AI-based method for automatic execution of conditions and actions for vehicle environment optimization and the system using it

Abstract

In the AI-based method for automatically executing conditions and operations for vehicle environment optimization according to the present invention, the method comprises the steps of: a service providing server collecting data from a user terminal and a vehicle to generate an artificial intelligence service model optimized for user patterns; and the service providing server controlling a vehicle based on the artificial intelligence service model. The step of the service providing server collecting data from a user terminal and a vehicle to generate an artificial intelligence service model optimized for user patterns comprises: a step in which the service providing server collects at least one of vehicle status data, external environment data, and vehicle linkage data from the vehicle; a step in which the service providing server collects user condition data from a user terminal; a step in which the service providing server analyzes time-based patterns of driving and user behavior based on at least one of vehicle status data, external environment data, vehicle linkage data, and user condition data; a step in which the service providing server analyzes situation-based patterns of driving and user behavior based on at least one of vehicle status data, external environment data, vehicle linkage data, and user condition data; and a step in which the service providing server generates at least one specific condition according to an operation based on the analyzed pattern to train the artificial intelligence service model.

Description

Technology Field

[0001] The present invention relates to an AI-based method for automatically executing conditions and actions for vehicle environment optimization and a system using the same. Specifically, it relates to an AI-based method for automatically executing conditions and actions for vehicle environment optimization and a system using the same that can automatically adjust by identifying the in-vehicle environment as well as conditions and actions pre-set by a user, and execute vehicle operations and services in real time. Background Technology

[0003] While automobiles provide various features for the driver's convenience and safety, most of these functions involve the inconvenience of having to be set manually or operated directly while driving. For example, controlling the vehicle's temperature, music playback, route guidance, and setting Do Not Disturb mode must be adjusted manually by the driver while driving, which can distract their concentration.

[0004] Furthermore, as the connection between smartphones and in-vehicle systems has strengthened, various apps capable of controlling vehicle functions have emerged; however, most systems require the user to manually launch the app. For instance, the system typically involves opening a specific app upon entering the vehicle to provide services necessary while driving. However, this approach entails the inconvenience of users having to launch a specific app or input conditions every time, and there is a lack of systems that automatically execute frequently used functions.

[0005] Prior art includes Korean registered patent No. 10-2625590 (method for setting an optimized driving environment for a user of a vehicle sharing platform and an electronic device for performing the same), but it only includes the steps of obtaining a command to automatically set a driving environment within a vehicle, obtaining an image of a user captured using a camera installed in the vehicle based on the command, obtaining body information of the user by analyzing the image, obtaining setting information for setting the driving environment based on the user's body information and information about the vehicle, and transmitting a command to the vehicle to set the driving environment based on the setting information. The problem to be solved

[0007] The problem that the present invention aims to solve is devised to resolve the issues of the conventional technology described above. It can provide a service optimized for the vehicle and the user based on conditions pre-set by the user and patterns of time periods and situations in which actions occur. By integrally analyzing the vehicle, the external environment, and the user's patterns, it can provide a customized service that automatically reflects the user's behavior and preferences. means of solving the problem

[0009] In the AI-based method for automatically executing conditions and operations for vehicle environment optimization according to the present invention, the method comprises the steps of: a service providing server collecting data from a user terminal and a vehicle to generate an artificial intelligence service model optimized for user patterns; and the service providing server controlling a vehicle based on the artificial intelligence service model. The step of the service providing server collecting data from a user terminal and a vehicle to generate an artificial intelligence service model optimized for user patterns comprises: a step in which the service providing server collects at least one of vehicle status data, external environment data, and vehicle linkage data from the vehicle; a step in which the service providing server collects user condition data from a user terminal; a step in which the service providing server analyzes time-based patterns of driving and user behavior based on at least one of vehicle status data, external environment data, vehicle linkage data, and user condition data; a step in which the service providing server analyzes situation-based patterns of driving and user behavior based on at least one of vehicle status data, external environment data, vehicle linkage data, and user condition data; and a step in which the service providing server generates at least one specific condition according to an operation based on the analyzed pattern to train the artificial intelligence service model.

