SYSTEM AND METHOD FOR ADAPTIVE SEAT HEATING OF A VEHICLE

Abstract

The present invention provides a system (110) and a method (700) for adaptive seat heating of a vehicle. The system (110) receives in real time one or more heating parameters associated with the seat configuration for one or more predetermined periods. In response to the reception, the system (110) determines a usage pattern of the one or more heating parameters, wherein the one or more heating parameters are set by an occupant of the seat configuration. The system (110) determines an occupant preference based on the usage pattern. Using a machine learning system, the system (110) dynamically predicts one or more optimal heating parameters for the seat configuration using the occupant preference and one or more variables. The prediction is based on a confidence level associated with the one or more optimal heating parameters.

Description

TECHNICAL AREA

[0001] The present invention relates generally to the field of automotive seating equipment. In particular, the present invention relates to a system and a method for adaptive seat heating of a vehicle using machine learning. BACKGROUND

[0002] US Patent 11634005 B2 discloses a system and a method for a seating arrangement in which a temperature-changing element is located within the seating arrangement. The temperature-changing element is configured to heat or cool an occupant of the seating assembly in which the temperature-changing element is located. Furthermore, a control unit communicates with the temperature-changing element and one or more data sources to control the activation of the temperature-changing element depending on the data. The control unit uses a variety of predefined predictive activation models, each of which establishes different rules for activating the temperature-changing element depending on the data generated by the one or more data sources.The control unit selects at least one of several predefined predictive activation models to control the activation of the temperature-changing element based on the data generated by one or more data sources. The selected predictive activation model automatically activates the temperature-changing element if the data collected during the acquisition step meets the rules of that model.

[0003] Furthermore, conventional systems include manual and automatic options, requiring passengers to constantly monitor their comfort and adjust the seat heating settings to find the best possible comfort under varying conditions (e.g., in winter). The seat heating is applied to the entire seat surface, even if a person isn't in contact with all parts of the seat, which can lead to energy loss. Maintaining a warm interior in winter requires excessive energy consumption, especially in electric vehicles, which can directly impact range. Additionally, conventional systems require manual intervention and are not user-friendly.

[0004] The aforementioned conventional systems and methods use a predetermined predictive activation model to control the activation of the temperature-changing element based on identifiable conditions; they are manual, time-dependent, and not user-friendly. Therefore, there is a well-known need to provide an improved system and method for vehicle seat heating. TASKS OF THE PRESENT INVENTION

[0005] A general object of the present invention is to provide a system and a method for adaptive seat heating that instantly records heating parameters associated with a seat arrangement for predetermined periods of time in order to determine a usage pattern of the heating parameters.

[0006] Another object of the present invention is to provide a system in which the mode of the seat heating can be changed when the system detects changes in the posture / position of the passenger / occupant after a certain time using weight / pressure sensors.

[0007] Another object of the present invention is to provide a system that calculates a resident preference based on the usage pattern for the predetermined time periods.

[0008] Another object of the present invention is to provide a system that correlates environmental variables and occupant preferences in order to dynamically predict optimal heating parameters for the seat assembly based on a confidence assessment.

[0009] Another object of the present invention is to provide a system that uses a deep Gauss model for the dynamic prediction of the optimal heating parameters. SUMMARY

[0010] Aspects of the present invention relate generally to the field of automotive seating equipment. In particular, the present invention relates to a system and a method for adaptive seat heating of a vehicle using machine learning.

[0011] In one aspect, the present invention relates to an adaptive seat heating system. The adaptive seat heating system comprises a processor that is communicatively connected to a vehicle's seating arrangement. The processor is operationally connected to a memory that stores instructions causing the processor to receive, in real time, one or more heating parameters associated with the seating arrangement for one or more predetermined periods. In response to this input, the processor determines a usage pattern for the one or more heating parameters, wherein the one or more heating parameters are set by an occupant of the seating arrangement. The processor calculates an occupant's preference based on this usage pattern.The processor uses a machine learning system to dynamically predict one or more optimal heating parameters for the seat assembly, taking into account the occupant's preference and one or more variables. The prediction is based on a confidence score associated with the one or more optimal heating parameters.

