Route generation device and method for generating waypoints for autonomous driving

Abstract

A path generation device and method for generating waypoints for autonomous driving are disclosed. A path generation method performed by a path generation device for generating a path for an autonomous driving device includes the steps of: inputting at least one of GPS information serving as a reference for generating a path and 3D image information and information related to the driving mode of the autonomous driving device; selecting a generation model corresponding to the driving mode among a plurality of previously learned generation models; generating a plurality of waypoints suitable for the driving mode by applying GPS information and 3D image information to the selected generation model and calculating a driving speed based on the generated plurality of waypoints; and generating a path based on the plurality of waypoints and the driving speed.

Description

Technology Field

[0001] The present invention relates to a path generation device, and more specifically, to a path generation device and method for generating waypoints for autonomous driving that improve work efficiency by optimizing the path and driving speed of an autonomous driving device in an agricultural environment. Background Technology

[0002] Modern agriculture is a complex industry that combines primary, secondary, and tertiary industries. Moving away from agriculture focused on simple food production, there is a demand for development into smart agriculture, bio-agriculture, eco-friendly agriculture, cultural agriculture, tourism agriculture, and space agriculture through convergence with advanced technologies such as IT (Information Technology), BT (Bio Technology), NT (Nano Technology), ET (Environment Technology), CT (Culture Technology), and ST (Space Technology).

[0003] Meanwhile, smart farms have recently been receiving significant attention as a solution to rural sustainability crises, such as stagnation in income, exports, and growth rates resulting from the aging population in agriculture and livestock farming, labor shortages for young successors to farms, a decrease in production area, and reduced investment. In particular, these smart farms require autonomous vehicles capable of eco-friendly mobility, and due to this need, research on autonomous vehicles is currently being actively conducted. Prior art literature

[0004] Korean Patent Publication No. 10-2405667 (May 31, 2022) The problem to be solved

[0005] The problem that the present invention aims to solve is to provide a path generation device and method for generating waypoints for autonomous driving that maximize work efficiency and stability by optimizing the path and driving speed of the autonomous driving device in various agricultural environments. means of solving the problem

[0006] A path generation method performed by a path generation device for generating a path of an autonomous driving device according to the present invention to solve the above problem comprises: a step of inputting at least one of a GPS (Global Positioning System) information serving as a reference for generating the path, 3D image information, and information related to the driving mode of the autonomous driving device; a step of selecting a generation model corresponding to the driving mode among a plurality of previously learned generation models; a step of generating a plurality of waypoints suitable for the driving mode by applying the GPS information and the 3D image information to the selected generation model, and calculating a driving speed based on the generated plurality of waypoints; and a step of generating the path based on the plurality of waypoints and the driving speed.

[0007] In addition, prior to the input step, the method further includes a step of learning the plurality of generation models, wherein the plurality of generation models include at least one of: a first generation model corresponding to the basic mode among the driving modes and generating a path including waypoints with uniform spacing based on the GPS information; a second generation model corresponding to the driving mode among the driving modes and generating a path that minimizes the time taken to reach a target position from a starting position and deviation from a reference line; a third generation model corresponding to the pest control mode among the driving modes and generating a path according to the density of trees included in the three-dimensional image information; and a fourth generation model corresponding to the harvesting mode among the driving modes and generating a path according to the density of harvested crops included in the three-dimensional image information.

[0008] In addition, the learning step is characterized by performing reinforcement learning according to the characteristics of each of the plurality of generative models.

[0009] In addition, the learning step described above is characterized by training the third generation model so that waypoints are generated more densely in sections with a large number of trees.

[0010] In addition, the learning step is characterized by training the fourth generation model so that waypoints are generated more densely in sections with a larger number of harvested items.

[0011] In addition, the learning step is characterized by training the plurality of generation models such that the driving speed decreases as the distance between neighboring waypoints among the plurality of waypoints narrows.

[0012] In addition, if the above driving mode related information is not input, the selecting step is characterized by predicting current task related status information based on the GPS information and the three-dimensional image information, and selecting a generating model corresponding to the status information among the plurality of generating models.

