coefficient and the historical movement trajectory time median correspond one-to-one; the historical movement trajectory frequency and the historical movement trajectory frequency coefficient correspond one-to-one.

[0265] In step S80541, the server performs multiplication calculation on the historical movement trajectory time coefficient and the historical movement trajectory time median of each group to obtain the historical movement trajectory time parameter.

[0266] In some examples, each historical moving trajectory time median corresponds to a first historical time period.

[0267] Illustratively, the server clusters the stored historical moving trajectory times according to preset time periods, to generate a plurality of first historical time periods. Each first historical time period corresponds to a category.

[0268] For example, the categories are morning peak, noon, or evening. The length of each first historical time period is 30 minutes.

[0269] In some examples, the multiplication calculation is performed on each set of historical moving trajectory time coefficients and historical moving trajectory time medians, to obtain historical moving trajectory time weight terms corresponding to each set of historical moving trajectory time coefficients and historical moving trajectory time medians; and the weighted sum is performed on the plurality of historical moving trajectory time weight terms, to obtain the historical moving trajectory time coefficients and the historical moving trajectory time medians.

[0270] Illustratively, the historical moving trajectory time coefficients are obtained based on the number of first historical time periods and the total number of stored historical moving trajectory times.

[0271] For example, the historical time coefficient = the number of first historical time periods / the total number of stored historical moving trajectory times.

[0272] Illustratively, the product of each set of historical moving trajectory time coefficients and historical moving trajectory time medians is calculated, and the weighted sum is performed on the product of each set of historical moving trajectory time coefficients and historical moving trajectory time medians, to obtain the historical moving trajectory time parameter.

[0273] Illustratively,

[0274]

[0275] In the above formula, ε k represents the kth historical moving trajectory time coefficient; is the length of the kth first historical time period.

[0276] wherein, may be dynamically adjusted. For example, when ε k is greater than 0.3, then is 15 min; when ε k is not greater than 0.3, then is 30 min.

[0277] In step S80542, the server performs multiplication calculation on each set of historical moving trajectory frequency and historical moving trajectory frequency coefficient, to obtain the historical moving trajectory frequency parameter.

[0278] In some examples, each historical movement trajectory frequency coefficient corresponds to a second historical time period. For example, each second historical time period is one day. The first second historical time period is the day 0 days ago, i.e., the day corresponding to the real-time movement trajectory. The second historical time period is the day 1 day ago.

[0279] In some examples, each group of historical movement trajectory frequencies and historical movement trajectory frequency coefficients are multiplied to obtain historical movement trajectory frequency weights corresponding to each group of historical movement trajectory frequencies and historical movement trajectory frequency coefficients; and multiple historical movement trajectory frequency weights are weighted and summed to obtain historical movement trajectory frequency parameters.

[0280] For example, if d is different, the corresponding historical movement trajectory frequency coefficient is also different.

[0281] For example, when d = 0, the historical movement trajectory frequency coefficient is 1. When 0 < d ≤ 30, the historical movement trajectory frequency coefficient is approximately 0.05. Generally, when d is greater than 30, all user historical behavior information corresponding to that day is deleted. This is because when d is greater than 30, the corresponding user historical behavior information has a lower reference value, which will reduce the accuracy of the selection probability.

[0282] For example,

[0283] In the above formula, N d Indicates the frequency of starting the target device within the day before d day.

[0284] In some examples, in some scenarios, the user may not activate the target device within a certain day. In this case, directly using the above formula to calculate the historical movement trajectory frequency parameter will reduce the accuracy of the historical movement trajectory frequency parameter. To solve this problem, the embodiment of the present application also counts the number of consecutive days M in which the target device is activated.

[0285] For example, when M < the threshold of consecutive days,

[0286] When M≥threshold of consecutive days,

[0287] For example, the consecutive days threshold is 3.

[0288] For example, if the user starts the target device for five consecutive days, then M=5; if the user does not start the target device within the next day, that is, there is an interruption in starting the target device for multiple consecutive days, then M=0, and then the value of M is recalculated.

[0289] Embodiments of the present application calibrate the historical movement trajectory frequency parameter through M, improve the accuracy of the historical behavior weight of the user, and further improve the selection probability of each binding device.

[0290] In step S80543, the server performs multiplication calculation on the historical movement trajectory time parameter and the historical movement trajectory time weight to obtain a historical movement trajectory time weight term.

[0291] For example, the historical movement trajectory time weight term = historical movement trajectory time parameter * historical movement trajectory time weight.

[0292] In step S80544, the server performs multiplication calculation on the historical movement trajectory frequency weight and the historical movement trajectory frequency parameter to obtain a historical movement trajectory frequency weight term.

