A method and system for adjusting fan operating parameters for an automotive seat

Abstract

The application relates to the technical field of intelligent cockpits, and discloses a fan working parameter control method and system for a car seat, wherein the method comprises the following steps: acquiring object features of a driver or passenger sitting on the car seat and environmental features in the cockpit; constructing a feature matrix for controlling fan working parameters based on the object features and the environmental features, and determining main feature factors and secondary feature factors in the feature matrix; adding different disturbance features to the secondary feature factors while keeping the main feature factors unchanged, and constructing multiple disturbance matrices; screening a target fan adjustment model from multiple fan adjustment models obtained through pre-training; acquiring fan adjustment parameters output by the target fan adjustment model for the feature matrix, and performing fan adjustment based on the acquired fan adjustment parameters. The method has the beneficial effect that an individualized fan adjustment process can be finely realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent cockpit, and particularly relates to a fan working parameter adjustment method and system for a car seat. BACKGROUND

[0002] The existing fan adjustment mode of the car seat is usually manually adjusted by the driver or passenger. For example, the cockpit provides different fan strengths, which are manually switched by the driver or passenger.

[0003] With the popularization of the intelligent cockpit, the demand for intelligent adjustment of the fan is increasingly urgent. However, the existing fan adjustment mode is only linked with the intelligent voice assistant, and the fan strength is adjusted by the voice of the driver or passenger. Although this adjustment mode is simple, it cannot be personalized for different driving scenarios. SUMMARY

[0004] The present application provides a fan working parameter adjustment method and system for a car seat, which can realize personalized fan adjustment mode to improve the riding experience of the driver or passenger.

[0005] In order to achieve the above purpose, the main technical scheme adopted by the present application includes:

[0006] In a first aspect, the present application provides a fan working parameter control method for a car seat, which comprises:

[0007] Obtaining the object features of the driver or passenger sitting on the car seat and the environmental features in the cockpit, wherein the object features are used to at least represent the gender, preference and contact area of the driver or passenger with the car seat, and the environmental features are used to represent the current actual temperature in the cockpit and the target adjustment temperature of the air conditioner;

[0008] Constructing a feature matrix for controlling the fan working parameter based on the object features and the environmental features, and determining the main feature factor and the secondary feature factor in the feature matrix;

[0009] Adding different disturbance features to the secondary feature factor while keeping the main feature factor unchanged, and constructing a plurality of disturbance matrices according to the main feature factor and the secondary feature factor after adding the disturbance features;

[0010] screening a target fan adjustment model from the plurality of pre-trained fan adjustment models, wherein the target fan adjustment model processes the feature matrix and each of the disturbance matrices and obtains respective fan adjustment parameters, and if a cluster analysis is performed on each of the fan adjustment parameters, the fan adjustment parameter corresponding to the feature matrix and the fan adjustment parameter corresponding to the disturbance matrix are located in different cluster clusters;

[0011] obtaining the fan adjustment parameter output by the target fan adjustment model for the feature matrix, and performing fan adjustment based on the obtained fan adjustment parameter.

[0012] In one embodiment, the preference is used to represent the preference degree of the driver or passenger for cold and warm, and the preference is represented by a numerical value in a specified interval, wherein the smaller the numerical value, the more the preference for cold, and the higher the numerical value, the more the preference for warm.

[0013] In one embodiment, constructing a feature matrix for controlling fan operating parameters based on the object features and the environment features comprises:

[0014] According to the preset sampling period, the object features and the environment features are collected, and the object features and the environment features with the same time stamp are spliced into a time stamp sequence;

[0015] According to the time sequence relationship, each of the time stamp sequences is arranged in sequence to form a feature matrix for controlling fan operating parameters.

[0016] In one embodiment, determining the main feature factor and the secondary feature factor in the feature matrix comprises:

[0017] Identifying feature data of each feature factor at different time stamps in the feature matrix;

[0018] For any feature data, the feature data is taken as index data, and query data and attribute data are determined from other feature data, and based on the association between the index data and the query data and attribute data, an index value of the index data is calculated;

[0019] According to the index value of each feature data, the main feature factor and the secondary feature factor are determined.

