KR102642036B1 - Method for providing artificial intelligence model for robot - Google Patents

Abstract

The present invention includes the steps of an artificial intelligence module performing machine learning using acquired data as training data; generating, by the artificial intelligence module, a prediction model for at least one operation of a robot based on the results of the machine learning; Installing the prediction model into a GPU (Graphics Processing Unit); A step where the GPU is mounted on a PCB, placed inside the body of a robot, and electrically connected to a sensing unit and a control unit of the robot; It relates to a method of providing an artificial intelligence model for a robot, including a step of the robot generating a driving path based on the prediction model and autonomously driving along the generated path.

Description

Method for providing artificial intelligence model for robot}

The present invention relates to a method of providing an artificial intelligence model for a robot.

Recently, robots have been put into practical use not only as industrial robots used in industries but also as household chores and office assistants in buildings such as homes, offices, and government offices. Representative examples of this include cleaning robots, guide robots, and crime prevention robots.

As artificial intelligence technology develops, artificial intelligence technology is also applied to robots. Robots equipped with artificial intelligence technology perform given tasks through machine learning using sensing data.

Robots equipped with such artificial intelligence technology have the problem of not being able to perform sufficient learning when sufficient sensing data is not secured at the time of manufacturing, and thus not showing efficient movement in performing given tasks.

Registration number 10-1987133

In order to solve the above problems, the purpose of the present invention is to provide a method of providing an artificial intelligence platform for robots so that they can efficiently perform given tasks from the time the robot is manufactured and delivered to the consumer.

The problems of the present invention are not limited to the problems mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the description below.

In order to achieve the above task, a method of providing an artificial intelligence model for a robot according to an embodiment of the present invention includes the steps of an artificial intelligence module performing machine learning using acquired data as training data; generating, by the artificial intelligence module, a prediction model for at least one operation of a robot based on the results of the machine learning; Installing the prediction model into a GPU (Graphics Processing Unit); A step where the GPU is mounted on a PCB, placed inside the body of a robot, and electrically connected to a sensing unit and a control unit of the robot; and a step of the robot generating a travel path based on the prediction model and autonomously driving along the generated path.

The electrically connecting step includes the step of the GPU exchanging signals with the control unit.

The method of providing an artificial intelligence model for a robot according to an embodiment of the present invention further includes the step of receiving, by the GPU, data generated by the sensing unit.

The method of providing an artificial intelligence model for a robot according to an embodiment of the present invention further includes the step of the GPU repeatedly performing machine learning based on data generated in the sensing unit.

The method of providing an artificial intelligence model for a robot according to an embodiment of the present invention further includes the step of the robot performing a preset operation based on the prediction model.

Specific details of other embodiments are included in the detailed description and drawings.

According to the present invention, one or more of the following effects are achieved.

First, it has the effect of efficiently performing the role of the robot without additional machine learning from the beginning of robot manufacturing.

Second, manufacturing and delivering robots with a certain degree of machine learning has the effect of increasing user convenience.

Third, depending on the operation of the robot, you can choose to install either a low-end GPU or a high-end GPU, creating a reasonable manufacturing cost.

The effects of the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description of the claims.

1 is a diagram showing the appearance of a robot according to an embodiment of the present invention.
Figure 2 is a control block diagram of a robot according to an embodiment of the present invention.
Figure 3 is a flow chart of a method for providing an artificial intelligence platform for robots according to an embodiment of the present invention.
Figure 4 is a diagram referenced to explain a method of providing an artificial intelligence platform for robots according to an embodiment of the present invention.
Figure 5 is a diagram referenced for explaining the charging operation of the robot according to an embodiment of the present invention.

Hereinafter, embodiments disclosed in the present specification will be described in detail with reference to the attached drawings. However, identical or similar components will be assigned the same reference numbers regardless of reference numerals, and duplicate descriptions thereof will be omitted. The suffixes “module” and “part” for components used in the following description are given or used interchangeably only for the ease of preparing the specification, and do not have distinct meanings or roles in themselves. Additionally, in describing the embodiments disclosed in this specification, if it is determined that detailed descriptions of related known technologies may obscure the gist of the embodiments disclosed in this specification, the detailed descriptions will be omitted. In addition, the attached drawings are only for easy understanding of the embodiments disclosed in this specification, and the technical idea disclosed in this specification is not limited by the attached drawings, and all changes included in the spirit and technical scope of the present invention are not limited. , should be understood to include equivalents or substitutes.

Terms containing ordinal numbers, such as first, second, etc., may be used to describe various components, but the components are not limited by the terms. The above terms are used only for the purpose of distinguishing one component from another.