[0293] For example, the historical movement trajectory frequency weight term = historical movement trajectory frequency weight * historical movement trajectory frequency parameter.

[0294] In step S80545, the server performs addition calculation on the historical movement trajectory frequency weight term and the historical movement trajectory time weight term to obtain a historical behavior weight term.

[0295] For example, the historical behavior weight term = historical movement trajectory frequency weight term + historical movement trajectory time weight term.

[0296] In step S8055, the server performs addition calculation on the size weight term, the angle weight term, the position weight term, and the historical behavior weight term to obtain the selection probability of the binding device.

[0297] In some examples, for each binding device, the sum of the size weight term, the angle weight term, the position weight term, and the historical behavior weight term corresponding to the binding device are taken as the selection probability of the binding device.

[0298] For example, the selection probabilities of each binding device are sorted to obtain a selection probability queue.

[0299] For example, the selection probabilities are sorted from large to small, and the highest at least one selection probability is selected from the head of the selection probability queue.

[0300] Embodiments of the present application select the highest at least one selection probability from the selection probabilities of each binding device, and then take the binding device corresponding to each selection probability in the highest at least one selection probability as a candidate device.

[0301] In step S806, the server selects the candidate device corresponding to the highest at least one selection probability from the plurality of binding devices.

[0302] In step S807, the server sends the indication information of each candidate device to the mobile terminal.

[0303] In step S808, the application displays the icon corresponding to the indication information of each candidate device on the operation interface.

[0304] In step S809, the user performs the icon selection operation.

[0305] In step S810, the application displays the remote controller corresponding to the icon selection operation on the operation interface in response to the icon selection operation.

[0306] In step S811, the user performs the remote controller operation.

[0307] In step S812, the target device corresponding to the remote controller is started in response to the remote controller operation.

[0308] According to the embodiments of the present application, the server receives the trajectory parameters corresponding to the mobile terminal to obtain the trajectory features, processes the trajectory features based on the preset size model to output the predicted size, calculates the predicted size to obtain the selection probability of each binding device, selects the candidate device corresponding to the highest selection probability from the plurality of binding devices, and sends the indication information of each candidate device to the mobile terminal. Compared with the related art which requires the user to select a target device from all the binding devices, the present application recommends fewer candidate devices to the user, greatly simplifies the user operation, and improves the user experience.

[0309] The above description has been made in conjunction with specific embodiments for the convenience of explanation. However, the above description in some embodiments is not intended to be exhaustive or to limit the embodiments to the specific forms disclosed. Various modifications and variations can be derived from the above teachings. The selection and description of the above embodiments are to better explain the content of the present disclosure, so that those skilled in the art can better use the embodiments.

Claims

1. A method for determining a candidate device, characterized in that: include: receiving trajectory parameters corresponding to the real-time movement trajectory of the mobile terminal sent by the mobile terminal; Processing the trajectory parameters based on a preset size model to obtain trajectory features; Processing the trajectory features based on the preset size model and outputting a predicted size; Calculating the predicted size to obtain a selection probability of each bound device; Selecting a candidate device corresponding to at least one highest selection probability from the plurality of bound devices; Sending indication information of each candidate device to the mobile terminal.

2. The method according to claim 1, characterized in that The trajectory parameters include a plurality of coordinate points; The trajectory features include standard trajectory length features, length features and shape features; The trajectory feature is used to characterize the movement trajectory of the mobile terminal; The step of processing the trajectory parameters based on a preset size model to obtain trajectory features includes: Processing each of the coordinate points based on the preset size model to obtain a trajectory length parameter, a spatial range parameter, and a shape parameter; Performing multiplication calculation on the trajectory length parameter and the size factor based on the preset size model to obtain a standard trajectory length feature; Performing division calculation on the spatial range parameter based on the preset size model to obtain length and width features; The shape parameters are divided based on the preset size model to obtain shape features.

3. The method according to claim 1, characterized in that Before calculating the predicted size and obtaining the selection probability of each bound device, the method further includes: Obtaining a real-time pitch angle, a real-time horizontal angle, a position parameter, a position weight, an angle weight, a size weight, a historical behavior parameter, and a historical behavior weight of the mobile terminal; The historical behavior parameter is obtained based on the user's historical behavior; The location parameter is obtained based on the location of the mobile terminal and the location of each of the bound devices.