[0020] In one embodiment, calculating the index value of the index data comprises:

[0021] According to the same time span, the index data, the query data and the attribute data are respectively divided to obtain respective local data;

[0022] For any first local data of the index data, a similarity between the first local data and each local data of the query data is calculated to obtain a plurality of similarities;

[0023] The plurality of similarities are taken as weight coefficients to weight each local data of the attribute data, and each weighted local data is spliced to obtain an index feature;

[0024] The index feature is input into an index value calculation model to obtain an index value corresponding to the index feature.

[0025] In one embodiment, each fan adjustment model is obtained by training in the following manner:

[0026] For any main feature factor, a training sample of the main feature factor is obtained, and a first label value of a fan adjustment parameter is set for the training sample;

[0027] On the basis of the training sample, a disturbance sample corresponding to the training sample is generated by adding a disturbance feature to a secondary feature factor, and a second label value of a fan adjustment parameter is set for each disturbance sample, wherein, when the first label value and the second label value are subjected to cluster analysis, the first label value and the second label value are located in different cluster clusters;

[0028] The training sample and the disturbance sample are used to train a fan adjustment model, so that the prediction results output by the trained fan adjustment model for the training sample and the disturbance sample correspond to the first label value and the second label value, respectively;

[0029] The trained fan adjustment model is taken as a fan adjustment model of the main feature factor.

[0030] In a second aspect, the application further provides a fan working parameter control system for a car seat, the system comprising:

[0031] A feature acquisition unit is configured to acquire object features of a driver or passenger seated on the car seat and environment features in a cabin, wherein the object features are used to at least represent a gender, a preference, and a contact area with the car seat of the driver or passenger, and the environment features are used to represent a current actual temperature in the cabin and a target adjustment temperature of an air conditioner;

[0032] A factor determination unit is configured to construct a feature matrix for controlling fan working parameters based on the object features and the environment features, and determine main feature factors and secondary feature factors in the feature matrix;

[0033] a disturbance unit, configured to add different disturbance features to the secondary characteristic factor while keeping the primary characteristic factor unchanged, and construct a plurality of disturbance matrices according to the primary characteristic factor and the secondary characteristic factor after adding the disturbance features;

[0034] a model screening unit, configured to screen a target fan adjustment model from a plurality of fan adjustment models pre-trained, wherein the target fan adjustment model processes the feature matrix and each of the disturbance matrices, and obtains respective fan adjustment parameters, and if clustering analysis is performed on each of the fan adjustment parameters, the fan adjustment parameter corresponding to the feature matrix and the fan adjustment parameter corresponding to the disturbance matrix are located in different clustering clusters;

[0035] a fan adjustment unit, configured to obtain a fan adjustment parameter output by the target fan adjustment model for the feature matrix, and perform fan adjustment based on the obtained fan adjustment parameter.

[0036] In an embodiment, the factor determination unit is specifically configured to identify feature data of each characteristic factor at different timestamps in the feature matrix; for any feature data, the feature data is taken as index data, and query data and attribute data are determined from other feature data, an index value of the index data is calculated based on an association between the index data and the query data and the attribute data, and the primary characteristic factor and the secondary characteristic factor are determined according to the index values of the feature data.

[0037] In an embodiment, the factor determination unit is specifically configured to divide the index data, the query data and the attribute data respectively according to the same time span to obtain respective local data; for any first local data of the index data, a plurality of similarities are obtained by calculating similarities between the first local data and each local data of the query data; each local data of the attribute data is weighted by taking the plurality of similarities as weight coefficients, and each weighted local data is spliced to obtain an index feature; and the index feature is input into an index value calculation model to obtain an index value corresponding to the index feature.

[0038] In an embodiment, the model screening unit trains each of the fan adjustment models in the following manner:

[0039] for any primary characteristic factor, a training sample of the primary characteristic factor is obtained, and a first label value of a fan adjustment parameter is set for the training sample;

[0040] On the basis of the training sample, a disturbance sample corresponding to the training sample is generated by adding a disturbance feature to a secondary characteristic factor, and a second label value of a fan adjustment parameter is set for each disturbance sample, wherein when the first label value and the second label value are subjected to clustering analysis, the first label value and the second label value are located in different clustering clusters;

[0041] The fan adjustment model is trained by using the training sample and the disturbance sample, so that the prediction results output by the trained fan adjustment model for the training sample and the disturbance sample correspond to the first label value and the second label value respectively;

[0042] The trained fan adjustment model is used as a fan adjustment model of the primary characteristic factor.