When a component is said to be "connected" or "connected" to another component, it is understood that it may be directly connected to or connected to the other component, but that other components may exist in between. It should be. On the other hand, when it is mentioned that a component is “directly connected” or “directly connected” to another component, it should be understood that there are no other components in between.

Singular expressions include plural expressions unless the context clearly dictates otherwise.

In this application, terms such as “comprise” or “have” are intended to designate the presence of features, numbers, steps, operations, components, parts, or combinations thereof described in the specification, but are not intended to indicate the presence of one or more other features. It should be understood that this does not exclude in advance the possibility of the existence or addition of elements, numbers, steps, operations, components, parts, or combinations thereof.

1 is a diagram showing the appearance of a robot according to an embodiment of the present invention.

Referring to the drawings, a robot 100 (hereinafter referred to as robot) to which an artificial intelligence algorithm according to an embodiment of the present invention is applied is a robot that performs multiple operations. Specifically, the robot 100 runs autonomously in the target space and can perform customer service operations, quarantine operations, sterilization operations, air purification operations, cleaning operations, etc. Meanwhile, the target space may be defined as a space where the robot 100 performs various operations while autonomously driving.

Meanwhile, the robot 100 receives electrical energy from a battery contained within. In order to receive electrical energy while performing multi-operations, charging must be done according to the situation. The robot can perform charging operations in addition to the above-described operations.

Meanwhile, depending on the embodiment, the scope of the robot 100 includes a medical delivery robot, a quarantine robot, a quarantine service robot, a mobile collaborative robot, a cleaning robot, and a device installed in a specific space (e.g., an access management system, a kiosk). may be included.

Figure 2 is a control block diagram of a robot according to an embodiment of the present invention.

Referring to the drawing, the robot 100 performs actions while moving on its own in a specific space. The robot 100 may be referred to as an autonomous mobile robot. The robot 100 can drive autonomously. The robot 100 may generate a driving path based on data generated by the sensing unit 110 and the artificial intelligence module 200 and autonomously drive along the generated path.

The robot 100 includes a sensing unit 110, an artificial intelligence module 200, a communication unit 125, an input unit 130, a memory 140, a control unit 170, a driving unit 150, an operating unit 155, and It may include a charging unit 190.

The sensing unit 110 can sense situations inside or outside the robot 100. The sensing unit 110 may provide data generated by at least one sensor to the control unit 170. The control unit 170 may perform operations based on data received from the sensing unit 110.

The sensing unit 110 can detect people in the target space. For example, the sensing unit 110 can detect a person in the target space through analysis of images captured by a camera.

The sensing unit 110 can measure the pollution level of the target space. For example, the sensing unit 110 may measure the degree of pollution of the target space through data generated by an environmental sensor.

Meanwhile, the sensing unit 110 can sense internal or external situations while the robot 100 is autonomously traveling.

The sensing unit 110 includes a plurality of sensors. The sensing unit 110 may include an ultrasonic sensor, lidar, radar, infrared sensor, camera, environmental sensor, and IMU (Inertial Measurement Unit).

The ultrasonic sensor can detect objects outside the robot 100 using ultrasonic waves.

The ultrasonic sensor may include an ultrasonic transmitter and a receiver. Depending on the embodiment, the ultrasonic sensor may further include at least one control unit that is electrically connected to the ultrasonic transmitter and receiver, processes the received signal, and generates data about the object based on the processed signal. The control unit function of the ultrasonic sensor may be implemented in the control unit 170.

The ultrasonic sensor can detect an object based on ultrasonic waves and detect the location of the detected object, the distance to the detected object, and the relative speed.

The ultrasonic sensor may be placed at an appropriate location outside the robot 100 to detect objects located in front, behind, or on the sides of the robot 100.

Lidar can detect objects outside the robot 100 using laser light.

LiDAR may include an optical transmitter and an optical receiver. Depending on the embodiment, the LIDAR may further include at least one control unit that is electrically connected to the light transmitting unit and the light receiving unit, processes the received signal, and generates data about the object based on the processed signal. . The control unit function of the LIDAR may be implemented in the control unit 170.

Lidar can be implemented in a time of flight (TOF) method or a phase-shift method.

Lidar can be implemented as a driven or non-driven type.

When implemented in a driven manner, the lidar is rotated by a motor and can detect objects around the robot 100.

When implemented in a non-driven manner, the LIDAR can detect objects located within a predetermined range based on the robot 100 through optical steering.

The robot 100 may include a plurality of non-driven LIDARs.

Lidar detects objects using laser light, based on the TOF (Time of Flight) method or phase-shift method, and determines the position of the detected object, the distance to the detected object, and the relative speed. It can be detected.