[0013] A path generation device for generating a path of an autonomous driving device according to the present invention comprises a waypoint generation unit that, when at least one of GPS information serving as a reference for generating the path and 3D image information and information related to the driving mode of the autonomous driving device is input, selects a generation model corresponding to the driving mode among a plurality of previously learned generation models, applies the GPS information and the 3D image information to the selected generation model to generate a plurality of waypoints suitable for the driving mode, and calculates a driving speed based on the generated plurality of waypoints, and a path generation unit that generates the path based on the plurality of waypoints and the driving speed. Effects of the invention

[0014] According to an embodiment of the present invention, a driving mode suitable for the current situation and conditions is selected, customized waypoints are generated based on the selected driving mode, and a driving speed corresponding to the generated waypoints is calculated, and then a path of the autonomous driving device can be generated.

[0015] This ensures both operational efficiency and safety, and allows for stability by driving at appropriate speeds in work sections, corner sections, and straight sections, respectively. Brief explanation of the drawing

[0016] FIG. 1 is a configuration diagram for explaining a path generation system according to an embodiment of the present invention. FIG. 2 is a block diagram illustrating a path generation device according to an embodiment of the present invention. FIG. 3 is a diagram illustrating a path generation process according to an embodiment of the present invention. FIG. 4 is a block diagram for explaining a control unit according to an embodiment of the present invention. Figure 5 is a block diagram for explaining a plurality of generation models according to an embodiment of the present invention. FIG. 6 is a diagram illustrating waypoint generation and path generation based on a basic mode according to an embodiment of the present invention. FIG. 7 is a diagram illustrating waypoint generation and path generation based on a driving mode according to an embodiment of the present invention. FIG. 8 is a diagram illustrating waypoint generation and path generation based on a control mode according to an embodiment of the present invention. FIG. 9 is a diagram illustrating waypoint generation and path generation based on a harvesting mode according to an embodiment of the present invention. FIG. 10 is a flowchart illustrating a path generation method according to an embodiment of the present invention. FIG. 11 is a block diagram illustrating a computing device according to an embodiment of the present invention. Specific details for implementing the invention

[0017] Embodiments of the present invention are described below with reference to the attached drawings so that those skilled in the art can easily implement them. However, the present invention may be embodied in various different forms and is not limited to the embodiments described herein. Furthermore, in order to clearly explain the present invention in the drawings, parts unrelated to the explanation have been omitted, and similar parts throughout the specification are denoted by similar reference numerals.

[0018] In this specification and drawings (hereinafter referred to as the 'this specification'), redundant descriptions of identical components are omitted.

[0019] Furthermore, when a component is described in this specification as being 'connected' or 'connected' to another component, it should be understood that it may be directly connected to or connected to the other component, or that there may be other components in between. On the other hand, when a component is described in this specification as being 'directly connected' or 'directly connected' to another component, it should be understood that there are no other components in between.

[0020] Furthermore, the terms used in this specification are used merely to describe specific embodiments and are not intended to limit the invention.

[0021] Additionally, in this specification, singular expressions may include plural expressions unless the context clearly indicates otherwise.

[0022] Furthermore, in this specification, terms such as 'comprising' or 'having' are intended merely to indicate the existence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not excluding in advance the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0023] Additionally, in this specification, the term "and / or" includes a combination of the plurality of described items or any of the plurality of described items. In this specification, "A or B" may include "A," "B," or "both A and B."

[0024] In addition, detailed descriptions of known functions and configurations that may obscure the essence of the invention will be omitted in this specification.

[0026] FIG. 1 is a configuration diagram for explaining a path generation system according to an embodiment of the present invention.

[0027] Referring to FIG. 1, the path generation system (400) optimizes the path and driving speed of an autonomous driving device in various agricultural environments to maximize work efficiency and stability. The path generation system (400) includes a path generation device (100), an autonomous driving device (200), and a user terminal (300).

[0028] The path generation device (100) generates an optimal path suitable for the work state performed by the autonomous driving device (200). At this time, the path includes information on the driving speed for each section. The path generation device (100) can generate a path using Global Positioning System (GPS) information and 3D image information. Here, the GPS information and 3D image information may be information collected by a user while driving directly at least once in the area where the path is to be generated, and the 3D image information may include Lidar information and image information.

[0029] The autonomous driving device (200) receives a path generated by the path generation device (100), drives autonomously according to the received path, and performs tasks when necessary. While performing autonomous driving and tasks, the autonomous driving device (200) collects information such as movement location, movement distance, movement speed, and task status, and can generate monitoring information based on the collected information. The autonomous driving device (200) transmits the monitoring information to a user terminal (300). The autonomous driving device (200) may be an autonomous driving vehicle or an autonomous driving robot. Preferably, the autonomous driving device (200) may include a module capable of performing farming tasks (e.g., a pest control machine, a robotic arm, etc.).