[0043] The technical solution provided by the application can reflect the current driving scene through multi-dimensional data by collecting the object characteristics of the driver and the environmental characteristics in the cabin. For these multi-dimensional data, the primary characteristic factor which mainly affects the fan adjustment and the secondary characteristic factor which secondarily affects the fan adjustment can be determined. In order to realize the fine personalized adjustment requirement, the secondary characteristic factor can be added with disturbance under the condition of having the same primary characteristic factor, so as to obtain multiple disturbance data corresponding to the actual multi-dimensional data. Subsequently, a suitable target fan adjustment model is screened out from the numerous fan adjustment models, and the required fan adjustment parameter is output through the target fan adjustment model. In this process, the selected target fan adjustment model can distinguish the actual multi-dimensional data and the generated disturbance data, so that the finally output fan adjustment parameter can be clustered into different clustering clusters. In this way, the different driving scenes have the same primary characteristic factor, but as long as the secondary characteristic factor changes, the final fan adjustment parameter can also change, so as to realize the fine personalized adjustment process. BRIEF DESCRIPTION OF DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the specific embodiments or prior art of the present application, the drawings needed in the specific embodiments or prior art description will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0045] Figure 1 A flowchart of a fan working parameter control method for an automobile seat provided by an embodiment of the present application;

[0046] Figure 2A schematic diagram of a fan working parameter control system for a car seat is provided for the embodiments of the present application.

[0047] Figure 3 A structural schematic diagram of a computer device is provided for the embodiments of the present application. DETAILED DESCRIPTION

[0048] To make the objectives, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described below in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0049] Please refer to Figure 1 The embodiments of the present application provide a fan working parameter control method for a car seat, which comprises the following steps.

[0050] S1: Obtain the object features of the driver or passenger seated on the car seat and the environmental features in the cabin, wherein the object features are used to at least represent the gender, preference, and contact area of the driver or passenger with the car seat, and the environmental features are used to represent the current actual temperature in the cabin and the target adjustment temperature of the air conditioner.

[0051] In the present embodiment, the intelligent cabin can collect the object features of the driver or passenger and the environmental features in the cabin through the information pre-recorded by the driver or passenger in the vehicle machine and various sensors. In actual application, the identity of the driver or passenger can be determined through facial feature recognition or voiceprint recognition, so as to obtain a series of information such as the gender and preference of the driver or passenger in the vehicle machine. At the same time, the contact area of the driver or passenger with the car seat can be determined through the pressure sensors distributed on the seat, which can be a ratio value between 0 and 1. The real-time temperature in the cabin can be collected through the temperature sensor, and the target adjustment temperature of the air conditioner can be obtained by the vehicle machine. The above information can comprehensively reflect the current fan adjustment scene, and subsequent personalized fan adjustment can be based on these information. Of course, more or less information can be obtained for fan adjustment in actual application, which is determined according to the requirements of the application scene.

[0052] In the present embodiment, the preference of the driver or passenger can be used to represent the preference degree of the driver or passenger for cold and warm, which is represented by a value in a specified interval, wherein the smaller the value, the more the driver or passenger prefers cold, and the higher the value, the more the driver or passenger prefers warm. For example, the value in the specified interval can be a value between 0 and 1.

[0053] S2: constructing a feature matrix for controlling fan working parameters based on the object features and the environment features, and determining main feature factors and secondary feature factors in the feature matrix.

[0054] In the embodiment, the object features and the environment features can be collected according to a preset sampling period, and the object features and the environment features with the same timestamp are spliced into a timestamp sequence. For example, at t1, each item of data in the object features and the environment features can be collected, and each item of collected data is arranged in a predetermined order, thereby constituting a timestamp sequence at t1. Each different time point can have a respective timestamp sequence. Subsequently, each of the timestamp sequences is arranged in a time sequence relationship, thereby constituting a feature matrix for controlling fan working parameters. In the feature matrix, different rows represent timestamp sequences at different time points.