[0010] A system utilizing an AI-based method for automatically executing conditions and actions for vehicle environment optimization according to the present invention comprises: a service providing server that collects at least one of vehicle status data, external environment data, and vehicle linkage data from a vehicle; collects user condition data from a user terminal; analyzes time-based patterns of driving and user behavior based on at least one of vehicle status data, external environment data, vehicle linkage data, and user condition data; analyzes situation-based patterns of driving and user behavior based on at least one of vehicle status data, external environment data, vehicle linkage data, and user condition data; generates at least one specific condition according to an action based on the analyzed pattern to train an artificial intelligence service model; and further trains the trained artificial intelligence service model to generate a recommendation service by analyzing user preferences to generate a final artificial intelligence service model; a user terminal that transmits user condition data to the service providing server; and a vehicle that transmits vehicle status data, external environment data, vehicle linkage data, and user condition data to the service providing server. Effects of the invention

[0012] According to an embodiment of the present invention, user convenience can be significantly improved by automatically optimizing the in-vehicle environment through real-time analysis of the user's behavioral patterns, the vehicle, and the external environment.

[0013] In addition, by providing AI-based customized services, vehicle settings are automatically adjusted, allowing users to utilize the vehicle more efficiently.

[0014] In addition, the automatic execution of in-vehicle apps and functions can provide users with a consistent and convenient vehicle environment.

[0015] In addition, integration with smart homes allows for integrated control of the vehicle and home environment, enabling more efficient and systematic management of the living space. Brief explanation of the drawing

[0017] FIG. 1 is a flowchart illustrating an AI-based condition and operation automatic execution method for vehicle environment optimization according to an embodiment of the present invention. FIG. 2 is a flowchart illustrating a method for a service providing server according to an embodiment of the present invention to collect data from a user terminal and a vehicle and generate an artificial intelligence service model optimized for user patterns. FIG. 3 is a configuration diagram of a system using an AI-based condition and operation automatic execution method for vehicle environment optimization according to an embodiment of the present invention. Specific details for implementing the invention

[0018] Specific structural or functional descriptions of embodiments according to the concept of the present invention disclosed herein are provided merely for the purpose of explaining embodiments according to the concept of the present invention, and embodiments according to the concept of the present invention may be implemented in various forms and are not limited to the embodiments described herein.

[0019] Embodiments according to the concept of the present invention may be subject to various modifications and may take various forms; therefore, embodiments are illustrated in the drawings and described in detail in this specification. However, this is not intended to limit the embodiments according to the concept of the present invention to specific disclosed forms, and includes all modifications, equivalents, or substitutions that fall within the spirit and scope of the present invention.

[0020] The terms used herein are used merely to describe specific embodiments and are not intended to limit the invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. In this specification, terms such as "comprising" or "having" are intended to indicate the existence of the features, numbers, steps, actions, components, parts, or combinations thereof described herein, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0021] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings attached to this specification.

[0023] FIG. 1 is a flowchart illustrating an AI-based method for automatically executing conditions and operations for vehicle environment optimization according to an embodiment of the present invention, and FIG. 2 is a flowchart illustrating a method in which a service provider server collects data from a user terminal and a vehicle and generates an artificial intelligence service model optimized for user patterns according to an embodiment of the present invention.

[0024] Referring to FIG. 1, a service providing server (200) collects data from a user terminal (100) and a vehicle (300) to generate an artificial intelligence service model optimized for user patterns (S101), and the service providing server (200) controls the vehicle based on the artificial intelligence service model (S103).

[0025] Referring to FIG. 2, a method for a service providing server (200) to collect data from a user terminal (100) and a vehicle (300) to generate an artificial intelligence service model optimized for a user pattern involves the service providing server (200) collecting at least one of vehicle status data, external environment data, and automatic natural movement data from the vehicle (300) (S201), and the service providing server (200) collecting user condition data from the user terminal (100) (S203). The vehicle status data may include at least one of the vehicle type, speed, location, temperature, battery status, and date, but is not necessarily limited thereto. The external environment data may include at least one of the temperature, weather, time of day, and road traffic conditions outside the vehicle, but is not necessarily limited thereto. The vehicle linkage data may include at least one service information linked to the vehicle, but is not necessarily limited thereto.