[0030] The user terminal (300) is a terminal used by a user (or administrator) and communicates with the path generation device (100) and the autonomous driving device (200). The user terminal (300) transmits at least one of GPS information, 3D image information, and driving mode-related information to the path generation device (100) to support the creation of a path in the area related to the received information by the path generation device (100). Additionally, the user terminal (300) receives monitoring information from the autonomous driving device (200) and outputs the received monitoring information to help the user check the content remotely.

[0031] Meanwhile, the path generation system (400) supports communication between the path generation device (100), the autonomous driving device (200), and the user terminal (300) by establishing a communication network (450). The communication network (450) may be composed of a backbone network and a subscriber network. The backbone network may be composed of one or more integrated networks among an X.25 network, a Frame Relay network, an ATM network, an MPLS (Multi-Protocol Label Switching) network, and a GMPLS (Generalized Multi-Protocol Label Switching) network. The subscriber network may be FTTH (Fiber To The Home), ADSL (Asymmetric Digital Subscriber Line), cable network, Zigbee, Bluetooth, Wireless LAN (IEEE 802.11b, IEEE 802.11a, IEEE 802.11g, IEEE 802.11n), Wireless Hart (ISO / IEC 62591-1), ISA 100.11a (ISO / IEC 62734), CoAP (Constrained Application Protocol), MQTT (Message Queuing Telemetry Transport), WIBro (Wireless Broadband), WiMAX, 3G, HSDPA (High Speed ​​Downlink Packet Access), 4G, 5G, and 6G, etc. In some embodiments, the communication network (450) may be an internet network and a mobile communication network. Additionally, the communication network (450) may include any other widely known or future-developed wireless or wired communication methods.

[0032] Additionally, although the path generation device (100) and the autonomous driving device (200) are depicted as separate devices in the drawing, the path generation device (100) may be implemented in a form that includes the autonomous driving device (200) depending on the situation in which the system is implemented.

[0034] FIG. 2 is a block diagram for explaining a path generation device according to an embodiment of the present invention, and FIG. 3 is a diagram for explaining a path generation process according to an embodiment of the present invention.

[0035] Referring to FIGS. 1 to 3, the path generating device (100) includes a communication unit (10), an input unit (30), a control unit (50), an output unit (70), and a storage unit (90).

[0036] The communication unit (10) communicates with the autonomous driving device (200) and the user terminal (300). The communication unit (10) transmits generated path-related information to the autonomous driving device (200) and can receive at least one of GPS information, 3D image information, and driving mode-related information from the user terminal (300) (Fig. 3 (b)). The communication unit (10) can transmit and receive information with each device through the communication network (450), and if the path generation device (100) is included inside the autonomous driving device (200), information exchange can be performed via an internal line.

[0037] The input unit (30) receives user input related to route generation directly from the user. When the user determines the area where the route will be generated (Fig. 3 (a)), the input unit (30) may receive at least one of GPS information, 3D image information, and driving mode information for the area (Fig. 3 (b)).

[0038] The control unit (50) performs overall control of the path generation device (100). When at least one of the GPS information serving as a reference for generating a path and 3D image information and information related to the driving mode of the autonomous driving device (200) is input, the control unit (50) selects a generation model corresponding to the driving mode among a plurality of previously learned generation models. Here, the driving mode may include a basic mode for general driving, a driving mode for driving purposes, a pest control mode for pest control purposes, and a harvesting mode for harvesting purposes. Additionally, the plurality of generation models may include at least one of a first generation model corresponding to the basic mode among the driving modes, a second generation model corresponding to the driving mode, a third generation model corresponding to the pest control mode, and a fourth generation model corresponding to the harvesting mode. The control unit (50) applies the GPS information and 3D image information to the selected generation model to generate a plurality of waypoints suitable for the corresponding driving mode, and calculates the driving speed based on the generated plurality of waypoints (Fig. 3 (c)). The control unit (50) generates a path based on a plurality of waypoints and driving speeds (Fig. 3 (d)).