4. The method according to claim 3, characterized in that Calculating the predicted size to obtain the selection probability of each bound device includes: For each of the bound devices, multiplying the predicted size and the size weight to obtain a size weight; multiplying the real-time pitch angle, the real-time horizontal angle, and the angle weight of the mobile terminal to obtain an angle weight; Performing multiplication calculation on the position parameter and the position weight to obtain a position weight; Multiplying the historical behavior parameter and the historical behavior weight to obtain a historical behavior weight; The size weight item, the angle weight item, the position weight item, and the historical behavior weight item are added together to obtain the selection probability of the bound device.

5. The method according to claim 4, characterized in that The predicted size includes a predicted length parameter and a predicted width parameter; The predicted size and the size weight are multiplied to obtain a size weight, including: Subtracting the predicted length parameter from the actual length parameter to obtain a length difference; Subtracting the predicted width parameter from the actual width parameter to obtain a width difference; Performing exponential calculation on the length difference and the width difference to obtain a size matching degree; The size matching degree and the size weight are multiplied to obtain the size weight item.

6. The method according to claim 4, characterized in that The angle weight includes a horizontal angle weight and a pitch angle weight; the angle weight includes a horizontal angle weight and a pitch angle weight; The real-time pitch angle, the real-time horizontal angle, and the angle weight of the mobile terminal are multiplied to obtain an angle weight, including: performing a matching calculation on the real-time horizontal angle and the historical horizontal angle to obtain a horizontal angle matching degree, and performing a multiplication calculation on the horizontal angle matching degree and the horizontal angle weight to obtain a horizontal angle weight; performing a matching calculation on the real-time pitch angle and the historical pitch angle to obtain a pitch angle matching degree, and performing a multiplication calculation on the pitch angle matching degree and the pitch angle weight to obtain a pitch angle weight; The pitch angle weight and the horizontal angle weight are added to obtain the angle weight.

7. The method according to claim 4, characterized in that The historical behavior parameters include a time parameter of each group of historical movement trajectories and a frequency parameter of the historical movement trajectories; the historical movement trajectories time parameters include a time coefficient of each group of historical movement trajectories and a median of the historical movement trajectories time; the historical movement trajectories frequency parameters include a frequency of each group of historical movement trajectories and a frequency coefficient of the historical movement trajectories; the historical behavior weight includes a time weight of the historical movement trajectories and a frequency weight of the historical movement trajectories; the time coefficient of the historical movement trajectories and the median of the historical movement trajectories time are in one-to-one correspondence; the frequency of the historical movement trajectories and the frequency coefficient of the historical movement trajectories are in one-to-one correspondence; The historical behavior parameter and the historical behavior weight are multiplied to obtain a historical behavior weight, including: Multiplying each group of the historical movement trajectory time coefficients and the historical movement trajectory time median to obtain the historical movement trajectory time parameter; Performing multiplication calculation on each group of the historical movement trajectory frequencies and the historical movement trajectory frequency coefficient to obtain the historical movement trajectory frequency parameter; Multiplying the historical movement trajectory time parameter and the historical movement trajectory time weight to obtain a historical movement trajectory time weight; Multiplying the historical movement trajectory frequency weight and the historical movement trajectory frequency parameter to obtain a historical movement trajectory frequency weight; The historical movement trajectory frequency weight item and the historical movement trajectory time weight item are added together to obtain the historical behavior weight item.

8. A method for determining a candidate device, characterized in that: include: receiving a real-time moving trajectory of the mobile terminal sent by the mobile terminal; Calculating the matching degree between the real-time movement trajectory of the mobile terminal and each bound device; Selecting at least one candidate device corresponding to the highest matching degree from the plurality of bound devices; Sending indication information of each candidate device to the mobile terminal.

9. A server, characterized in that: include: a communicator configured to communicate with a mobile terminal; The controller is configured as: controlling the communicator to receive trajectory parameters corresponding to the real-time movement trajectory of the mobile terminal sent by the mobile terminal; Processing the trajectory parameters based on a preset size model to obtain trajectory features; Processing the trajectory features based on the preset size model and outputting a predicted size; Calculating the predicted size to obtain a selection probability of each bound device; Selecting a candidate device corresponding to at least one highest selection probability from the plurality of bound devices; The communicator is controlled to send indication information of each candidate device to the mobile terminal.

10. A server, characterized in that: include: a communicator configured to communicate with a mobile terminal; The controller is configured as: Controlling the communicator to receive the real-time moving trajectory of the mobile terminal sent by the mobile terminal; Calculating the matching degree between the real-time movement trajectory of the mobile terminal and each bound device; Selecting at least one candidate device corresponding to the highest matching degree from the plurality of bound devices; The communicator is controlled to send indication information of each candidate device to the mobile terminal.

Images