Lidar may be placed at an appropriate location outside the robot 100 to detect objects located in front, behind, or on the sides of the robot 100.

Radar may include an electromagnetic wave transmitter and a receiver. Radar can be implemented as a pulse radar or continuous wave radar based on the principle of transmitting radio waves. Among the continuous wave radar methods, the radar can be implemented in the FMCW (Frequency Modulated Continuous Wave) method or FSK (Frequency Shift Keying) method depending on the signal waveform.

Radar detects objects using electromagnetic waves based on TOF (Time of Flight) method or phase-shift method, and detects the position of the detected object, the distance to the detected object, and the relative speed. You can.

The radar may be placed at an appropriate location outside the robot 100 to detect objects located in front, behind, or on the sides of the robot 100.

The infrared sensor may include an infrared transmitting unit and a receiving unit. The infrared sensor can detect an object based on infrared light, and detect the position of the detected object, the distance to the detected object, and the relative speed.

The infrared sensor may be placed at an appropriate location outside the robot 100 to detect objects located in front, behind, or on the sides of the robot 100.

The camera can capture external images of the robot 100.

The camera can generate information about objects outside the robot 100 using images. The camera may include at least one lens, at least one image sensor, and at least one control unit that is electrically connected to the image sensor, processes a received signal, and generates data about the object based on the processed signal.

The camera may be at least one of a mono camera and a stereo camera.

The camera can obtain location information of the object, distance information to the object, or relative speed information to the object using various image processing algorithms.

For example, the camera may obtain distance information and relative speed information from the acquired image based on changes in the size of the object over time.

For example, the camera can obtain distance information and relative speed information to an object through a pinhole model, road surface profiling, etc.

For example, the camera may obtain distance information and relative speed information to an object based on disparity information in a stereo image acquired from a stereo camera.

The camera may be mounted in a position where a field of view (FOV) can be secured in order to photograph the outside of the robot 100.

The robot 100 may include a plurality of cameras. For example, the robot 100 may include a four-channel camera consisting of a front camera, a rear camera, a left camera, and a right camera.

The environmental sensor can sense the level of pollution in the target space. For example, an environmental sensor can sense the presence or concentration of biochemical pollutants that are subject to quarantine in the target space.

The IMU (Inertial Measurement Unit) can measure the inertia of the robot 100. An IMU can be described as an electronic device that measures specific forces, angular rates, and sometimes magnetic fields surrounding the robot 100 using a combination of accelerometers, tachometers, and sometimes magnetometers. The control unit 170 may generate information about the posture of the robot 100 based on data received from the IMU.

The IMU may include at least one of an acceleration sensor, a gyro sensor, and a magnetic sensor.

The artificial intelligence module 200 can learn the properties of objects, spaces, and robots 100 through machine learning. Machine learning means that a computer learns from data without a person directly instructing the computer to use logic, and through this, the computer solves problems on its own.

Deep Learning is. It is an artificial intelligence technology that teaches computers how to think like humans, based on Artificial Neural Networks (ANN) to form artificial intelligence. It is an artificial intelligence technology that allows computers to learn like humans on their own without a person teaching them. .

The artificial neural network (ANN) may be implemented in software form or in hardware form such as a chip.

The artificial intelligence module 200 may include an artificial neural network (ANN) in the form of software or hardware in which properties of objects, such as spatial properties and obstacles, are learned.

For example, the artificial intelligence module 200 is a deep neural network (DNN) such as CNN (Convolutional Neural Network), RNN (Recurrent Neural Network), and DBN (Deep Belief Network) learned through deep learning. may include.

The artificial intelligence module 200 can determine the properties of space and objects included in input image data based on weights between nodes included in the deep neural network (DNN).

The artificial intelligence module 200 can recognize the properties of space and obstacles included in the selected specific viewpoint image based on data previously learned through machine learning.

Meanwhile, the memory 140 may store space, input data for determining object properties, and data for learning the deep neural network (DNN).

The memory 140 may store the original image acquired by the camera and the extracted images from which a predetermined area is extracted.

Additionally, depending on the embodiment, weights and biases forming the deep neural network (DNN) structure may be stored in the memory 140.

Alternatively, depending on the embodiment, weights and biases constituting the deep neural network structure may be stored in the embedded memory of the artificial intelligence module 200.

Meanwhile, the artificial intelligence module 200 may perform a learning process using data received through the sensing unit 110 as training data.

The robot 100 may receive data related to machine learning from the given server through the communication unit 125. In this case, the robot 100 may update the artificial intelligence module 200 based on data related to machine learning received from the predetermined server.