[0026] The service providing server (200) analyzes the time-based patterns of driving and user behavior based on at least one of vehicle status data, external environment data, vehicle linkage data, and user condition data (S205). If the service providing server (200) analyzes at least one of vehicle status data, external environment data, and vehicle linkage data and determines that the user controls the vehicle status so that it is repeated a certain number of times or more in similar time periods, it can determine that the vehicle status is controlled periodically in that time period. Additionally, if the service providing server (200) analyzes at least one of vehicle status data, external environment data, and vehicle linkage data and determines that the user controls the vehicle status so that it is repeated a certain number of times or more in similar time periods regarding the environment outside the vehicle, it can determine that the vehicle status is controlled periodically in that time period. Additionally, if the service providing server (200) analyzes at least one of vehicle status data, external environment data, and vehicle linkage data and determines that the user uses the service in that time period so that it is repeated a certain number of times or more in similar time periods from at least one service information linked to the vehicle, it can determine that the service is used periodically in that time period. Additionally, the service providing server (200) can analyze at least one of vehicle status data, external environment data, vehicle linkage data, and user condition data, and if there is a pattern related to the time period in the conditions set by the user, it can determine that the user has a pattern in that time period.

[0027] The service providing server (200) analyzes situational patterns regarding driving and user behavior based on at least one of vehicle status data, external environment data, automatic natural movement data, and user condition data (S207). If the service providing server (200) analyzes at least one of vehicle status data, external environment data, and vehicle linkage data and determines that the user controls the vehicle status so that it is repeated a certain number of times or more in similar situations, it can determine that the vehicle status is controlled periodically in that situation. Additionally, if the service providing server (200) analyzes at least one of vehicle status data, external environment data, and vehicle linkage data and determines that the user controls the vehicle status so that it is repeated a certain number of times or more in similar situations regarding the environment outside the vehicle, it can determine that the vehicle status is controlled periodically during that time period. Additionally, if the service providing server (200) analyzes at least one of vehicle status data, external environment data, and vehicle linkage data and determines that the user uses the service in that situation so that it is repeated a certain number of times or more in similar situations from at least one service information linked to the vehicle, it can determine that the service is used periodically in that situation. Additionally, the service providing server (200) can analyze at least one of vehicle status data, external environment data, vehicle linkage data, and user condition data, and if there is a pattern related to the situation in the condition set by the user, it can determine that the user has a pattern in the situation.

[0028] Based on the pattern analyzed by the service providing server (200), at least one specific condition according to the operation is generated to train an artificial intelligence service model (S209). After generating at least one condition as described above, the service providing server (200) can train the artificial intelligence service model to learn the probability of the occurrence of the condition by time period or situation period by using the condition by time period and situation period as training data. The artificial intelligence service model is a prediction and control model for user-customized environment automation, and may use a machine learning or deep learning model, but is not necessarily limited thereto. The service providing server (200) further trains the artificial intelligence service model that has completed training to generate a recommendation service by analyzing the user's preferences to create a final artificial intelligence service model (S211). In addition to predicting conditions by time period and situation period, the service providing server (200) can provide a recommendation service tailored to the user's specific preferences and environment, further personalize the vehicle (300) environment, and maximize the service quality that the user can experience. The service providing server (200) can collect data regarding the user's preferences from the user terminal (100). At this time, the data regarding preferences varies according to the individual user's behavior, choices, and preferences, and may mainly include at least one of the vehicle's driving patterns, vehicle state control, and reactions to the external environment.

[0030] FIG. 3 is a configuration diagram of a system using an AI-based condition and operation automatic execution method for vehicle environment optimization according to an embodiment of the present invention.

[0031] Referring to FIG. 3, a system (10) using an AI-based condition and operation automatic execution method for vehicle environment optimization consists of a user terminal (100), a service provider server (200), and a vehicle (300).

[0032] The user terminal (100) can transmit user condition data to the service provider server (200). The user terminal (100) can transmit data collected from the user to the server through communication with the service provider server (200). In one embodiment, the user terminal (100) can collect time-based conditions, situation-based conditions, and other user condition data set by the user and transmit them to the service provider server. The user terminal (100) can receive settings that match personal needs or preferences from the user, and at this time, the user condition data may include at least one element among time-based, situation, and location.

[0033] The service provision server (200) is composed of a data collection unit (210), a model creation unit (220), a model execution unit (230), a communication unit (240), a storage unit (250), and a control unit (260).

[0034] The data collection unit (210) collects user condition data from the user terminal (100) and can collect at least one of vehicle status data, external environment data, and vehicle linkage data from the vehicle (300). The user condition data may refer to data regarding conditions pre-set by the user. The vehicle status data may include at least one of the vehicle type, speed, location, temperature, battery status, and date. The external environment data may include at least one of the temperature, weather, time of day, and road traffic conditions outside the vehicle. The vehicle linkage data may include at least one service information linked to the vehicle.