[0055] In the embodiment, after the feature matrix is generated, the feature matrix can be analyzed to determine the main feature factors and the secondary feature factors. Specifically, each feature factor at different timestamps can be identified in the feature matrix. The feature factors can refer to gender, preference, contact area, actual temperature, target adjustment temperature, etc. in step S1. For any feature data, the feature data can be used as index data, and query data and attribute data can be determined from other feature data. Subsequently, an index value of the index data can be calculated based on the association between the index data and the query data and the attribute data. Finally, according to the index values of each feature data, the main feature factors and the secondary feature factors can be determined. For example, the feature factor corresponding to the feature data with the highest index value can be used as the main feature factor, and other feature factors except the main feature factor can be used as the secondary feature factors. Of course, the feature factors with index values less than a certain threshold can also be filtered out, and these filtered feature factors are neither main feature factors nor secondary feature factors.

[0056] In the embodiment, after the selected feature data is used as index data, the query data and the attribute data can be determined in order from high to low according to the relevance of the feature data. For example, if the feature data of the contact area is used as the index data, then the gender and the preference can be used as the query data, and the actual temperature and the target adjustment temperature can be used as the attribute data.

[0057] In one embodiment, when calculating the index value of the index data, the index data, the query data and the attribute data can be divided respectively according to the same time span (for example, with a time length t as the span), to obtain respective local data. For example, the index data, the query data and the attribute data all contain data of N timestamps, and now they are divided according to a time length t as the span. Assuming that the time length t contains m timestamps, then the three kinds of data can all be divided to obtain N / m local data.

[0058] When calculating the index value, for any first local data of the index data, the similarity between the first local data and each local data of the query data can be calculated, to obtain a plurality of similarities. Since the divided local data are all in the form of vectors, the similarity between two vectors can be obtained by calculating the Euclidean distance between the vectors. The smaller the distance, the higher the similarity.

[0059] After calculating each similarity, each similarity can be used as a weight coefficient (which needs to be normalized in some scenarios) to weight each local data of the attribute data, and the weighted local data are spliced to obtain the index feature. For example, the current similarities are k1, k2 and k3, and the local data of the attribute data are D1, D2 and D3. Then, k1, k2 and k3 can be used as the weight coefficients of D1, D2 and D3 respectively for weighting, and after weighting, the three data are spliced to obtain the index feature.

[0060] Subsequently, the index feature is input into an index value calculation model, and the index value corresponding to the index feature can be obtained. The index value calculation model is trained based on a large number of samples. The samples can be a plurality of index feature samples, each of which is labeled with a standard index value, so as to perform supervised training.

[0061] S3: In the case of keeping the main feature factor unchanged, different perturbation features are added to the secondary feature factor, and a plurality of perturbation matrices are constructed according to the main feature factor and the secondary feature factor after adding the perturbation features.

[0062] In the embodiment, in order to realize a fine personalized fan adjustment process, the secondary feature factor can be perturbed. Thus, even if different fan adjustment scenarios have the same main feature factor, as long as the perturbation of the secondary feature factor is large enough, different fan adjustment parameters will be generated.

[0063] Specifically, the added disturbance feature can be realized by a disturbance weight, and the disturbance weight can be obtained based on big data analysis for different secondary characteristic factors. The disturbance weight selected herein needs to represent sufficient disturbance to the secondary characteristic factor, so as to obtain a disturbance matrix different from the characteristic matrix.

[0064] S4: selecting a target fan adjustment model from the plurality of fan adjustment models pre-trained, wherein the target fan adjustment model processes the characteristic matrix and each disturbance matrix, and obtains respective fan adjustment parameters, and if clustering analysis is performed on each fan adjustment parameter, the fan adjustment parameter corresponding to the characteristic matrix and the fan adjustment parameter corresponding to the disturbance matrix are located in different clustering clusters.

[0065] In the embodiment, in order to obtain accurate fan adjustment parameters, different fan adjustment model sets can be selected for different primary characteristic factors. That is, for any primary characteristic factor, multiple different fan adjustment models can be provided, which can significantly distinguish secondary characteristic factors, so as to output different fan adjustment parameters in the case of large disturbance of secondary characteristic factors.