Data obtained through the operation of the robot 100 may be transmitted to the server. The robot 100 may transmit data related to space, objects, and usage to the server. Here, data related to space and objects are data related to the recognition of space and objects recognized by the robot 100, or data related to space and objects acquired by the camera. It may be image data for In addition, usage-related data is data acquired according to the use of a certain product, for example, the robot 100, and may include usage history data, sensing data obtained from the sensing unit 110, etc. You can.

Meanwhile, the artificial intelligence module 200 of the robot 100 may be equipped with a deep neural network (DNN) structure such as CNN (Convolutional Neural Network). The learned deep neural network structure (DNN) can receive input data for recognition, recognize the properties of objects and spaces included in the input data, and output the results. In addition, the learned deep neural network structure (DNN) receives input data for recognition, analyzes and learns data related to the usage of the robot 100, and can recognize usage patterns, usage environments, etc. .

Meanwhile, data related to space, objects, and usage may be transmitted to the server through the communication unit 125. The server can generate a configuration of learned weights, and the server can learn a deep neural network (DNN) structure using training data. The server may learn a deep neural network (DNN) based on the received data and then transmit the updated deep neural network (DNN) structure data to the robot 100 to update it.

Accordingly, the robot 100 becomes increasingly smarter and can provide a user experience (UX) that evolves as it is used.

The artificial intelligence module 200 may use the data received through the sensing unit 110 as training data to perform a learning process for detecting an object. Here, an object may be defined as an object that directly or indirectly affects the movement of the robot 100, such as objects, people, or structures around the robot 100.

The artificial intelligence module 200 can use the data received through the sensing unit 110 as training data to perform a learning process for customer service operations, quarantine operations, or learning operations.

The artificial intelligence module 200 may use the data received through the sensing unit 110 as training data to perform a learning process for scheduling operations.

The artificial intelligence module 200 may be implemented with a GPU (Graphics Processing Unit). The artificial intelligence module 200 implemented with GPU can be physically installed in the robot 100 and become part of the robot system.

The GPU on which the artificial intelligence module 200 is implemented may be mounted on a printed circuit board (PCB), placed inside the body of the robot, and electrically connected to other components of the robot 100. In particular, the artificial intelligence module 200 is electrically connected to the control unit 170, thereby enabling smart control of the robot 100 by exchanging signals. In this case, the artificial intelligence module 200 enables motion control of the robot 100 based on machine learning.

The artificial intelligence module 200 may be implemented in software or hardware form. When the artificial intelligence module 200 is implemented with a processor (eg, GPU), the artificial intelligence module 200 may be referred to as the processor on which the artificial intelligence module 200 is implemented.

The communication unit 125 may exchange signals with electronic devices external to the robot 100 (eg, a user terminal, a server, another mobile robot, a charging station, etc.).

The communication unit 125 can exchange data with an external electronic device.

The communication unit 125 may include at least one of a transmitting antenna, a receiving antenna, a radio frequency (RF) circuit capable of implementing various communication protocols, and an RF element to perform communication.

The input unit 130 is used to receive information from the user, and the data collected by the input unit 130 can be analyzed by the control unit 170 and processed as a user's control command.

The input unit 130 may include a voice input unit and a touch input unit. Depending on the embodiment, it may include a gesture input unit or a mechanical input unit.

The voice input unit may convert the user's voice input into an electrical signal. The converted electrical signal may be provided to the control unit 170. The voice input unit may include one or more microphones.

The touch input unit may convert the user's touch input into an electrical signal. The converted electrical signal may be provided to the control unit 170.

The touch input unit may include a touch sensor for detecting a user's touch input.

Depending on the embodiment, the touch input unit may be formed integrally with the display 181, thereby implementing a touch screen. This touch screen can provide both an input interface and an output interface between the robot 100 and the user.

The gesture input unit may convert the user's gesture input into an electrical signal. The converted electrical signal may be provided to the control unit 170.

The gesture input unit may include at least one of an infrared sensor and an image sensor for detecting a user's gesture input.

The mechanical input unit may include at least one of a button, a dome switch, a jog wheel, and a jog switch. The electrical signal generated by the mechanical input unit may be provided to the control unit 170.

The memory 140 is electrically connected to the control unit 170. The memory 140 can store basic data for the unit, control data for controlling the operation of the unit, and input/output data. The memory 140 may store data processed by the control unit 170. The memory 140 may be comprised of at least one of ROM, RAM, EPROM, flash drive, and hard drive in terms of hardware. The memory 140 may store various data for the overall operation of the robot 100, such as programs for processing or controlling the control unit 170. The memory 140 may be implemented integrally with the control unit 170. Depending on the embodiment, the memory 140 may be classified as a sub-component of the control unit 170.

