Method and system for determining and controlling autonomous vehicle based on artificial intelligence model

Abstract

An embodiment of the present disclosure provides a method for controlling an autonomous vehicle based on an artificial intelligence model. The present method includes the steps of: collecting environmental data by using a data collection vehicle including at least one first sensor; generating an artificial intelligence model for outputting control information for controlling driving of the vehicle by using the environmental data as learning data; and controlling an autonomous vehicle including a second sensor on the basis of the artificial intelligence model.

Description

Method and system for judgment and control of autonomous vehicles based on artificial intelligence models

[0001] The present invention relates to a method and system for judgment and control of an autonomous vehicle based on an artificial intelligence model, and more specifically, to a method and system for collecting data using a data collection vehicle equipped with a LiDAR sensor, generating an artificial intelligence model to predict control information for controlling the driving of a vehicle based on the collected data, and enabling an autonomous vehicle to perform judgment and control without a separate LiDAR sensor based on the generated artificial intelligence model.

[0002] Autonomous driving technology has been the subject of research and development for decades and has made significant technological progress in recent years, but it still faces many challenges in the process of commercialization and popularization. One of the main causes of these difficulties is the bottleneck in decision design. For autonomous driving technology to be realistically introduced, legal, regulatory, and environmental requirements must be established through prototypes; however, conversely, a stalemate arises where decision design can only proceed once these environments are clearly defined.

[0003] Approaches utilizing artificial intelligence models are being actively researched to solve this problem, but the current limitation is that most methods require data that is difficult to obtain in reality.

[0004] To analyze these issues more specifically, the requirements of existing autonomous driving technologies can be divided into data and sensor aspects. In terms of data, existing methods have limitations in that they require data that is difficult to acquire, such as bird's-eye views of the vehicle, while in terms of sensors, there is a limitation in that finished autonomous vehicles necessarily require high-cost sensors such as LIDAR.

[0005] Accordingly, there is a need for a method that enables autonomous vehicles to perform decision-making and control without using high-cost sensors such as LiDAR.

[0006] The present disclosure aims to solve the problems of the aforementioned prior art by collecting data using a data collection vehicle equipped with a lidar sensor, generating an artificial intelligence model to predict control information for controlling the driving of a vehicle based on the collected data, and providing a method and system for an autonomous vehicle to perform judgment and control without a separate lidar sensor based on the generated artificial intelligence model.

[0007] The technical problems that the present invention aims to solve are not limited to the technical problems described above, and other technical problems of the present invention may be derived from the following description.

[0008] As a technical means for solving the aforementioned technical problem, an embodiment according to the first aspect of the present disclosure provides a method for controlling an autonomous vehicle based on an artificial intelligence model. The method comprises the steps of: collecting environmental data using a data collection vehicle comprising at least one first sensor; generating an artificial intelligence model that outputs control information for controlling the driving of a vehicle using the environmental data as training data; and controlling an autonomous vehicle comprising a second sensor based on the artificial intelligence model.

[0009] As a technical means for solving the technical problem described above, an embodiment according to a second aspect of the present disclosure provides a control device for an autonomous vehicle based on an artificial intelligence model. The device comprises a communication module that performs transmission and reception with at least one of a data collection vehicle and an autonomous vehicle, at least one processor, and a memory electrically connected to said processor and storing at least one code executed by said processor. The memory stores a code that, when executed through said processor, causes said processor to generate an artificial intelligence model that collects environmental data using said data collection vehicle including at least one first sensor and outputs control information for controlling the driving of a vehicle using said environmental data as training data, and causes said autonomous vehicle including a second sensor to control said artificial intelligence model.

[0010] According to the present invention, the production cost of autonomous vehicles can be significantly reduced by using LIDAR only in a limited way for data collection vehicles.

[0011] In addition, according to the present invention, stable driving is possible even in complex environments through technologies such as Look-ahead Point, weighting by road type, and cooperative driving.

[0012] In addition, according to the present invention, the flexibility of the technology can be increased through the separate design of the data collection vehicle and the autonomous driving vehicle.

[0013] In addition, according to the present invention, the model can be adapted to changes in road conditions, weather, and obstacles, and can be utilized in various road environments.

[0014] The effects of the present invention are not limited to the effects described above, but include all effects understood from the following description.

[0015] FIG. 1 is a drawing illustrating a server and external devices connected to it in communication according to an embodiment of the present invention.

[0016] Figure 2 is a diagram illustrating the detailed configuration of the server shown in Figure 1.