[0066] Specifically, each fan adjustment model corresponding to the primary characteristic factor is obtained in the following manner:

[0067] For any primary characteristic factor, the training sample of the primary characteristic factor is obtained, and the first label value of the fan adjustment parameter is set for the training sample. It should be noted that the first label value of the fan adjustment parameter corresponding to the training sample obtained herein is the same, that is, although the specific values of the secondary characteristic factors in the training sample can be different, the difference is small and does not affect the fan adjustment parameter. Subsequently, on the basis of the training sample, the disturbance sample corresponding to the training sample is generated by adding a disturbance feature to the secondary characteristic factor, and the second label value of the fan adjustment parameter of each disturbance sample is set. It should be noted that the disturbance sample herein is a sample essentially different from the training sample, and therefore the second label value added is different from the first label value, and in actual application, the value of the second label value can also have multiple values as the number of disturbance samples increases. When the first label value and the second label value are subjected to clustering analysis, the first label value and the second label value are located in different clustering clusters, thereby embodying the essential difference between the training sample and the disturbance sample.

[0068] The fan adjustment model is trained by using the training sample and the perturbation sample, so that the prediction results output by the trained fan adjustment model correspond to the first label value and the second label value respectively. Finally, the trained fan adjustment model is taken as a fan adjustment model of the main feature factor. In this way, for the training sample and the perturbation sample with significant differences, the results output by the fan adjustment model are also significantly different (located in different clustering clusters), and as the number of perturbation samples increases, the number of trained fan adjustment models also increases.

[0069] After training the plurality of fan adjustment models corresponding to the main feature factor, in actual application, each fan adjustment model can be used to process the feature matrix and each perturbation matrix. If it is found that a certain fan adjustment model processes the feature matrix and each perturbation matrix and obtains respective fan adjustment parameters, and if clustering analysis is performed on each fan adjustment parameter, the fan adjustment parameters corresponding to the feature matrix and the fan adjustment parameters corresponding to the perturbation matrix are located in different clustering clusters, it is indicated that the selected fan adjustment model is correct, and the selected fan adjustment model can be used to output the result of the feature matrix.

[0070] S5: Obtain the fan adjustment parameter output by the target fan adjustment model for the feature matrix, and perform fan adjustment based on the obtained fan adjustment parameter.

[0071] The technical solution provided by the application can collect the object features of the driver and the environmental features in the cabin, and can reflect the current driving scene through multi-dimensional data. For these multi-dimensional data, the main feature factor which mainly affects the fan adjustment and the secondary feature factor which secondarily affects the fan adjustment can be determined. In order to realize the fine personalized adjustment requirement, the secondary feature factor can be added with perturbation under the condition of having the same main feature factor, so as to obtain multiple perturbation data corresponding to the actual multi-dimensional data. Subsequently, a suitable target fan adjustment model is selected from a plurality of fan adjustment models, and the required fan adjustment parameter is output by the target fan adjustment model. In this process, the selected target fan adjustment model can distinguish the actual multi-dimensional data and the generated perturbation data, so that the finally output fan adjustment parameter can be clustered into different clustering clusters. In this way, different driving scenes have the same main feature factor, but as long as the secondary feature factor changes, the final fan adjustment parameter can also change, so that the fine personalized adjustment process is realized.

[0072] Please refer to Figure 2 The application also provides a fan working parameter control system for an automobile seat, which comprises:

[0073] characteristics of an occupant seated on a car seat and environmental characteristics in a cabin, wherein the object characteristics are used to at least represent gender, preference, and contact area of the occupant with the car seat, and the environmental characteristics are used to represent current actual temperature and target adjustment temperature of air conditioning in the cabin;

[0074] a factor determination unit configured to construct a feature matrix for controlling fan working parameters based on the object characteristics and the environmental characteristics, and determine primary feature factors and secondary feature factors in the feature matrix;

[0075] a perturbation unit configured to add different perturbation features to the secondary feature factors while keeping the primary feature factors unchanged, and construct a plurality of perturbation matrices according to the primary feature factors and the secondary feature factors after adding the perturbation features;

[0076] a model screening unit configured to screen a target fan adjustment model from a plurality of pre-trained fan adjustment models, wherein the target fan adjustment model processes the feature matrix and each of the perturbation matrices, and obtains respective fan adjustment parameters, and if clustering analysis is performed on each of the fan adjustment parameters, the fan adjustment parameter corresponding to the feature matrix and the fan adjustment parameters corresponding to the perturbation matrices are located in different clustering clusters;

[0077] a fan adjustment unit configured to obtain a fan adjustment parameter output by the target fan adjustment model for the feature matrix, and perform fan adjustment based on the obtained fan adjustment parameter.