[0012] In one embodiment, the one or more variables may include one or more environmental variables surrounding the vehicle and one or more environmental variables associated with the vehicle.

[0013] In one embodiment, the one or more optimal heating parameters may include one or more temperature settings for the seat assembly, based on the confidence value of the one or more optimal heating parameters.

[0014] In one embodiment, the processor can correlate one or more environmental variables with the occupant's preference to dynamically predict one or more optimal heating parameters for the seat assembly.

[0015] In one embodiment, the machine learning system can include a deep Gauss model that enables the processor to calculate the confidence value based on one or more environmental variables and the occupant's preference for predicting an optimal heating parameter for a predetermined period among one or more predetermined time periods.

[0016] In one aspect, the present invention relates to a method for adaptive seat heating. The method comprises receiving, in real time, one or more heating parameters assigned to the seat arrangement for one or more predetermined periods by a processor connected to an adaptive seat heating system. The one or more heating parameters are set by an occupant of the seat arrangement. The method comprises the processor determining a usage pattern of the one or more heating parameters in response to the reception, wherein the one or more heating parameters are set by an occupant of the seat arrangement. The method comprises the processor determining an occupant's preference based on the usage pattern.The procedure involves the dynamic prediction of one or more optimal heating parameters for the seat arrangement by the processor via a machine learning machine using the occupant's preference and one or more variables, with the prediction being based on a confidence value associated with the one or more optimal heating parameters.

[0017] In one embodiment, the one or more variables may include one or more environmental variables in the environment of the vehicle and one or more environmental variables inside the vehicle.

[0018] In one embodiment, the one or more optimal heating parameters may include one or more temperature settings for the seat assembly, based on the confidence value of the one or more optimal heating parameters.

[0019] In one embodiment, the method may include correlating the one or more environmental variables with the occupant's preference by the processor in order to dynamically predict the one or more optimal heating parameters for the seat assembly based on the confidence value.

[0020] In one embodiment, the method may include the calculation of the confidence value by the processor based on one or more environmental variables and the occupant's preference for predicting an optimal heating parameter by a deep Gauss model for a predetermined period among one or more predetermined time periods. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The accompanying drawings serve to further understand the present invention and are an integral part of this description. The drawings illustrate exemplary embodiments of the present invention and, together with the description, serve to explain the principles of the present invention. Fig. Figure 1 shows an example representation 100 of a proposed system 110 according to an embodiment of the present invention. Fig. Figure 2 shows an example block diagram 200 of the proposed system 110 according to an embodiment of the present invention. Fig. Figure 3 shows an exemplary flowchart 300 of a usage pattern calculated by the proposed system 110 according to an embodiment of the present invention. Fig. Figure 4 shows an example of a vehicle seat assembly 400 according to an embodiment of the present invention. Fig. Figure 5 shows an example representation 500 of a coil which is connected to the vehicle seat assembly 400 of Fig. 4 was realized in accordance with an embodiment of the present invention. Fig. Figure 6 shows an example representation 600 of a utility pattern calculated by the proposed system 110 according to an embodiment of the present invention. Fig. Figure 7 shows an exemplary flowchart of a method 700 for implementing the proposed system 110 in accordance with an embodiment of the present invention. Fig. Figure 8 shows an exemplary computer system 800 in which or with which the proposed system 110 can be implemented in accordance with an embodiment of the present invention. DETAILED DESCRIPTION

[0022] A detailed description of embodiments of the invention, illustrated in the accompanying drawings, follows. The embodiments are described in sufficient detail to clearly convey the invention. However, this level of detail is not intended to limit foreseeable variations of embodiments; rather, it is intended to cover all modifications, equivalents, and alternatives that fall within the spirit and scope of the present invention as defined by the accompanying claims.

[0023] The embodiments described herein relate generally to the field of automotive seating equipment. In particular, the present invention relates to a system and a method for adaptive seat heating using machine learning.