[0017] FIG. 3 is a diagram illustrating an example of a recognition area of ​​a data collection vehicle and an autonomous driving vehicle using a sensor according to an embodiment of the present invention.

[0018] FIG. 4 is a flowchart illustrating the sequence of a control method for an artificial intelligence model-based autonomous vehicle according to another embodiment of the present invention.

[0019] FIG. 5 is a diagram illustrating an example of a control process for an autonomous driving vehicle based on an artificial intelligence model according to another embodiment of the present invention.

[0020] The present disclosure will be described in detail below with reference to the attached drawings. However, the present disclosure may be implemented in various different forms and is not limited to the embodiments described herein. Furthermore, the attached drawings are intended only to facilitate understanding of the embodiments disclosed herein, and the technical concept disclosed herein is not limited by the attached drawings. All terms used herein, including technical and scientific terms, should be interpreted in the sense generally understood by those skilled in the art to which the present disclosure pertains. Terms defined in advance should be interpreted as having additional meanings consistent with relevant technical literature and the present disclosure, and should not be interpreted in a highly ideal or restrictive sense unless otherwise defined.

[0021] In order to clearly explain the present disclosure in the drawings, parts unrelated to the explanation have been omitted, and the size, form, and shape of each component shown in the drawings may be varied. Throughout the specification, identical or similar parts are denoted by identical or similar reference numerals.

[0022] In describing the embodiments disclosed in this specification, detailed descriptions of related prior art have been omitted where it is determined that such detailed descriptions could obscure the essence of the embodiments disclosed in this specification.

[0023] Throughout the specification, when a part is described as "including (providing or providing)" a certain component, this means that, unless specifically stated otherwise, it does not exclude other components but rather may additionally "include (providing or providing)" other components.

[0024] Terms indicating ordinal numbers, such as first, second, etc., used herein are used solely for the purpose of distinguishing one component from another and do not limit the order or relationship of the components. For example, the first component of the present disclosure may be named the second component, and similarly, the second component may be named the first component. Singular forms used herein should be interpreted to include plural forms unless explicitly to the contrary.

[0025] FIG. 1 is a drawing illustrating a server and external devices connected to it in communication according to an embodiment of the present invention.

[0026] Referring to FIG. 1, an artificial intelligence model-based autonomous driving vehicle control system may include a server (100), a data collection vehicle (200), and an autonomous driving vehicle (300).

[0027] The server (100) can be connected to communicate with at least one of the data collection vehicle (200) and the autonomous vehicle (300) through a pre-configured network.

[0028] The server (100) may be an artificial intelligence model-based autonomous driving vehicle control device.

[0029] A server (100) collects environmental data using a data collection vehicle that includes at least one first sensor. Here, the first sensor may include at least one of a Light Detection and Ranging (LIDAR), a Global Navigation Satellite System (GNSS), and an Inertial Measurement Unit (IMU).

[0030] A server (100) generates an artificial intelligence model that outputs control information for controlling the driving of a vehicle using environmental data as training data. Here, the environmental data may include image data, destination coordinates, navigation commands, LiDAR data, speed data, and acceleration data. The control information may include at least one of the vehicle's speed, acceleration, steering angle, a target point (Look-ahead Point) on the future driving path, and motor control parameters.

[0031] The server (100) controls an autonomous vehicle including a second sensor based on an artificial intelligence model. Here, the second sensor may include at least one of a physical sensor such as a speed sensor and an acceleration sensor, a camera, a Global Navigation Satellite System (GNSS), and an Inertial Measurement Unit (IMU).

[0032] The data collection vehicle (200) can collect high-quality environmental data necessary for training an artificial intelligence model.

[0033] The data collection vehicle (200) can accumulate data in various driving situations by being equipped with a precise sensor compared to the autonomous driving vehicle (300). Here, the precise sensor may be a LiDAR sensor, and the driving situation may be environmental information and driving records.

[0034] The data collection vehicle (200) may include at least one of a LIDAR, a camera, a Global Navigation Satellite System (GNSS), and an Inertial Measurement Unit (IMU).

[0035] The data collection vehicle (200) can record all sensor data generated while driving and store it as a dataset that can be used to train an artificial intelligence model.

[0036] The data collection vehicle (200) can record actions performed by the actual driver of the vehicle and use them as imitation learning data. Here, the actions performed by the actual driver of the vehicle may include at least one of speed, acceleration, and steering angle.

[0037] The autonomous vehicle (300) can be equipped with a generated artificial intelligence model.