[0035] The above model generation unit (220) is composed of a time-based pattern analysis module (221), a situation-based pattern analysis module (222), a condition generation module (223), and an additional learning module (224).

[0036] The above time-based pattern analysis module (221) can analyze time-based patterns of driving and user behavior based on the vehicle status data, external environment data, vehicle linkage data, and user condition data. The time-based pattern analysis module (221) can determine that the vehicle status is controlled periodically during the corresponding time period if it analyzes at least one of the vehicle status data, external environment data, and vehicle linkage data and determines that the user controls the vehicle status so that it is repeated a certain number of times or more in similar time periods. At this time, the certain number of times may be 5 or more, but is not necessarily limited thereto. In one embodiment, if the user sets an electric charging station as a waypoint 5 or more times before going to work between 7:00 AM and 8:00 AM on weekdays when the type of vehicle (300) is an electric vehicle and the battery level is 30% or less, the time-based pattern analysis module (221) can determine that the user has a pattern during the corresponding time period.

[0037] Additionally, the time-based pattern analysis module (221) can analyze at least one of vehicle status data, external environment data, and vehicle linkage data, and if it analyzes that the user controls the vehicle status so that it is repeated a certain number of times or more in a similar time period with respect to the environment outside the vehicle, it can determine that the user controls the vehicle status periodically in that time period. At this time, the certain number of times may be 5 or more, but is not necessarily limited thereto. In one embodiment, if the user turns on the heater power 5 or more times when the temperature outside the vehicle (300) is 10℃ or lower between 7:00 AM and 8:00 AM on weekdays, the time-based pattern analysis module (221) can determine that the user has a pattern in that time period.

[0038] Additionally, the time-based pattern analysis module (221) can determine that the user uses the service periodically during that time period by analyzing at least one of the vehicle status data, external environment data, and vehicle linkage data, and by analyzing at least one service information linked to the vehicle, if the user uses the service repeatedly at similar times during that time period. At this time, the certain number of times may be 5 or more, but is not necessarily limited thereto. In one embodiment, if music of the 'jazz' genre is played 5 or more times in a music app (Spotify, Apple Music, Melon, etc.) linked to the vehicle (300) between 7:00 AM and 8:00 AM on weekdays, the time-based pattern analysis module (221) can determine that the user has a pattern during that time period.

[0039] Additionally, the time-based pattern analysis module (221) analyzes at least one of vehicle status data, external environment data, vehicle linkage data, and user condition data, and if a pattern related to the time period exists in the conditions set by the user, it can determine that the user has a pattern in that time period. In one embodiment, the user condition data may be data that automatically sets the desired temperature of the heater inside the vehicle (300) to 24℃ when the engine of the vehicle (300) is turned on during December to February, and in this case, the time-based pattern analysis module (221) can determine that the user has a pattern in that time period. In another embodiment, the user condition data may be data that automatically plays the radio when the engine of the vehicle (300) is turned on between 7:00 AM and 8:00 AM on weekdays, and in this case, the time-based pattern analysis module (221) can determine that the user has a pattern in that time period.

[0040] The above situational pattern analysis module (222) can analyze situational patterns regarding driving and user behavior based on the vehicle status data, external environment data, vehicle linkage data, and user condition data. The situational pattern analysis module (222) can determine that the vehicle status is controlled periodically in a given situation if it analyzes at least one of the vehicle status data, external environment data, and vehicle linkage data and determines that the user controls the vehicle status so that it is repeated a certain number of times or more in similar situations. At this time, the certain number of times may be 5 or more, but is not necessarily limited thereto. In one embodiment, if the user turns on the cruise mode 5 or more times while driving at a speed of 100 km / s or more for 5 minutes or more, the situational pattern analysis module (222) can determine that the user has a pattern in that situation.

[0041] Additionally, the situational pattern analysis module (222) analyzes at least one of vehicle status data, external environment data, and vehicle linkage data, and if it analyzes that the user controls the vehicle status so that it is repeated a certain number of times or more in a similar situation regarding the environment outside the vehicle, it can determine that the vehicle status is controlled periodically during that time period. At this time, the certain number of times may be 5 or more, but is not necessarily limited thereto. In one embodiment, if the internal humidity is controlled 5 or more times on a rainy day, the situational pattern analysis module (222) can determine that the user has a pattern in that situation.