[0024] Various embodiments of the present invention are described in detail with reference to the Fig. 1-8.

[0025] Fig. Figure 1 shows an example representation 100 of a proposed system 110 according to an embodiment of the present invention.

[0026] In one embodiment, the proposed system 110 can be communicatively connected to a vehicle's seating assembly. The system 110 can receive, in real time, one or more heating parameters associated with the seating arrangement for one or more predetermined periods. The one or more heating parameters can be set by an occupant of the seating assembly. In response to the received data, the system 110 can determine a usage pattern for the one or more heating parameters. Furthermore, the system 110 can determine an occupant's preference based on this usage pattern.

[0027] The System 110 can correlate one or more environmental variables and one or more environmental variables with an occupant preference to dynamically predict one or more optimal heating parameters for the seat assembly based on the confidence value.

[0028] In one embodiment, the system 110 can calculate the occupant's preference based on the usage pattern. The system 110 can determine the usage pattern by recording the one or more heating parameters set by the occupant for one or more predetermined time periods, where the one or more heating parameters are recorded as k units over the one or more predetermined time periods. The one or more predetermined time periods can depend on a histogram function or a frequency distribution. Furthermore, the system 110 can dynamically predict one or more optimal heating parameters for the seat assembly using the occupant preference and one or more variables via a machine learning engine.The machine learning engine can include a deep Gaussian model that calculates a confidence score based on one or more environmental variables and occupant preference to predict an optimal heating parameter for a predetermined period. The one or more optimal heating parameters can include one or more temperature settings for the seat assembly, based on the confidence score of the one or more optimal heating parameters. The prediction can be based on a confidence score associated with the one or more optimal heating parameters.

[0029] In one embodiment, an anomaly associated with one or more sensors configured with the vehicle can lead to a lower confidence level.

[0030] As in Fig. As shown in Figure 1, in one embodiment, the system 110 can receive one or more environmental variables 106 surrounding the vehicle and one or more environmental variables 108 associated with the vehicle. The system 110 can determine the occupant's preference (also referred to as user preference) 104 based on the usage pattern. The system 110 can correlate the one or more environmental variables 106 and the one or more environmental variables 108 with the occupant preference 104 and generate one or more optimal heating parameters 112 for the vehicle seat assembly using a machine learning engine. The one or more optimal heating parameters 112 can be generated based on a confidence value associated with the one or more optimal heating parameters 112.

[0031] In one embodiment, the system 110 can identify a location on the seat cushion and backrest of the vehicle seat assembly for the placement of one or more temperature sensors. The system 110 can also enable pressure measurement on the seat cushion and backrest using one or more pressure sensors. During testing, the system 110 can collect data from various drivers. Furthermore, the system 110 can measure various temperatures, such as the ambient temperature, the interior temperature, the seat surface temperature, body temperature, and humidity, depending on whether the seat heating is switched on or off. The system 110 can measure the duration of the seat heating option, the intensity of the seat heating option (e.g., power or current), and the occupant's contact area with the vehicle seat assembly. The system 110 can record various parameters in conjunction with a test vehicle under winter and cold temperatures.

[0032] In one embodiment, the system 110 can analyze the temperature measurement, the pressure measurement, the duration and the frequency of the seat heating and pass these inputs to a model for simulation.

[0033] In one embodiment, the system can record 110 pressure data points associated with various mannequins using a pressure mat under static conditions. Furthermore, anthropometric data from research studies can also be used as a reference to create an algorithm for simulating the model.

[0034] In one embodiment, the system 110 can determine a GPS location (or the last known location) in latitude and longitude (L lat , L lon ) determine. Either using the geographical location or the on-board temperature sensor, the system 110 can estimate an ambient temperature outside the vehicle (T). outSystem 110 can estimate the correlation with temperature changes and the rate of temperature changes (t) for one or more predetermined periods. Therefore, the environmental variables can be represented as X env = {(L lat , L lon ), T out The environmental variables are represented as , t} and calculated at different time intervals based on seat heating usage. The frequency of calculation for the environmental variables may be lower than for the environmental variables, as the environmental variables exhibit a low degree of change and sensitivity to model performance due to natural factors affecting them.