[0038] The autonomous vehicle (300) can recognize the driving environment and perform judgment and control using an artificial intelligence model based on data collected from the data collection vehicle (200).

[0039] The autonomous vehicle (300) can analyze the environment in real time and perform various autonomous driving tasks such as path planning, obstacle avoidance, and lane change.

[0040] The autonomous vehicle (300) may include at least one of a camera, a speed sensor, an acceleration sensor, a Global Navigation Satellite System (GNSS), and an Inertial Measurement Unit (IMU).

[0041] The autonomous vehicle (300) may include a controller that adjusts the acceleration, deceleration, and steering of the vehicle in real time. Here, the controller may be at least one of PID (Proportional-Integral-Derivative) and MPC (Model Predictive Control).

[0042] Figure 2 is a diagram illustrating the detailed configuration of the server shown in Figure 1.

[0043] Referring to FIG. 2, the server (100) may include a communication module (110), a processor (120), and a memory (130).

[0044] The communication module (110) can receive environmental data from a data collection vehicle. In addition, the communication module (110) can receive current driving data from an autonomous vehicle and transmit an artificial intelligence model and a control signal generated based on the artificial intelligence model to the autonomous vehicle.

[0045] The communication module (110) may include a device comprising hardware and software necessary to transmit and receive signals, such as control signals or data signals, through a wired or wireless connection with another network device.

[0046] The processor (120) may include various types of devices for controlling and processing data. The processor (120) may refer to a data processing device embedded in hardware having a physically structured circuit to perform a function expressed by code or instructions included in a program.

[0047] In one example, the processor (120) may be implemented in the form of a microprocessor, a central processing unit (CPU), a processor core, a multiprocessor, an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), etc., but the scope of the invention is not limited thereto.

[0048] The processor (120) performs operations according to the code stored in memory (130).

[0049] The memory (130) can store at least one of the information and data input to the communication module (110), the information and data required for the function performed by the processor (120), and the data generated according to the execution of the processor (120).

[0050] Memory (130) should be interpreted as a general term for non-volatile storage devices that retain stored information even when power is not supplied, and volatile storage devices that require power to retain stored information. In addition to volatile storage devices that require power to retain stored information, memory (130) may include cloud storage, SSD, magnetic storage media, or flash storage media, but the scope of the present invention is not limited thereto.

[0051] Memory (130) is electrically connected to the processor (120) and stores at least one code executed by the processor (120). Memory (130) stores code that causes the processor (120) to perform the following functions and procedures when executed through the processor (120).

[0052] A memory (130) stores code that causes environmental data to be collected using a data collection vehicle that includes at least one first sensor. Here, the first sensor may include at least one of a Light Detection and Ranging (LIDAR), a Global Navigation Satellite System (GNSS), and an Inertial Measurement Unit (IMU).

[0053] Environmental data is driving data of a data collection vehicle and may include image data captured through a camera, destination coordinates, navigation commands, LiDAR data, speed data, and acceleration data. However, it is not limited thereto, and environmental data may include at least one of the coordinates, heading, speed, target speed, target coordinates, commands, a list of next destination coordinates, and previous control values ​​of the data collection vehicle. Here, the target coordinates may be at the intersection level, and the previous control values ​​of the data collection vehicle may be at least one of throttle, steer, and brake.

[0054] In the memory (130), code is stored that causes an artificial intelligence model to be generated to output control information for controlling the driving of a vehicle using environmental data as training data. Here, the control information may include at least one of the vehicle's speed, acceleration, steering angle, target point on the future driving path, and motor control parameters. The training data may include actual driving data and, at the same time, additionally include virtual data generated in a simulation environment.

[0055] The memory (130) may store code that causes the LiDAR data to be converted into a two-dimensional plane.

[0056] The memory (130) may store code that causes the generated training data by combining converted LiDAR data, velocity data, and acceleration data.

[0057] Memory (130) may store code that causes the optimization of training data by removing duplicate or unnecessary data from environment data and filtering necessary data.

[0058] For example, in cases where an obstacle appears repeatedly at a specific location in the LIDAR data and part of the camera image is of low quality due to light reflection, the memory (130) may store code that causes the same obstacle to be recorded multiple times in the collected data to be removed, the low-quality camera image to be filtered, the data unsuitable for deep learning to be excluded, and the remaining data to be cleaned up to perform at least one of labeling and normalization to generate optimal training data.