[0042] Additionally, the situational pattern analysis module (222) analyzes at least one of vehicle status data, external environment data, and vehicle linkage data, and can determine that the user uses the service periodically in a given situation if the user uses the service repeatedly more than a certain number of times in a similar situation from at least one service information linked to the vehicle. At this time, the certain number of times may be 5 or more, but is not necessarily limited thereto. In one embodiment, if the user moves to a specific destination via a specific route 5 or more times, the situational pattern analysis module (222) can determine that the user has a pattern in that situation.

[0043] Additionally, the situational pattern analysis module (221) can analyze at least one of vehicle status data, external environment data, vehicle linkage data, and user condition data, and if a pattern related to the situation exists in the conditions set by the user, it can determine that the user has a pattern in that situation. In one embodiment, the user condition data may be data that causes the emergency lights of the vehicle (300) to automatically light up for 5 seconds when emergency braking occurs, and in this case, the situational pattern analysis module (222) can determine that the user has a pattern in that situation.

[0044] The above condition generation module (223) can train an artificial intelligence service model by generating at least one specific condition according to an operation based on the pattern analyzed by the time-based pattern analysis module (221) and the situation-based pattern analysis module (222). In one embodiment, the condition generation module (223) can generate a condition that if the time-based pattern analysis module (221) determines that there is a time-based pattern because there are 5 or more instances where the user sets an electric charging station as a waypoint before going to work when the type of vehicle (300) is electric and the battery level is 30% or less between 7:00 AM and 8:00 AM on weekdays, the electric charging station is set as a waypoint before going to work when the battery level is 30% or less between 7:00 AM and 8:00 AM on weekdays. In another embodiment, the condition generation module (223) can generate a condition in which the cruise mode is automatically turned on when the vehicle (300) drives at a speed of 100 km / h or higher for 5 minutes or more, if the situational pattern analysis module (222) determines that there is a situational pattern because there are 5 or more instances where the user turns on the cruise mode when driving at a speed of 100 km / h or higher for 5 minutes or more.

[0045] The condition generation module (223) can generate at least one condition as described above, and then use the conditions by time period and situation period as training data to train the artificial intelligence service model to learn the probability of the conditions occurring by time period or situation period. The artificial intelligence service model is a prediction and control model for user-customized environment automation, and may use machine learning or deep learning models, but is not necessarily limited thereto. The artificial intelligence service model may use a regression analysis model that predicts the probability of conditions occurring by time period or situation period, or a classification model that classifies situations satisfying specific conditions. In addition, the artificial intelligence service model may use a neural network that learns complex patterns through a multilayer neural network to predict and automatically control when conditions occur in the future based on the information learned by the neural network learning the pattern of a user repeatedly performing a specific action at a specific time period, or a recurrent neural network (RNN) model that is advantageous for analyzing patterns occurring over time and can be used to predict conditions by time period and situation period, but is not necessarily limited thereto.

[0046] The condition generation module (223) can preprocess at least one generated condition through normalization, categorization, and missing value processing. The condition generation module (223) can train an artificial intelligence service model through a machine learning algorithm or a deep learning model based on the preprocessed data. During the training process, the artificial intelligence service model can learn user patterns and predict actions that satisfy the corresponding conditions.

[0047] In this specification, the terms neural network, network function, and neural network may be used interchangeably. A neural network may consist of a set of interconnected computational units, which may generally be referred to as nodes. These nodes may also be referred to as neurons. A neural network is composed of at least one node. The nodes (or neurons) constituting the neural network may be interconnected by one or more links. Within a neural network, one or more nodes connected via links may form a relative relationship between an input node and an output node. The concepts of input and output nodes are relative; any node in an output node relationship with respect to one node may be in an input node relationship with respect to another node, and vice versa. As described above, the input node versus output node relationship may be formed around links. One or more output nodes may be connected to a single input node via links, and vice versa.

[0048] In a relationship between an input node and an output node connected through a single link, the value of the output node's data can be determined based on the data input to the input node. Here, the link interconnecting the input node and the output node may have a weight. The weight can be variable and can be varied by the user or an algorithm to enable the neural network to perform the desired function. For example, if one or more input nodes are interconnected to a single output node by respective links, the output node's value can be determined based on the values ​​input to the input nodes connected to the output node and the weights set on the links corresponding to each input node.