[0035] In one embodiment, the system 110 can be a pressure matrix or a two-dimensional (2D) grid matrix of the standardized pressure values ​​of the pressure sensor (P). matSystem 110 can calculate an energy matrix in which the current and voltage applied to the heating mat at various points can be recorded. System 110 can record the interior temperature of the vehicle (T). in The System 110 can also record a number of seconds associated with the activated heating (t heat System 110 can only record the times for which active heating is predicted. Furthermore, System 110 can record the environmental variables as X. amb = {P mat , E mat , T in , t heat} calculate. These values ​​can be calculated in real time based on feedback from one or more sensors configured in the vehicle and provided as input for the algorithm to estimate heat / voltage predictions for the heating mat.

[0036] In one embodiment, the system 110 can record a user profile associated with the vehicle occupant. The system 110 can use the user profile as a categorical variable instead of setting different models for different users. (U i Furthermore, the System 110 can record the average usage time of the seat heating function in seconds (t avg System 110 can record an occupant's usage pattern in the form of a simple k-length histogram (k=4), whereby System 110 can calculate the occupant's heating setting for different k temperature ranges, weighted by usage time. This can be represented as a vector of length k with normalized time values ​​at k different temperature levels (P k System 110 can represent the user's preference as a combined variable P. user = {Ui, t avg , P k} calculate. This variable can be stored offline in the memory of a processor 202 connected to the System 110 and updated based on the use of the seat heating.

[0037] In one embodiment, the machine learning machine can include a deep Gauss model that enables the system 110 to calculate the confidence value based on one or more environmental variables and the occupant's preference for predicting an optimal heating parameter for a predetermined period among one or more predetermined time periods.

[0038] In one embodiment, deep Gaussian processes can be an extension of Gaussian processes that incorporates the principles of deep learning. Deep Gaussian processes can be represented as a deep belief network based on Gaussian process mappings. In these models, the data can be modeled as the output of a multivariate Gaussian process (GP) using a neural network. The inputs to this Gaussian process can be controlled by another GP. A single-layer model can correspond to a standard GP or the GP-LVM latent variable model, where inference can be performed within the model using approximate variational marginalization.

[0039] In one embodiment, the Gaussian process can be a collection of random variables, a finite number of which have a common Gaussian distribution. This distribution can be defined by its mean function m(x) and covariance function k(x, x'), where the mean function can represent the expected value of the function at any point x. m(x)=E[f(x)]. A covariance function, also called a kernel, can measure the similarity between two points $x$ and $x'$. The covariance function can determine a correlation between the function values ​​at these two points and can be expressed as follows. k(x,x')=E[(f(x)−m(x))(f(x')−m(x'))]. Given a set of training data, the predictive distribution can be calculated. In the Deep GP implementation, the system can specify the number of layers in the neural network architecture and the training data, and initialize the network. The network can predict an estimate of the heating parameters along with the variance, which serves as a proxy for uncertainty. These predictions can then be used for the seat heating. Given a set of training data D = {xi, y} ii=1 N , is the prediction distribution for a new input x* is Gaussian with: μ(x*)=k(x*,X)[K+σn2 I]y−1 σ(2x*)=k(x*,x*)−k(x*,X)[K+σn2 I]−1k(X,x*), where K is the covariance matrix of the training data and σn2 the variance of the noise.

[0040] In one embodiment, System 110 can be used in various applications with differing requirements. This allows System 110 to effectively improve the customer experience and reduce manual intervention. For example, the occupant can decide whether it is too cold (approx. 5 °C) and whether they wish to use the seat heating to a certain degree, e.g., warm up for 10 minutes at level 2. The current level can be manually changed to level 2 (medium) as needed.