[0059] The memory (130) may store code that causes an artificial intelligence model to be trained to output control values ​​of an autonomous vehicle to improve the control accuracy of the autonomous vehicle by applying different weights according to a preset road type. Here, the preset road type may include at least one of a highway, an urban area, and an intersection.

[0060] For example, when an autonomous vehicle alternately drives on a highway and a city road, the memory (130) may store code that causes the vehicle to drive on the highway with priority weights such as maintaining distance between vehicles and centering the lane, and to drive on the city road with high priority weights such as traffic light detection, intersection detection, and pedestrian detection.

[0061] The memory (130) may store code that causes an artificial intelligence model to be trained to predict the behavior of surrounding vehicles by receiving environmental data.

[0062] The memory (130) stores code that causes an autonomous vehicle including a second sensor to be controlled based on an artificial intelligence model. Here, the second sensor may include at least one of a camera, a speed sensor, an acceleration sensor, a Global Navigation Satellite System (GNSS), and an Inertial Measurement Unit (IMU).

[0063] The memory (130) may store code that causes the autonomous vehicle to predict the surrounding environment and set the driving path based on a target point on the future driving path. For example, if the autonomous vehicle is driving on a curved road and a sharp curve appears ahead, the memory (130) may store code that causes the vehicle to set a target point on the future driving path, analyze the lane curvature and obstacles at the target point, calculate the optimal speed based on the analysis results, and pre-adjust the steering angle.

[0064] The memory (130) may store code that compares the current driving data of the autonomous vehicle with a preset target state and causes the autonomous vehicle to be controlled so that the current driving data of the autonomous vehicle reaches the preset target state.

[0065] The memory (130) may store code that causes the autonomous vehicle to share data in real time with other autonomous vehicles or robots and to perform cooperative judgment and control.

[0066] For example, when multiple autonomous trucks perform platooning on a highway, the memory (130) may store code that receives data on its position, speed, acceleration, and traffic conditions from the lead vehicle and provides it to the subsequent vehicle in real time, causing the subsequent vehicle to maintain an accurate distance and perform energy-efficient driving based on the data from the lead vehicle.

[0067] The memory (130) may store code that causes the artificial intelligence model to be updated by continuously learning data accumulated during the operation of the autonomous vehicle.

[0068] FIG. 3 is a diagram illustrating an example of a recognition area of ​​a data collection vehicle and an autonomous vehicle using a sensor according to an embodiment of the present invention. More specifically, FIG. 3(a) may be a recognition area of ​​a data collection vehicle including a lidar sensor and a camera, and FIG. 3(b) may be a diagram illustrating an example of a recognition area of ​​an autonomous vehicle including a camera.

[0069] Referring to FIG. 3, the data collection vehicle can collect environmental data of the data collection vehicle using both a lidar sensor and a camera.

[0070] The data collection vehicle can detect the full 360-degree range around the vehicle through a LiDAR sensor. A portion of the LiDAR area detected by the LiDAR sensor is superimposed with the camera area detected by the camera, allowing for more detailed detection of the surroundings of the data collection vehicle.

[0071] Autonomous vehicles can collect driving data using cameras.

[0072] Autonomous vehicles can use cameras to detect the entire 360-degree range around the autonomous vehicle. Cameras can detect the surrounding environment as visual image data.

[0073] FIG. 4 is a flowchart illustrating the sequence of a control method for an artificial intelligence model-based autonomous vehicle according to another embodiment of the present invention.

[0074] The control method of an autonomous vehicle based on an artificial intelligence model described below can be performed by the control device or server (100) of the autonomous vehicle based on an artificial intelligence model described above with reference to FIGS. 1 to 3. Accordingly, the content of the embodiment of the present disclosure described above with reference to FIGS. 1 to 3 can be applied in the same way to the embodiment described below, and content that overlaps with the description above will be omitted. The steps described below do not necessarily have to be performed in order, the order of the steps can be set in various ways, and the steps may be performed almost simultaneously.

[0075] Referring to FIG. 4, the control method of an autonomous vehicle based on an artificial intelligence model includes an environment data collection step (S100), an artificial intelligence model creation step (S200), and an autonomous vehicle control step (S300).

[0076] The environmental data collection step (S100) is a step of collecting environmental data using a data collection vehicle comprising at least one first sensor. Here, the first sensor includes at least one of a LIDAR (Light Detection and Ranging), a GNSS (Global Navigation Satellite System), and an IMU (Inertial Measurement Unit), and the environmental data may include LIDAR data, velocity data, and acceleration data.