[0049] As described above, a neural network consists of one or more nodes interconnected through one or more links, forming input-output node relationships within the network. The characteristics of a neural network can be determined by the number of nodes and links within the network, the association relationships between the nodes and links, and the weight values ​​assigned to each link. For example, if two neural networks exist with the same number of nodes and links but different link weight values, the two neural networks can be recognized as being different from each other.

[0050] A neural network can be composed of a set of one or more nodes. A subset of nodes constituting a neural network can form a layer. Some of the nodes constituting a neural network can form a layer based on their distances from an initial input node. For example, a set of nodes with a distance of n from an initial input node can form n layers. The distance from the initial input node can be defined by the minimum number of links that must be traversed to reach that node from the initial input node. However, this definition of a layer is arbitrary for illustrative purposes, and the degree of a layer within a neural network can be defined in a way different from that described above. For example, a layer of nodes may be defined by its distance from a final output node.

[0051] Initial input nodes may refer to one or more nodes within a neural network to which data is directly input without passing through links in their relationships with other nodes. Alternatively, in terms of link-based relationships between nodes within the neural network, they may refer to nodes that do not have other input nodes connected by links. Similarly, final output nodes may refer to one or more nodes within a neural network that do not have output nodes in their relationships with other nodes. Furthermore, hidden nodes may refer to nodes constituting the neural network that are neither initial input nodes nor final output nodes.

[0052] A neural network according to one embodiment of the present invention may have the number of nodes in the input layer equal to the number of nodes in the output layer, and may be a neural network in which the number of nodes decreases and then increases again as it progresses from the input layer to the hidden layer. Additionally, a neural network according to another embodiment of the present disclosure may have the number of nodes in the input layer less than the number of nodes in the output layer, and may be a neural network in which the number of nodes decreases as it progresses from the input layer to the hidden layer. Additionally, a neural network according to yet another embodiment of the present disclosure may have the number of nodes in the input layer greater than the number of nodes in the output layer, and may be a neural network in which the number of nodes increases as it progresses from the input layer to the hidden layer. A neural network according to yet another embodiment of the present disclosure may be a neural network in which the above-described neural networks are combined.

[0053] A deep neural network (DNN) can refer to a neural network that includes multiple hidden layers in addition to input and output layers. Deep neural networks allow for the identification of the latent structures of data. In other words, they can identify the latent structures of photos, text, videos, audio, and music (for example, what objects are present in a photo, what the content and emotions of the text are, what the content and emotions of the audio are, etc.). Deep neural networks include convolutional neural networks (CNN) and recurrent neural networks (RNN).

[0054] It may include recurrent neural networks, auto encoders, Generative Adversarial Networks (GANs), restricted Boltzmann machines (RBMs), deep belief networks (DBNs), Q networks, U networks, Siamese networks, Generative Adversarial Networks (GANs), etc. The description of deep neural networks described above is merely illustrative and the present disclosure is not limited thereto.

[0055] In one embodiment of the present invention, the network function may include an autoencoder. The autoencoder may be a type of artificial neural network for outputting output data similar to the input data. The autoencoder may include at least one hidden layer, and an odd number of hidden layers may be placed between the input and output layers. The number of nodes in each layer may be reduced from the number of nodes in the input layer to an intermediate layer called a bottleneck layer (encoding), and then expanded symmetrically from the bottleneck layer to the output layer (symmetric to the input layer). The autoencoder may perform non-linear dimensionality reduction. The number of input and output layers may correspond to the dimension after preprocessing of the input data. In the autoencoder structure, the number of nodes in the hidden layer included in the encoder may have a structure in which the number of nodes decreases as it moves away from the input layer. If the number of nodes in the bottleneck layer (the layer with the fewest nodes located between the encoder and decoder) is too small, a sufficient amount of information may not be transmitted, so it may be maintained at a certain number or more (e.g., more than half of the input layer).

[0056] Neural networks can be trained using at least one of supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning. The training of a neural network may be the process of applying knowledge to the neural network to perform a specific action.