[0041] In one embodiment, climatic conditions can change during the journey, for example, when driving from the city to a rural area in winter. The intensity of the seat heating can be adjusted according to the changing weather conditions. The current can be monitored and manually set to different heating levels, depending on the occupant's comfort. In all applications, the System 110 automatically decides on and off, the intensity, and the duration of use based on the environmental conditions and the occupant's contact area with the seat surface. The pressure sensor data can also be used to further develop other functions, such as improving customer comfort, particularly during a massage.

[0042] In one embodiment, confidence estimates 114 from the deep Gaussian model of system 110 can be used to determine whether or not to use the prediction from the current timestamp. The following equation is used to convert the variance estimate into a confidence value: C=e(−α*σ*) , where α is the scaling factor, which is multiplied by the standard deviation σ* The function is multiplied and exponentially applied. The exponential function ensures that the confidence level lies between 0 and 1, where 0 represents the lowest and 1 the highest confidence level. System 110 can use the model's output if the confidence level is above 0.7 (70%) to provide a consistent prediction of occupant comfort.

[0043] In one embodiment, System 110 can use adaptive seat heating with combined variables available from the user, the environment, and the vehicle's sensors. System 110 can use the deep Gaussian model for a robust implementation of the predictive model. System 110 can use occupant behavior and the usage pattern model to make the deep Gaussian model insensitive to fluctuations and changes in the occupant, allowing System 110 to adapt to multiple occupants. Furthermore, System 110 can use a confidence estimation factor from the deep Gaussian model, enabling System 110 to fall back on standard temperatures.

[0044] In one embodiment, the System 110 can provide an analysis of user behavior, enabling highly personalized heating settings that automatically adapt to individual preferences and improve comfort without manual intervention. The System 110 can continuously learn from user behavior and improve its predictions over time to ensure optimal comfort and efficiency. This adaptability to user behavior, combined with the correlations of other variables in the data, can make the System 110 truly robust and dynamic.

[0045] Fig. Figure 2 shows an example block diagram 200 of the proposed system 110 according to an embodiment of the present invention.

[0046] As in Fig. As shown in Figure 2, the System 110 can include one or more processors / processing devices 202, which can be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, logic circuits, and / or any devices that process data based on operating instructions. Among other capabilities, the one or more processor(s) 202 can be configured to retrieve and execute computer-readable instructions stored in a memory 204 of the System 110. The memory 204 can be configured to store one or more computer-readable instructions or routines on a non-volatile, computer-readable storage medium that can be retrieved and executed to create or exchange data packets over a network service. The memory 204 can include any non-volatile storage device, such as...a volatile memory such as random access memory (RAM) or a non-volatile memory such as erasable programmable read-only memory (EPROM), flash memory, and the like.

[0047] In one embodiment, the system 110 may include one or more interfaces 206. The interface(s) 206 may include a variety of interfaces, such as interfaces for data input and output (I / O) devices, storage devices, and the like. The interface(s) 206 may also provide a communication path for one or more components of the system 110. Examples of such components include a processing machine 208 and a database 210.

[0048] The processing machine(s) 208 can be implemented as a combination of hardware and programming (e.g., programmable instructions) to implement one or more functions of the processing machine(s). In the examples described here, such combinations of hardware and programming can be implemented in various ways. For example, the programming for the processing machine(s) 208 can consist of processor-executable instructions stored on a non-volatile, machine-readable storage medium, and the hardware for the processing machine(s) 208 can include a processing resource (e.g., one or more processors) to execute such instructions. In the examples presented here, the machine-readable storage medium can store instructions which, when executed by the processing resource, implement the processing machine(s) 208.In such examples, the system may include the machine-readable storage medium that stores the instructions and the processing resource for executing the instructions, or the machine-readable storage medium may be separate but accessible to both the system and the processing resource. In other examples, the processing machine(s) 208 may be implemented by electronic circuits. The processing module 208 may include a machine learning module 212.