[0077] The artificial intelligence model generation step (S200) is a step of generating an artificial intelligence model that outputs control information for controlling the driving of a vehicle using environmental data as training data. Here, the control information may be at least one of the vehicle's speed, acceleration, steering angle, and motor control parameters.

[0078] The artificial intelligence model generation step (S200) may include a step of converting LiDAR data into a two-dimensional plane and a step of generating training data by combining the converted LiDAR data, velocity data, and acceleration data.

[0079] The artificial intelligence model generation step (S200) may include a step of optimizing training data by removing duplicate or unnecessary data from environmental data and filtering necessary data.

[0080] The artificial intelligence model generation step (S200) may include the step of generating an artificial intelligence model trained to improve the control accuracy of an autonomous vehicle by applying different weights according to a preset road type.

[0081] The autonomous driving vehicle control step (S300) is a step of controlling an autonomous driving vehicle including a second sensor based on an artificial intelligence model. Here, the second sensor may include at least one of a camera, a speed sensor, an acceleration sensor, a Global Navigation Satellite System (GNSS), and an Inertial Measurement Unit (IMU).

[0082] The autonomous vehicle control step (S300) may include a step of comparing the current state of the autonomous vehicle with a preset target state and controlling the autonomous vehicle so that the current state of the autonomous vehicle reaches the preset target state.

[0083] FIG. 5 is a diagram illustrating an example of a control process for an autonomous driving vehicle based on an artificial intelligence model according to another embodiment of the present invention.

[0084] Referring to FIG. 5, the control process of an artificial intelligence model-based autonomous vehicle may include a process for determining the type of data to be used for input and output (S410), a process for configuring output data acquisition sensors (S421), a process for designing data collection vehicle sensors (S422), a process for vehicle operation and data collection (S423), a process for learning a judgment / control model based on data (S424), a process for configuring input data acquisition sensors (S431), a process for designing autonomous vehicle sensors (S432), and a process for testing an autonomous vehicle (S440).

[0085] The process of determining the types of data to be used for input and output (S410) may be an initial step for determining the types of data required for model training, judgment, and control.

[0086] The process of determining the types of data to be used for input and output (S410) can determine which input data the autonomous vehicle will use to recognize the environment and perform a judgment, and can determine the output data that the autonomous vehicle must predict for judgment and control.

[0087] The output data acquisition sensor configuration process (S421) is a step of configuring a sensor to collect output data, and may be a process of designing a sensor so that the autonomous vehicle can accurately acquire the necessary information.

[0088] For example, the output example may include at least one of LIDAR z-axis projection, target speed, steering angle, target acceleration, and throttle control parameter.

[0089] The data collection vehicle sensor design process (S422) is a process of designing and installing sensors required for a data collection vehicle, and can perform the role of recording environmental data and driving data.

[0090] The data collection vehicle sensor design process (S422) can be performed based on the output data acquisition sensor configuration process (S421) and the input data acquisition sensor configuration process (S431).

[0091] The vehicle operation and data collection process (S423) may be a process of driving a data collection vehicle and actually collecting environmental data necessary for learning and model training.

[0092] The process of learning a judgment / control model with data (S424) may be a process of developing judgment and control functions of an autonomous vehicle by training an artificial intelligence model based on collected data.

[0093] The input data acquisition sensor configuration process (S431) may be a process of configuring a sensor for an autonomous vehicle to collect input data in real time.

[0094] For example, the input example may include at least one of front / rear / side cameras, speed status, coordinate status, destination coordinates, and navigation commands.

[0095] The autonomous vehicle sensor design process (S432) may be a process of designing a simplified sensor for the autonomous vehicle to perform judgment and control.

[0096] The autonomous vehicle test process (S440) may be a process of verifying the performance of an autonomous vehicle by testing it based on a learned model and designed sensors.

[0097] A person skilled in the art to which this disclosure pertains will understand that, based on the foregoing description, other specific forms can be easily modified without altering the technical spirit or essential features of this disclosure. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive. The scope of this disclosure is defined by the claims set forth below, and all modifications or variations derived from the meaning and scope of the claims and equivalents thereof should be interpreted as being included within the scope of this disclosure. The scope of this application is defined by the claims set forth below rather than by the foregoing detailed description, and all modifications or variations derived from the meaning and scope of the claims and equivalents thereof should be interpreted as being included within the scope of this application.

[0098] The form for carrying out the invention is substantially the same as the best form for carrying out the invention mentioned above.

[0099] The present invention has industrial applicability as it can be utilized in products to which autonomous driving is applied.