[0057] Neural networks can be trained to minimize the error in their output. The training process involves repeatedly inputting training data into the network, calculating the error between the network's output and the target for the training data, and updating the weights of each node by backpropagating the error from the output layer to the input layer in a direction that reduces the error. In the case of supervised learning, training data is used where the correct answer is labeled for each data point (i.e., labeled training data), whereas in the case of unsupervised learning, the correct answer may not be labeled for each training data point. For instance, in the case of supervised learning for data classification, the training data may consist of data where each training point is labeled with a category. The labeled training data is input into the neural network, and the error can be calculated by comparing the network's output (category) with the label of the training data. As another example, in the case of unsupervised learning for data classification, the error can be calculated by comparing the input training data with the neural network's output. The calculated error is backpropagated in the neural network (i.e., from the output layer to the input layer), and through backpropagation, the connection weights of each node in each layer of the neural network can be updated. The amount of change in the connection weights of each node being updated can be determined by the learning rate. The neural network's calculation of the input data and the backpropagation of the error can constitute a learning cycle (epoch). The learning rate can be applied differently depending on the number of iterations of the neural network's learning cycle. For example, a high learning rate can be used in the early stages of training to quickly achieve a certain level of performance and increase efficiency, while a low learning rate can be used in the later stages to improve accuracy.

[0058] In the training of neural networks, training data is generally a subset of real-world data (i.e., the data intended to be processed using the trained neural network); therefore, a training cycle may exist where errors on the training data decrease but errors on real-world data increase. Overfitting is a phenomenon in which errors on real-world data increase due to excessive training on the training data. For example, a neural network trained on cats by showing it yellow cats may fail to recognize cats other than yellow ones as cats, which can be a type of overfitting.

[0059] Overfitting can cause an increase in errors in machine learning algorithms. Various optimization methods can be used to prevent such overfitting. To prevent overfitting, methods such as increasing the training data, regularization, dropout (which disables some nodes in the network during training), and the use of batch normalization layers can be applied.

[0060] The additional learning module (224) can generate a final artificial intelligence service model by further training the artificial intelligence service model trained by the condition generation module (223) to generate a recommendation service by analyzing the user's preferences. In addition to predicting conditions by time of day and situation, the additional learning module (224) can provide a recommendation service tailored to the user's specific preferences and environment, further personalize the vehicle (300) environment, and maximize the quality of service that the user can experience. The additional learning module (224) can collect data regarding the user's preferences from the user terminal (100). At this time, the data regarding preferences varies according to the user's individual behavior, choices, and preferences, and may mainly include at least one of the vehicle's driving patterns, vehicle state control, and reactions to the external environment. The additional learning module (224) can expand the existing learning dataset by reinforcing data related to the user's preferences. In one embodiment, additional data occurring at specific conditions or at time of day preferred by the user can be collected to enable the model to make more precise predictions. Additionally, the additional learning module (224) can readjust the existing model to match the user's preferences and reflect user characteristics in the learned model. Additionally, the additional learning module (224) can generate a recommendation service based on the learned user preferences, and the recommendation service may include automated function control of the vehicle and optimize user convenience. In one embodiment, the final artificial intelligence service model additionally trained by the additional learning module (224) can learn the user's preference of always setting the vehicle temperature to 24 degrees during a specific time period and generate a recommendation service so that the vehicle automatically sets the temperature to 24℃ during that time period.

[0061] The above model execution unit (230) can control the vehicle based on at least one of the artificial intelligence service model and the final artificial intelligence service model generated by the above model generation unit (220).

[0062] The communication unit (240) enables each component of the user terminal (100), vehicle (300), and service provider server (200) to communicate through a network. The network means may include at least one CDMA-based (or HSDPA-based) mobile communication network, and / or an IEEE 802.16x-based high-speed wireless internet, and / or an IEEE 802.11x-based wireless LAN communication network, but is not necessarily limited thereto. The base unit (250) can store the contents of each component of the service provider server (200). The control unit (260) can control each component of the service provider server (200).

[0063] The vehicle (300) is composed of a driving unit (310), a service management unit (320), a communication unit (330), and a control unit (340).

[0064] The driving unit (310) is a core component related to the actual driving of the vehicle (300) and can perform a function that supports the vehicle (300) to move smoothly on the road. The driving unit (310) controls at least one of acceleration, deceleration, turning, and braking of the vehicle (300), and can determine the state of the vehicle by continuously collecting data such as the vehicle's position and speed. The driving unit (310) can extract at least one of vehicle state data and external environment data and transmit it to the service provider server (200).