[0049] In one embodiment, the processor 202 can receive one or more heating parameters in real time, which are associated with the seat assembly for one or more predetermined time periods. The processor 202 can store the one or more heating parameters in the database 210. The one or more heating parameters can be set by an occupant of the seat assembly. In response to the reception, the processor 202 can determine a usage pattern of the one or more heating parameters, wherein the one or more heating parameters can be set by an occupant of the seat assembly. Based on the usage pattern, the processor 202 can determine an occupant's preference.

[0050] In one embodiment, the processor 202 can determine a resident's preference based on the usage pattern. The processor 202 can determine the usage pattern by recording the one or more heating parameters set by the resident for one or more predetermined time periods, which comprise k sets of the one or more heating parameters over the one or more predetermined time periods.

[0051] In one embodiment, the processor 202 can dynamically predict one or more optimal heating parameters for the seat assembly via a machine learning machine 212 by using the occupant's preference and one or more variables, the prediction being based on a confidence value associated with the one or more optimal heating parameters. The machine learning machine 212 can include a deep Gaussian model that enables the processor 202 to compute the confidence value based on the one or more environmental variables and the occupant's preference in order to predict an optimal heating parameter for a predetermined period among one or more predetermined time periods.

[0052] In one embodiment, the one or more variables may include one or more environmental variables surrounding the vehicle and one or more environmental variables associated with the vehicle. The one or more optimal heating parameters may include one or more temperature settings for the seat assembly based on the confidence value of the one or more optimal heating parameters.

[0053] In one embodiment, the processor 202 can correlate the one or more environmental variables with the occupant preference to dynamically predict the one or more optimal heating parameters for the seat assembly based on the confidence value.

[0054] Fig. Figure 3 shows an exemplary flowchart 300 of a usage pattern calculated by the proposed system 110 according to an embodiment of the present invention.

[0055] As in Fig. As shown in Figure 3, the flowchart 300 can include the following steps.

[0056] In step 302: The System 110 can receive one or more heating parameters, such as the current temperature at which the user is operating the heating system (temp_current), in degrees Celsius, and the duration for which the seat heating was active during the session (time_spent), in seconds. The System 110 can record this information in a histogram. P_k denotes the existing histogram vector for the user (initially all zeros). TEMP_BINS denotes defined temperature ranges for each bin. The histogram can contain the one or more heating parameters as k units of P_k in TEMP_BINS.

[0057] In step 304: System 110 can determine the bin index k for temp_current. In step 306: System 110 can determine whether the current temperature of the seat heater is lower than the temperature recorded in the container. In step 308: If the determination in step 306 is positive, System 110 can set the k-value to zero and proceed to step 314. In step 310: In response to a negative determination from step 306, System 110 can determine whether temp_current is greater than TEMP_BINS[k-1][1].

[0058] In step 312: In response to a positive determination in step 310, system 110 can set k to K-1. In step 314: In response to a negative determination from step 310, system 110 can increment P_k[k] by the time consumed. In step 316: System 110 can calculate the total time as a sum of P_k. In step 318: System 110 can determine if the total time is greater than zero. In step 320: In response to a positive determination in step 318, system 110 can normalize each bin P_k. In step 322: In response to a negative determination from step 318, system 110 can output updated P_k. This process in flowchart 300 helps in tracking and analyzing the use of the seat heating system based on different temperature ranges and durations.

[0059] Fig. Figure 4 shows an example of a vehicle seat assembly 400 according to an embodiment of the present invention.

[0060] As in Fig. As shown in Figure 4, the vehicle seat assembly 400 can, in one embodiment, include a seat backrest 402 that supports the occupant's back. The vehicle seat assembly 400 can also include a seat cushion 404 that supports the occupant's lower body.

[0061] Fig. Figure 5 shows an example representation 500 of a coil which is connected to the vehicle seat assembly 400 of Fig. 4 was realized in accordance with an embodiment of the present invention.

[0062] As in Fig. As shown in Figure 5, in one embodiment the vehicle seat assembly can include a heating coil 502 arranged under the seat cover of the vehicle seat assembly. When the heating coil 502 is switched on, current can flow from a battery to the heating coil 502 in the cushion 506 to heat the cushion 506.