[0039] The output unit (70) outputs information that is received, input, generated, or produced during the process of generating a path. For example, the output unit (70) can output GPS information, 3D image information, multiple waypoints, paths, etc. The output unit (70) may be a display, and the display may include at least one of a liquid crystal display (LCD), a thin film transistor-liquid crystal display (TFT LCD), an organic light-emitting diode (OLED), a flexible display, and a 3D display.

[0040] The storage unit (90) stores a program or algorithm for operating the path generation device (100). The storage unit (90) stores information that is received, input, generated, or produced during the process of generating a path. For example, the storage unit (90) may store GPS information, 3D image information, information related to driving modes, multiple waypoints, paths, etc. Additionally, the storage unit (90) stores multiple generation models to which artificial intelligence technology is applied. The storage unit (90) may include at least one storage medium among a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (e.g., SD or XD memory, etc.), RAM (Random Access Memory), SRAM (Static Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), PROM (Programmable Read-Only Memory), magnetic memory, a magnetic disk, and an optical disk.

[0042] FIG. 4 is a block diagram for explaining a control unit according to an embodiment of the present invention, FIG. 5 is a block diagram for explaining a plurality of generation models according to an embodiment of the present invention, FIG. 6 is a diagram for explaining waypoint generation and path generation based on a basic mode according to an embodiment of the present invention, FIG. 7 is a diagram for explaining waypoint generation and path generation based on a driving mode according to an embodiment of the present invention, FIG. 8 is a diagram for explaining waypoint generation and path generation based on a pest control mode according to an embodiment of the present invention, and FIG. 9 is a diagram for explaining waypoint generation and path generation based on a harvesting mode according to an embodiment of the present invention.

[0043] Referring to FIGS. 1 to 9, the control unit (50) includes a learning unit (51), a waypoint generation unit (53), and a path generation unit (55).

[0044] The learning unit (51) learns a plurality of generative models used to generate paths. Here, each of the plurality of generative models is a model to which artificial intelligence technology is applied, and includes a neural network, preferably a Convolutional Neural Network (CNN), but is not limited thereto. The plurality of generative models includes a first generative model (61), a second generative model (63), a third generative model (65), and a fourth generative model (67).

[0045] The first generation model (61) corresponds to the basic mode among the driving modes of the autonomous driving device (200) and generates a path containing waypoints at uniform intervals based on GPS information. That is, the first generation model (61) uses GPS information to generate efficient waypoints and adjusts the speed.

[0046] The second generation model (63) corresponds to the driving mode among the driving modes of the autonomous driving device (200) and generates a path that minimizes the time taken to reach a target position from a starting position and the deviation from the reference line. That is, the second generation model (63) generates many waypoints in corner sections and generates a minimum number of waypoints in straight sections to ensure optimal driving.

[0047] The third generation model (65) corresponds to the pest control mode among the driving modes of the autonomous driving device (200) and generates a path according to the density of trees included in the three-dimensional image information. That is, the third generation model (65) generates a path so that it stays for a long time in sections with many trees and passes quickly in sections without trees.

[0048] The fourth generation model (67) corresponds to the harvesting mode among the driving modes of the autonomous driving device (200) and generates a path according to the density of harvested items included in the three-dimensional image information. That is, the fourth generation model (67) generates a path such that it stays for a long time in sections with many harvested items and passes quickly in sections without harvested items. Harvested items include fruit trees, flowers, vegetables, crops, etc.

[0049] The learning unit (51) can perform reinforcement learning according to the characteristics of each of the first generation model (61) to the fourth generation model (67). Through this, the learning unit (51) supports the generation of an optimal path suitable for the driving mode of the autonomous driving device (200).

[0050] For example, when the learning unit (51) learns the second generative model (63), it can perform reinforcement learning using a reward function that minimizes the time taken to reach the starting point and the end point and minimizes deviation from the reference line. When the learning unit (51) learns the third generative model (65), it can perform reinforcement learning using a reward function that adjusts the density and speed of waypoints according to the density of trees recognized in the three-dimensional image information. Additionally, when the learning unit (51) learns the fourth generative model (67), it can perform reinforcement learning using a reward function that adjusts the density and speed of waypoints according to the density of harvested objects recognized in the three-dimensional image information.

[0051] Here, the learning unit (51) can train each of the multiple generation models such that the driving speed decreases as the distance between neighboring waypoints among the multiple waypoints becomes narrower. That is, the learning unit (51) can train the model to reduce the driving speed as the distance between neighboring waypoints becomes narrower.