[0078] In an embodiment, the factor determination unit is specifically configured to identify feature data of each feature factor at different time stamps in the feature matrix; for any feature data, the feature data is taken as index data, and query data and attribute data are determined from other feature data, an index value of the index data is calculated based on an association between the index data and the query data and attribute data; and the primary feature factors and the secondary feature factors are determined according to the index values of each feature data.

[0079] In one embodiment, the factor determining unit is specifically configured to divide the index data, the query data and the attribute data respectively according to the same time span to obtain respective local data; for any first local data of the index data, the similarity between the first local data and each local data of the query data is calculated to obtain a plurality of similarities; the plurality of similarities are taken as weight coefficients to weight each local data of the attribute data, and the weighted local data are spliced to obtain an index feature; and the index feature is input into an index value calculation model to obtain an index value corresponding to the index feature.

[0080] In one embodiment, the model screening unit trains each fan adjustment model in the following manner:

[0081] For any main feature factor, a training sample of the main feature factor is obtained, and a first label value of a fan adjustment parameter is set for the training sample;

[0082] On the basis of the training sample, a disturbance sample corresponding to the training sample is generated by adding a disturbance feature to a secondary feature factor, and a second label value of a fan adjustment parameter is set for each disturbance sample, wherein, when the first label value and the second label value are subjected to cluster analysis, the first label value and the second label value are located in different cluster clusters;

[0083] The training sample and the disturbance sample are used to train a fan adjustment model, so that the prediction results output by the trained fan adjustment model for the training sample and the disturbance sample correspond to the first label value and the second label value, respectively;

[0084] The trained fan adjustment model is taken as a fan adjustment model of the main feature factor.

[0085] The further function of each module and unit described above is the same as that of the corresponding embodiment described above, and will not be described here.

[0086] The unit in this embodiment refers to an ASIC (Application Specific Integrated Circuit, Application Specific Integrated Circuit) circuit, a processor and a memory executing one or more software or fixed programs, and / or other devices that can provide the above functions.

[0087] Please refer to Figure 3 , Figure 3 A structural schematic diagram of a computer device provided in the embodiments of the present application is shown in FIG. 1. Figure 3As shown, the computer device includes one or more processors 10, memory 20, and interfaces 30 for external devices such as a keyboard and a mouse and peripheral devices such as disk devices or other storage devices. One or more busses 10 can be used to implement the interface between the various internal and external components and can be implemented using any one or more of a variety of bus technologies including a System bus, PCI, SCSI, AGP, Super- I / O bus, etc. Furthermore, various buses can be used in front side buses, back side buses, and other bus configurations based on any bus or messaging technology known to those skilled in the art. Figure 3 The processor 10 is used in the embodiments below as an example.

[0088] The processor 10 can be a central processing unit, a network processor, or a combination thereof. The processor 10 can further include a hardware chip. The hardware chip can be an application specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device can be a complex programmable logic device, a field programmable logic device, a general array logic, or any combination thereof.

[0089] The memory 20 stores instructions that can be executed by the at least one processor 10, so that the at least one processor 10 implements the method shown in the above embodiments.

[0090] The memory 20 can include a program region and a data region. The program region can store an operating system and application programs required by at least one function. The data region can store data created according to the use of the computer device, etc. In addition, the memory 20 can include a high-speed random access memory, and can further include a non-transitory memory such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state memory device. In some alternative embodiments, the memory 20 can optionally include a memory disposed remotely with respect to the processor 10, and these remote memories can be connected to the computer device through a network. Examples of the network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0091] The memory 20 can include a volatile memory such as a random access memory, and can also include a non-volatile memory such as a flash memory, a hard disk, or a solid state disk. The memory 20 can further include a combination of the above-mentioned kinds of memories.

[0092] The computer device further includes a communication interface 30 for communication with other devices or communication networks.