[0063] Fig. Figure 6 shows an example representation 600 of a usage pattern calculated by the proposed system 110 according to an embodiment of the present invention. It represents the approximately continuous evaluation of the user's heating preferences. In the diagram of Fig. The x-axis represents the temperature and the y-axis represents the proportion of time the seat heating is used.

[0064] As in Fig. As shown in Figure 6, period P1 represents the duration of a user's initial use of the seat heating system. Similarly, periods P2 and P3 represent subsequent periods of use by the same user. For example, during period P1, the user uses the seat heating within a temperature range of 3 to 11 degrees for the entire duration. However, after this initial experience, the same user might next use the seat heating for only half the duration within a temperature range of 3 to 11 degrees and for half the duration within a range of 11 to 17 degrees, depending on their comfort preferences. Similarly, on the third use, they might use the seat heating for the same duration within 3 to 11 degrees, 11 to 17 degrees, and 17 to 23 degrees, again depending on their comfort. In practice, the proportion of usage time may not be the same for each temperature range and may also vary from user to user.The diagram in . Fig. Section 6 serves only to understand how user behavior may develop during subsequent uses.

[0065] Fig. Figure 7 shows an exemplary flowchart of a method 700 for implementing the proposed system 110 in accordance with an embodiment of the present invention.

[0066] As in Fig. As shown in Figure 7, the procedure 700 can include the following steps.

[0067] In step 702: The procedure may involve a System 110 receiving in real time one or more heating parameters associated with the seat assembly for one or more predetermined time periods. In step 704: The procedure may involve the System 110 determining a usage pattern of the one or more heating parameters in response to the receipt, wherein the one or more heating parameters may be set by an occupant of the seat assembly. In step 706: The procedure may involve the System 110 determining an occupant's preference based on the usage pattern.In step 708: The procedure can include the dynamic prediction of one or more optimal heating parameters for the seat assembly by the system 110 via a machine learning machine (212) using the occupant's preference and one or more variables, the prediction being based on a confidence value associated with the one or more optimal heating parameters.

[0068] Fig. Figure 8 shows an exemplary computer system 800 in which or with which the proposed system 110 can be implemented in accordance with an embodiment of the present invention.

[0069] As in Fig.As shown in Figure 8, the computer system 800 can comprise an external storage device 810, a bus 820, main memory 830, read-only memory 840, a mass storage device 850, one or more communication ports 860, and a processor 870. A person skilled in the art will understand that the computer system 800 can comprise more than one processor and communication ports. The communication port(s) 860 can be chosen depending on a network, such as a local area network (LAN), wide area network (WAN), or any network to which the computer system 800 is connected. The main memory 830 can be random access memory (RAM) or any other dynamic storage device generally known in the art. The read-only memory 840 can be any static storage device, such as a PROM (programmable read-only memory) chip for storing static information, e.g.,Startup or BIOS (Basic Input / Output System) commands for, but not limited to, the 870 processor. The 850 mass storage can be any current or future mass storage solution that can be used to store information and / or instructions.

[0070] Bus 820 can connect processor 870 to the other memory, storage, and communication blocks. Optionally, operator and management interfaces, such as a screen, keyboard, and cursor control device, can also be connected to bus 820 to support direct operator interaction with computer system 800. Other operator and management interfaces can be provided via network connections connected through communication port(s) 860. The exemplary computer system 800 mentioned above is not intended to limit the scope of the present invention in any way.

[0071] While the foregoing describes various embodiments of the invention, other and further embodiments of the invention can be developed without departing from the fundamental scope of the invention. The scope of the invention is defined by the following claims. The invention is not limited to the described embodiments, versions, or examples, provided that they are included to enable a person with ordinary technical knowledge to manufacture and use the invention when combined with the information and knowledge available to that person. BENEFITS OF THE PRESENT DISCLOSURE

[0072] The present disclosure offers individual adjustment of the seat heating for different customer types by calculating an occupant preference for predetermined time periods.