[0065] The above service management unit (320) is configured to adjust the service functions of the vehicle (300) and can be linked with an external service (app), and can extract vehicle linkage data including at least one service information linked with the vehicle and transmit it to the service provision server (200).

[0066] The communication unit (330) enables each component of the vehicle (300) and the service provider server (200) to communicate through a network. The network means may include at least one CDMA-based (or HSDPA-based) mobile communication network, and / or an IEEE 802.16x-based high-speed wireless internet, and / or an IEEE 802.11x-based wireless LAN communication network, but is not necessarily limited thereto. The control unit (340) can control each component of the vehicle (300).

[0068] The invention has been described with reference to embodiments illustrated in the drawings, but this is merely illustrative, and those skilled in the art will understand that various modifications and equivalent alternative embodiments are possible therefrom. Accordingly, the true technical scope of protection of the invention should be determined by the technical spirit of the appended claims. Explanation of the symbols

[0070] 10; System using an AI-based method for automatic execution of conditions and actions for vehicle environment optimization 100; User terminal 200; Service provider server 210; Data Collection Department 220; Model creation section 221; Time-based pattern analysis module 222; Situational Pattern Analysis Module 223; Condition generation module 224; Additional learning module 230; Model Execution Unit 240; Communications Department 250; storage section 260; control unit 300; Vehicle 310; Driving part 320; Service Management Department 330; Communications Department 340; control unit

Claims

Claim 1 A method for automatically executing AI-based conditions and actions for optimizing a vehicle environment, comprising: a step in which a service providing server collects data from a user terminal and a vehicle to generate an artificial intelligence service model optimized for user patterns; and a step in which the service providing server controls a vehicle based on the artificial intelligence service model, wherein the step in which the service providing server collects data from a user terminal and a vehicle to generate an artificial intelligence service model optimized for user patterns comprises: a step in which the service providing server collects at least one of vehicle status data, external environment data, and vehicle linkage data from the vehicle; a step in which the service providing server collects user condition data from the user terminal; a step in which the service providing server analyzes time-based patterns of driving and user behavior based on at least one of vehicle status data, external environment data, vehicle linkage data, and user condition data; a step in which the service providing server analyzes situation-based patterns of driving and user behavior based on at least one of vehicle status data, external environment data, vehicle linkage data, and user condition data; and a step in which the service providing server generates at least one specific condition according to an action based on the analyzed pattern and trains the artificial intelligence service model. Claim 2 In claim 1, the step of the service providing server collecting data from a user terminal and a vehicle to generate an artificial intelligence service model optimized for user patterns further includes the step of the service providing server generating a final artificial intelligence service model by additionally training the artificial intelligence service model that has completed training to generate a recommendation service by analyzing user preferences. This describes an AI-based method for automatically executing conditions and actions for vehicle environment optimization. Claim 3 A method for automatically executing AI-based conditions and operations for vehicle environment optimization according to claim 1, wherein the vehicle status data includes at least one of the vehicle type, speed, location, temperature, battery status, and date, the external environment data includes at least one of the external temperature, weather, time of day, and road traffic conditions, and the vehicle linkage data includes at least one service information linked to the vehicle. Claim 4 A system utilizing an AI-based method for automatically executing conditions and actions for vehicle environment optimization, comprising: a service providing server that collects at least one of vehicle status data, external environment data, and vehicle linkage data from a vehicle; collects user condition data from a user terminal; analyzes time-based patterns of driving and user behavior based on at least one of vehicle status data, external environment data, vehicle linkage data, and user condition data; analyzes situation-based patterns of driving and user behavior based on at least one of vehicle status data, external environment data, vehicle linkage data, and user condition data; generates at least one specific condition according to an action based on the analyzed pattern to train an artificial intelligence service model; and further trains the trained artificial intelligence service model to generate a recommendation service by analyzing user preferences to generate a final artificial intelligence service model; a user terminal that transmits user condition data to the service providing server; and a vehicle that transmits vehicle status data, external environment data, vehicle linkage data, and user condition data to the service providing server. Claim 5 A system using an AI-based method for automatically executing conditions and operations for vehicle environment optimization, characterized in that, in paragraph 4, the vehicle status data includes at least one of the vehicle type, speed, location, temperature, battery status, and date, the external environment data includes at least one of the external temperature, weather, time of day, and road traffic conditions, and the vehicle linkage data includes at least one service information linked to the vehicle.

Images