[0093] The embodiments of the present application further provide a computer readable storage medium, and the method according to the embodiments of the present application can be implemented in hardware, firmware, or recorded in a storage medium, or be implemented as computer code originally stored in a remote storage medium or a non-transitory machine readable storage medium and downloaded to a local storage medium and stored in the local storage medium, so that the method described herein can be processed by such software on a storage medium using a general purpose computer, a special purpose processor, or programmable or special hardware. The storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, or a solid state disk, etc. Further, the storage medium can also include a combination of the above-mentioned memories. It can be understood that the computer, the processor, the microprocessor controller, or the programmable hardware includes a storage component that can store or receive software or computer code, when the software or computer code is accessed and executed by the computer, the processor, or the hardware, the method shown in the above embodiments is implemented.

[0094] The system, the device, and the unit illustrated in the above embodiments can be specifically implemented by a computer chip or an entity, or by a product having certain functions. A typical implementation device is a computer. Specifically, the computer can be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0095] For the convenience of description, the above device is described as various units respectively described in functions. Of course, the functions of the units can be implemented in the same or multiple software and / or hardware in the implementation of the present application.

[0096] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a device. Therefore, the present application can adopt a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt a computer program product in the form of being implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program code.

[0097] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 one or more flow or blocks

[0098] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the function specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 one or more flow or blocks

[0099] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 one or more flow or blocks

[0100] It should be further understood that the terms "comprise", "comprising", or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can also include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises a" does not, without more constraints, foreclose the existence of additional identical elements in the process, method, article, or apparatus that comprises the recited element.

[0101] Each of the embodiments in the present specification is described in a progressive manner, and the same or similar parts between each embodiment can be mutually referred to, and each embodiment focuses on the difference from other embodiments. In particular, for the system embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the part of the method embodiments.

[0102] The above merely provides an example of the present application, but is not intended to limit the present application. The present application can have various modifications and changes for those skilled in the art. Any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present application shall be included in the scope of claims of the present application.

[0103] Although the embodiments of the present application are described with reference to the drawings, various modifications and changes can be made by those skilled in the art without departing from the spirit and scope of the present application, and such modifications and changes shall fall within the scope defined by the appended claims.

Claims

1. A method for controlling fan operating parameters for an automotive seat, the method comprising: determining a fan operating parameter; and adjusting the fan operating parameter based on a seat temperature. The method comprises: acquiring object features of a driver or passenger seated on a car seat and environment features in a cabin, wherein the object features are used to at least represent gender, preference, and contact area of the driver or passenger with the car seat, and the environment features are used to represent current actual temperature and target air conditioning temperature adjustment in the cabin; constructing a feature matrix for controlling fan working parameters based on the object features and the environment features, and determining main feature factors and secondary feature factors in the feature matrix; adding different perturbation features to the secondary feature factors while keeping the main feature factors unchanged, and constructing a plurality of perturbation matrices based on the main feature factors and the secondary feature factors after adding the perturbation features; selecting a target fan adjustment model from a plurality of pre-trained fan adjustment models, wherein the target fan adjustment model processes the feature matrix and each of the perturbation matrices, and obtains respective fan adjustment parameters, and if each of the fan adjustment parameters is subjected to cluster analysis, the fan adjustment parameters corresponding to the feature matrix and the fan adjustment parameters corresponding to the perturbation matrices are located in different cluster clusters; acquiring fan adjustment parameters output by the target fan adjustment model for the feature matrix, and performing fan adjustment based on the acquired fan adjustment parameters.

2. The method of claim 1, wherein, The preference is used to represent the preference degree of the driver or passenger for cold and warm, and the preference is represented by a numerical value in a specified interval, wherein the smaller the numerical value, the more the driver or passenger prefers cold, and the higher the numerical value, the more the driver or passenger prefers warm.

3. The method of claim 1, wherein, Constructing a feature matrix for controlling fan working parameters based on the object features and the environment features comprises: collecting the object features and the environment features according to a preset sampling period, and splicing the object features and the environment features with the same time stamp into a time stamp sequence; sequentially arranging each of the time stamp sequences according to a time sequence relationship to form a feature matrix for controlling fan working parameters.

4. The method according to claim 1 or 3, characterized in that, Determining main feature factors and secondary feature factors in the feature matrix comprises: identifying feature data of each feature factor at different time stamps in the feature matrix; for any feature data, taking the feature data as index data, and determining query data and attribute data from other feature data, calculating an index value of the index data based on the association between the index data and the query data and attribute data; determining main feature factors and secondary feature factors according to the index values of each feature data.