[0052] When at least one of the GPS information, 3D image information, and driving mode information of the autonomous driving device (200) that serves as a reference for generating a route in a region desired by the user is input, the waypoint generation unit (53) selects a generation model corresponding to the driving mode among a plurality of generation models learned from the learning unit (51). When driving mode information is input, the waypoint generation unit (53) can select a generation model corresponding to the driving mode included in the driving mode information. In addition, when driving mode information is not input, the waypoint generation unit (53) can predict current operation-related status information based on GPS information and 3D image information and select a generation model corresponding to the predicted status information among a plurality of generation models. For example, the waypoint generation unit (53) can use GPS information and 3D image information to determine whether the region where a route is to be generated currently requires monitoring, pest control, or harvesting, and select a generation model according to the determined result. That is, the waypoint generation unit (53) can select a second generation model (63) corresponding to the driving mode if it determines that monitoring is necessary, select a third generation model (65) corresponding to the pest control mode if it determines that pest control is necessary, and select a fourth generation model (67) corresponding to the harvesting mode if it determines that harvesting is necessary. Additionally, the waypoint generation unit (53) can select a first generation module (61) corresponding to the basic mode if it determines that monitoring, pest control, and harvesting are not all necessary.

[0053] The waypoint generation unit (53) generates multiple waypoints suitable for the driving mode by applying GPS information and 3D image information input into the selected generation model, and calculates the driving speed based on the generated multiple waypoints. For example, when the waypoint generation unit (53) uses the first generation model (61), it can generate waypoints (W) at uniform intervals in all sections passing through the tree (T) and calculate the driving speed for all sections equally (Fig. 6(a)). When the waypoint generation unit (53) uses the second generation model (63), it can generate many waypoints (W) in corner sections and generate a minimum number of waypoints (W) in straight sections among the sections passing through the tree (T), and calculate the driving speed based on the spacing between neighboring waypoints (Fig. 7(a)). When the waypoint generation unit (55) uses the third generation model (65), it generates more waypoints (W) in sections where trees (T) are planted more densely, generates fewer waypoints (W) in sections where trees (T) are planted less densely, and can calculate the driving speed based on the spacing between neighboring waypoints (Fig. 8(a)). When the waypoint generation unit (55) uses the fourth generation model (65), it generates more waypoints (W) in sections where there are more harvested fruits (H) on the trees (T), generates fewer waypoints (W) in sections where there are fewer harvested fruits (H), and can calculate the driving speed based on the spacing between neighboring waypoints (Fig. 9(a)).

[0054] The path generation unit (55) generates a path (P) based on a plurality of waypoints generated from the waypoint generation unit (55) and the calculated driving speed. The path generation unit (55) can generate a path (P) by connecting adjacent waypoints into one and matching the driving speed for each connected section (Fig. 6 (b) to Fig. 9 (b)). The path generation unit (55) transmits the generated path-related information to the autonomous driving device (200) to help the autonomous driving device (200) perform optimized driving.

[0056] FIG. 10 is a flowchart illustrating a path generation method according to an embodiment of the present invention.

[0057] Referring to FIGS. 1 and FIGS. 10, the path generation method selects a driving mode suitable for the current situation and conditions and generates customized waypoints based on the selected driving mode. The path generation method can generate a path for the autonomous driving device (200) after calculating a driving speed corresponding to the generated waypoints. Through this, the path generation method can simultaneously ensure the efficiency and safety of the operation and secure stability by driving at an appropriate driving speed in the work section, corner section, and straight section, respectively.

[0058] In step S101, the path generation device (100) initializes the system. The path generation device (100) initializes previously recorded or stored information.

[0059] In step S103, the path generation device (100) receives GPS information and 3D image information. The path generation device (100) may receive GPS information and 3D image information of the area where a path is to be generated. Here, the GPS information and 3D image information may be information collected by the user while driving directly at least once in the area where the path is to be generated, and the 3D image information may include LiDAR information and image information.

[0060] In step S105, the path generation device (100) selects a driving mode of the autonomous driving device (200). When driving mode-related information of the autonomous driving device (200) is input, the path generation device (100) selects a driving mode included in the driving mode-related information, and when driving mode-related information is not input, it selects a driving mode using GPS information and 3D image information. The driving mode may include at least one of a basic mode, a driving mode, a pest control mode, and a harvesting mode.