Claims

1. A method for controlling an autonomous vehicle based on an artificial intelligence model performed by a server, a) collecting environmental data using a data collection vehicle comprising at least one first sensor; b) a step of generating an artificial intelligence model that outputs a control value for controlling the driving of a vehicle using the above-mentioned environment data as training data; and c) A method for controlling an autonomous vehicle based on an artificial intelligence model, comprising the step of controlling the autonomous vehicle including a second sensor based on the artificial intelligence model.

2. In Paragraph 1, The first sensor above includes at least one of a Light Detection and Ranging (LIDAR), a Global Navigation Satellite System (GNSS), and an Inertial Measurement Unit (IMU). A control method for an autonomous vehicle based on an artificial intelligence model, wherein the second sensor comprises at least one of a shooting device, a speed sensor, an acceleration sensor, a Global Navigation Satellite System (GNSS), and an Inertial Measurement Unit (IMU).

3. In Paragraph 1, The above environmental data includes image data, destination coordinates, navigation commands, LiDAR data, speed data, and acceleration data, and The above step b) is, A step of converting the above LiDAR data into a two-dimensional plane; and A control method for an artificial intelligence model-based autonomous vehicle, comprising the step of generating training data by combining the converted lidar data, the velocity data, and the acceleration data.

4. In Paragraph 1, The above step b) is, A method for controlling an autonomous vehicle based on an artificial intelligence model, comprising the step of optimizing the training data by removing duplicate or unnecessary data from the environment data and filtering necessary data.

5. In Paragraph 1, The above artificial intelligence model is, A control method for an autonomous vehicle based on an artificial intelligence model, which is trained to improve the control accuracy of the autonomous vehicle by applying different weights according to a preset road type.

6. In Paragraph 1, A method for controlling an autonomous vehicle based on an artificial intelligence model, wherein the above control information is at least one of the vehicle's speed, acceleration, steering angle, and motor control parameters.

7. In Paragraph 1, The above step c) is, A method for controlling an autonomous vehicle based on an artificial intelligence model, comprising the step of comparing current driving data of the autonomous vehicle with a preset target state and controlling the autonomous vehicle so that the current driving data of the autonomous vehicle reaches the preset target state.

8. A communication module that performs transmission and reception with at least one of a data collection vehicle and an autonomous driving vehicle; At least one processor; and It includes a memory electrically connected to the processor and storing at least one code executed in the processor, When the above memory is executed through the above processor, the processor, A control device for an AI model-based autonomous vehicle, which collects environmental data using a data collection vehicle including at least one first sensor, generates an AI model that outputs control information for controlling the driving of a vehicle using the environmental data as training data, and stores code that causes the autonomous vehicle including a second sensor to be controlled based on the AI ​​model.

9. In Paragraph 8, The first sensor above includes at least one of a Light Detection and Ranging (LIDAR), a Global Navigation Satellite System (GNSS), and an Inertial Measurement Unit (IMU). A control device for an artificial intelligence model-based autonomous vehicle, wherein the second sensor comprises at least one of a shooting device, a speed sensor, an acceleration sensor, a Global Navigation Satellite System (GNSS), and an Inertial Measurement Unit (IMU).

10. In Paragraph 8, The above environmental data includes image data, destination coordinates, navigation commands, LiDAR data, speed data, and acceleration data, and The above memory allows the processor, A control device for an artificial intelligence model-based autonomous vehicle, which stores a code that converts the above LiDAR data into a two-dimensional plane and causes the generated training data by combining the converted LiDAR data, the velocity data, and the acceleration data.

11. In Paragraph 8, The above memory allows the processor, A control device for an artificial intelligence model-based autonomous vehicle that stores code causing the optimization of the training data by removing duplicate or unnecessary data from the environment data and filtering necessary data.

12. In Paragraph 8, The above artificial intelligence model is, A control device for an autonomous vehicle based on an artificial intelligence model, which is trained to improve the control accuracy of the autonomous vehicle by applying different weights according to a preset road type.

13. In Paragraph 8, A control device for an artificial intelligence model-based autonomous vehicle, wherein the above control information is at least one of the vehicle's speed, acceleration, steering angle, and motor control parameters.

14. In Paragraph 8, The above memory allows the processor, A control device for an artificial intelligence model-based autonomous vehicle, comprising the step of comparing current driving data of the autonomous vehicle with a preset target state and controlling the autonomous vehicle so that the current driving data of the autonomous vehicle reaches the preset target state.

Images