5. The method of claim 4, wherein, Calculating the index value of the index data comprises: dividing the index data, the query data, and the attribute data respectively according to the same time span to obtain respective local data; for any first local data of the index data, calculating the similarity between the first local data and each local data of the query data to obtain a plurality of similarities; taking the plurality of similarities as weight coefficients to weight each local data of the attribute data, and splicing each weighted local data to obtain an index feature; inputting the index feature into an index value calculation model to obtain an index value corresponding to the index feature.

6. The method of claim 1, wherein, Each of the fan adjustment models is trained in the following manner: For any primary feature factor, a training sample of the primary feature factor is obtained, and a first label value of a fan adjustment parameter is set for the training sample; On the basis of the training sample, a perturbation sample corresponding to the training sample is generated by adding a perturbation feature to a secondary feature factor, and a second label value of the fan adjustment parameter is set for each perturbation sample, wherein, when the first label value and the second label value are subjected to cluster analysis, the first label value and the second label value are located in different cluster clusters; The fan adjustment model is trained by using the training sample and the perturbation sample, so that the prediction results output by the trained fan adjustment model for the training sample and the perturbation sample correspond to the first label value and the second label value respectively; The trained fan adjustment model is used as a fan adjustment model of the primary feature factor.

7. A fan operating parameter control system for a vehicle seat, characterized by, The system comprises: A feature acquisition unit is configured to acquire object features of a driver or a passenger seated on a seat of a vehicle and environment features in a cabin, wherein the object features are used to represent at least gender, preference, and contact area of the driver or the passenger with the seat, and the environment features are used to represent current actual temperature and target adjustment temperature of an air conditioner in the cabin; A factor determination unit is configured to construct a feature matrix for controlling a fan working parameter based on the object features and the environment features, and determine primary feature factors and secondary feature factors in the feature matrix; A perturbation unit is configured to add different perturbation features to the secondary feature factors while keeping the primary feature factors unchanged, and construct a plurality of perturbation matrices according to the primary feature factors and the secondary feature factors after adding the perturbation features; A model screening unit is configured to screen a target fan adjustment model from a plurality of fan adjustment models trained in advance, wherein the target fan adjustment model processes the feature matrix and each of the perturbation matrices, and obtains respective fan adjustment parameters, and if each of the fan adjustment parameters is subjected to cluster analysis, the fan adjustment parameter corresponding to the feature matrix and the fan adjustment parameter corresponding to the perturbation matrix are located in different cluster clusters; A fan adjustment unit is configured to obtain a fan adjustment parameter output by the target fan adjustment model for the feature matrix, and perform fan adjustment based on the obtained fan adjustment parameter.

8. The system of claim 7, wherein, The factor determination unit is specifically configured to identify feature data of each feature factor at different time stamps in the feature matrix, take any feature data as index data, determine query data and attribute data from other feature data, calculate an index value of the index data based on an association between the index data and the query data and the attribute data, and determine the primary feature factors and the secondary feature factors according to the index values of the feature data.

9. The system of claim 8, wherein, The factor determination unit is specifically configured to: divide the index data, the query data, and the attribute data respectively according to a same time span to obtain respective local data; for any first local data of the index data, calculate similarities between the first local data and each local data of the query data to obtain a plurality of similarities; and use the plurality of similarities as weight coefficients to weight each local data of the attribute data, and splice the weighted local data to obtain an index feature. The index feature is input into an index value calculation model to obtain an index value corresponding to the index feature.

10. The system of claim 7, wherein, The model screening unit trains each fan adjustment model in the following manner: For any main feature factor, a training sample of the main feature factor is obtained, and a first label value of a fan adjustment parameter is set for the training sample; On the basis of the training sample, a disturbance sample corresponding to the training sample is generated by adding a disturbance feature to a secondary feature factor, and a second label value of the fan adjustment parameter is set for each disturbance sample, wherein, when performing cluster analysis on the first label value and the second label value, the first label value and the second label value are located in different cluster clusters; The training sample and the disturbance sample are used to train a fan adjustment model, so that the prediction results output by the trained fan adjustment model for the training sample and the disturbance sample correspond to the first label value and the second label value, respectively; The trained fan adjustment model is used as a fan adjustment model of the main feature factor.

Images