[0061] In step S107, the path generating device (100) determines whether the driving mode is the default mode. If the driving mode is the default mode, the path generating device (100) performs step S109, and if it is not the default mode, performs step S111.

[0062] In step S109, the path generation device (100) generates a plurality of waypoints based on a basic mode and calculates the driving speed. At this time, the path generation device (100) may use a pre-learned first generation model corresponding to the basic mode. When the generation of a plurality of waypoints and the calculation of the driving speed are completed, the path generation device (100) performs step S123.

[0063] In step S111, the path generation device (100) determines whether the driving mode is a driving mode. If the driving mode is a driving mode, the path generation device (100) performs step S113, and if it is not a driving mode, performs step S115.

[0064] In step S113, the path generation device (100) generates a plurality of waypoints based on the driving mode and calculates the driving speed. At this time, the path generation device (100) may use a previously learned second generation model corresponding to the driving mode. When the generation of the plurality of waypoints and the calculation of the driving speed are completed, the path generation device (100) performs step S123.

[0065] In step S115, the path generating device (100) determines whether the driving mode is a pest control mode. If the driving mode is a pest control mode, the path generating device (100) performs step S117, and if it is not a pest control mode, performs step S119.

[0066] In step S117, the path generation device (100) generates a plurality of waypoints based on the control mode and calculates the driving speed. At this time, the path generation device (100) may use a previously learned third generation model corresponding to the control mode. When the generation of the plurality of waypoints and the calculation of the driving speed are completed, the path generation device (100) performs step S123.

[0067] In step S119, the path generating device (100) determines whether the driving mode is a harvesting mode. If the driving mode is a harvesting mode, the path generating device (100) performs step S121, and if it is not a pest control mode, it performs step S101.

[0068] In step S121, the path generation device (100) generates a plurality of waypoints based on the harvesting mode and calculates the driving speed. At this time, the path generation device (100) may use a previously learned fourth generation model corresponding to the harvesting mode. When the generation of the plurality of waypoints and the calculation of the driving speed are completed, the path generation device (100) performs step S123.

[0069] In step S123, the path generation device (100) generates a path for the autonomous driving device (200). The path generation device (100) can generate a path based on multiple waypoints and driving speeds. The path generation device (100) transmits the generated path-related information to the autonomous driving device (200) to support the autonomous driving device (200) in performing optimized driving.

[0071] FIG. 11 is a block diagram illustrating a computing device according to an embodiment of the present invention.

[0072] Referring to FIG. 11, the computing device (TN100) may be a device described in this specification (e.g., a path generation device, an autonomous driving device, a user terminal, etc.).

[0073] The computing device (TN100) may include at least one processor (TN110), a transceiver (TN120), and a memory (TN130). Additionally, the computing device (TN100) may further include a storage device (TN140), an input interface device (TN150), an output interface device (TN160), etc. The components included in the computing device (TN100) may be connected by a bus (TN170) to communicate with each other.

[0074] The processor (TN110) can execute a program command stored in at least one of the memory (TN130) and the storage device (TN140). The processor (TN110) may refer to a central processing unit (CPU), a graphics processing unit (GPU), or a dedicated processor on which methods according to embodiments of the present invention are performed. The processor (TN110) may be configured to implement the procedures, functions, and methods described in relation to embodiments of the present invention. The processor (TN110) can control each component of the computing device (TN100).

[0075] Each of the memory (TN130) and the storage device (TN140) can store various information related to the operation of the processor (TN110). Each of the memory (TN130) and the storage device (TN140) may be composed of at least one of a volatile storage medium and a non-volatile storage medium. For example, the memory (TN130) may be composed of at least one of read-only memory (ROM) and random access memory (RAM).

[0076] The transmitting and receiving device (TN120) can transmit or receive wired or wireless signals. The transmitting and receiving device (TN120) can be connected to a network to perform communication.

[0078] Meanwhile, embodiments of the present invention are not limited to being implemented only through the apparatus and / or methods described so far, but may also be implemented through a program that realizes a function corresponding to the configuration of the embodiments of the present invention or a recording medium on which such program is recorded, and such implementation can be easily achieved by a person skilled in the art to which the present invention belongs based on the description of the embodiments described above.