[0073] The present disclosure offers an intelligent and self-learning function instead of a standardized option through the Deep Gauss model.

[0074] The present disclosure offers an improved customer experience and greater thermal comfort, as the system dynamically predicts optimal heating parameters for the seat assembly using the Deep Gauss model.

[0075] The present disclosure correlates the environmental variables and the occupant preferences to dynamically predict the optimal heating parameters for the seat assembly based on a confidence assessment.

[0076] The present invention provides a system that reduces the need for manual operation / override, as the system dynamically predicts the optimal heating parameters for the seat assembly based on the occupant's preferences and the variables.

[0077] The present disclosure offers a system with efficiency and energy savings, as the system only heats certain areas of the vehicle seat.

[0078] The present disclosure provides a system that uses a resident behavior and usage pattern model to make the deep Gauss model insensitive to fluctuations and changes in the resident, and allows the system to adapt to multiple residents via a user profile variable. QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] US 11634005 B2

[0002]

Claims

[1] Adaptive seat heating system (110) comprising the following: a processor (202) that is communicatively coupled with a seating arrangement of a vehicle, wherein the processor (202) is operationally coupled with a memory (204), wherein the memory (204) stores instructions that cause the processor (202) to: receive one or more heating parameters in real time, which are linked to the seat assembly for one or more predetermined time periods; to determine a usage pattern of one or more heating parameters in response to the reception, wherein the one or more heating parameters are set by an occupant of the seating arrangement; Determining an inmate's preference based on usage patterns; and Dynamic prediction of one or more optimal heating parameters for the seat assembly via a machine learning machine using the occupant's preference and one or more variables, wherein the prediction is based on a confidence value associated with the one or more optimal heating parameters. [2] System (110) according to claim 1, wherein the one or more variables comprise one or more environmental variables surrounding the vehicle and one or more environmental variables associated with the vehicle. [3] System (110) according to claim 1, wherein the one or more optimal heating parameters comprise one or more temperature settings for the seat assembly based on the confidence value of the one or more optimal heating parameters. [4] System (110) according to claim 2, wherein the processor (202) correlates the one or more environmental variables and the one or more environmental variables with the occupant's preference to dynamically predict the one or more optimal heating parameters for the seating arrangement. [5] System (110) according to claim 2, wherein the machine learning machine comprises a deep Gauss model enabling the processor (202) to calculate the confidence value based on the one or more environmental variables and the occupant's preference for predicting an optimal heating parameter for a predetermined period among the one or more predetermined periods. [6] Method (700) for adaptive seat heating, the method comprising: Received in real time (702) from a processor (202) connected to an adaptive seat heating system (110) of one or more heating parameters associated with the seat arrangement for one or more predetermined periods; Determining (704) a usage pattern of one or more heating parameters by the processor (202) in response to the reception, wherein the one or more heating parameters are set by an occupant of the seating arrangement; Determine (706), by the processor (202), an inmate's preference based on the usage pattern; and dynamic prediction (708) by the processor (202) via a machine learning machine of one or more optimal heating parameters for the seat assembly using the occupant's preference and one or more variables, wherein the prediction is based on a confidence value associated with the one or more optimal heating parameters. [7] Method (700) according to claim 6, wherein the one or more variables comprise one or more environmental variables surrounding the vehicle and one or more environmental variables within the vehicle. [8] Method (700) according to claim 6, wherein the one or more optimal heating parameters comprise one or more temperature settings for the seat assembly based on the confidence value of the one or more optimal heating parameters. [9] Method (700) according to claim 7, comprising correlating the one or more environmental variables and the one or more environmental variables with the occupant's preference by the processor (202) to dynamically predict the one or more optimal heating parameters for the seat assembly based on the confidence value. [10] Method (700) according to claim 7, comprising calculating the confidence value by the processor (202) on the basis of the one or more environmental variables, the one or more environmental variables and the occupant's preference for predicting an optimal heating parameter by a deep Gaussian model for a predetermined period from the one or more predetermined periods.

Images