[0080] Although embodiments of the present invention have been described in detail above, the scope of the present invention is not limited thereto, and various modifications and improvements made by a person skilled in the art using the basic concept of the present invention as defined in the following claims also fall within the scope of the present invention. Explanation of the symbols

[0081] 10: Communications Department 30: Input section 50: Control unit 51: Mode Selection 53: Waypoint Generation Unit 55: Path generation section 70: Output section 90: Storage section 100: Path generation device 200: Autonomous driving device 300: User terminal 400: Path generation system 450: Communication network

Claims

Claim 1 A path generation method performed by a path generation device for generating a path of an autonomous driving device, comprising: a step of learning a plurality of generation models for generating said paths; a step of inputting at least one of GPS (Global Positioning System) information serving as a reference for generating said paths, 3D image information, and information related to the driving mode of said autonomous driving device; a step of selecting a generation model corresponding to said driving mode among the plurality of previously learned generation models; a step of generating a plurality of waypoints suitable for said driving mode by applying the GPS information and the 3D image information to the selected generation model, and calculating a driving speed based on the generated plurality of waypoints. and a step of generating the path based on the plurality of waypoints and the driving speed; wherein the plurality of generating models include: a first generating model corresponding to the basic mode among the driving modes and generating a path including waypoints at uniform intervals based on the GPS information; a second generating model corresponding to the driving mode among the driving modes and generating a path that minimizes the time taken to reach a target position from a starting position and deviation from a reference line; a third generating model corresponding to the pest control mode among the driving modes and generating a path according to the density of trees included in the 3D image information; and a fourth generating model corresponding to the harvesting mode among the driving modes and generating a path according to the density of harvested fruits on the trees included in the 3D image information.A path generation method comprising at least one of the following, wherein the learning step comprises training the second generative model such that relatively more waypoints are generated in corner sections than in straight sections, training the third generative model such that waypoints are generated more densely in sections with more trees so that the time spent in sections with many trees becomes relatively longer than in sections without trees, and training the fourth generative model such that waypoints are generated more densely in sections with more harvests so that the time spent in sections with many harvests becomes relatively longer than in sections without harvests. Claim 2 delete Claim 3 A path generation method according to claim 1, wherein the learning step is characterized by performing reinforcement learning according to the characteristics of each of the plurality of generative models. Claim 4 delete Claim 5 delete Claim 6 A path generation method according to claim 1, wherein the learning step is characterized by training the plurality of generation models such that the driving speed decreases as the distance between neighboring waypoints among the plurality of waypoints narrows. Claim 7 A path generation method according to claim 1, wherein if the driving mode-related information is not input, the selecting step is characterized by predicting current task-related status information based on the GPS information and the 3D image information, and selecting a generation model corresponding to the status information among the plurality of generation models. Claim 8 A path generation device for generating a path for an autonomous driving device comprises: a learning unit for learning a plurality of generation models for generating the path; a waypoint generation unit that, when at least one of GPS information serving as a reference for generating the path and 3D image information and information related to the driving mode of the autonomous driving device is input, selects a generation model corresponding to the driving mode among the learned plurality of generation models, applies the GPS information and the 3D image information to the selected generation model to generate a plurality of waypoints suitable for the driving mode, and calculates a driving speed based on the generated plurality of waypoints; and a path generation unit for generating the path based on the plurality of waypoints and the driving speed; wherein the plurality of generation models include: a first generation model corresponding to a basic mode among the driving modes and generating a path including waypoints with uniform spacing based on the GPS information; a second generation model corresponding to a driving mode among the driving modes and generating a path that minimizes the time taken to reach a target position from a starting position and deviation from a reference line; and a third generation model corresponding to a pest control mode among the driving modes and generating a path according to the density of trees included in the 3D image information. and a fourth generation model corresponding to the harvesting mode among the above driving modes and generating a path according to the density of harvested items opened on the tree included in the three-dimensional image information;A path generation device comprising at least one of the following, wherein the learning unit trains the second generation model such that relatively more waypoints are generated in corner sections than in straight sections, trains the third generation model such that waypoints are generated more densely in sections with more trees so that the time spent in sections with many trees becomes relatively longer than in sections without trees, and trains the fourth generation model such that waypoints are generated more densely in sections with more harvests so that the time spent in sections with many harvests becomes relatively longer than in sections without harvests.

Images