The operation unit 155 may perform an operation suitable for the task given to the robot 100. The operation unit 155 may perform at least one of a customer service operation, a quarantine operation, a sterilization operation, an air purification operation, and a cleaning operation.

The gesture of serving customers may be an action of interacting with a person within the target space.

The quarantine operation is an operation to prevent the outbreak or epidemic of an infectious disease in advance, and may be an operation of performing at least one of chemical disinfection and physical disinfection while autonomously driving within the target space.

The sterilization operation may be an operation of spraying a chemical solution for sterilization while autonomously traveling within the target space.

The air cleaning operation may be an operation that cleans the air in the target space while autonomously driving within the target space.

The cleaning operation may be an operation of sucking up foreign substances or wiping the floor with a wet mop while autonomously driving within the target space.

The operation unit 155 may be equipped with hardware and software to perform each of the above-described operations.

The charging unit 190 may receive electrical energy from a charging station. The charging part 190 is. The supplied electrical energy can be stored in a battery.

The control unit 170 can control the overall operation of each unit of the robot 100. The control unit 170 may be called an Electronic Control Unit (ECU). The control unit 170 is electrically connected to each unit of the robot 100.

The control unit 170 includes application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), processors, and controllers. It may be implemented using at least one of controllers, micro-controllers, microprocessors, and other electrical units for performing functions.

Figure 3 is a flow chart of a method for providing an artificial intelligence platform for robots according to an embodiment of the present invention.

Referring to the drawing, the artificial intelligence module 200 can acquire data (S310). Here, the data may be data obtained from a sensing unit included in the first robot. Alternatively, the data may be data learned by the first robot using data acquired from the sensing unit.

The artificial intelligence module 200 may perform machine learning using the acquired data as training data (S320).

The artificial intelligence module 200 may generate a prediction model for at least one operation of the robot based on the results of machine learning (S330). The prediction model may be a model using at least one artificial neural network (ANN). A prediction model is a result of learning past data, and can be understood as a model that outputs prediction data for that time and space when input values for time and space are input.

The artificial intelligence module 200 may be implemented with a GPU (Graphics Processing Unit) (S340). In this case, the prediction model is implemented on GPU. For example, when the artificial intelligence module 200 is implemented as software, the artificial intelligence module 200 can be implemented as a GPU by installing software on the GPU. Meanwhile, depending on the embodiment, if the artificial intelligence module 200 is implemented as hardware, step S340 may be omitted.

As the artificial intelligence module 200 is implemented with GPU, distributed processing becomes possible, thereby increasing computational efficiency.

Meanwhile, in step S340, GPU specifications may be determined based on the type of operation of the robot 100.

The artificial intelligence module 200 can be installed on the robot 100 (S350). In this case, the prediction model is installed in the robot 100. The GPU on which the artificial intelligence module 200 is implemented is mounted inside the robot 100 and electrically connected to the unit included in the robot 100, so that the artificial intelligence module 200 can be installed in the robot 100. there is. The artificial intelligence module 200 includes a sensing unit 110, a communication unit 125, an input unit 130, a memory 140, a control unit 170, a driving unit 150, an operating unit 155, and a charging unit 190. Signals can be exchanged directly or indirectly with at least one of them.

Meanwhile, the robot 100 on which the artificial intelligence module 200 is installed may be a second robot that is different from the first robot that is the source of data acquisition in step S310.

Step S350 may include a step where the artificial intelligence module 200 is electrically connected to the control unit 170 of the robot 100 and exchanges signals.

With the artificial intelligence module 200 installed in the robot 100, the robot 100 can perform an operation (S360). The robot 100 may perform operations based on a prediction model.

The artificial intelligence module 200 may input data about time and area into a prediction model to generate prediction information about the floating population and pollution level for that time and area. Meanwhile, the area can be described as a unit space divided by dividing the target space based on a preset standard.

The artificial intelligence module 200 may generate an operation schedule based on prediction information. The artificial intelligence module 200 may generate a schedule for at least one of a customer service operation, a quarantine operation, a sterilization operation, an air purification operation, a cleaning operation, and a charging operation, based on the prediction information.

The control unit 170 may perform operations according to the created schedule.

The artificial intelligence module 200 can receive data generated by the sensing unit 110 (S370).

The artificial intelligence module 200 may repeatedly perform machine learning based on data generated by the sensing unit 110 (S380).

Meanwhile, by installing the artificial intelligence module 200 in the robot 100, the artificial intelligence module 200 can be utilized when processing operations related to ROS (Robot Operating System), Darknet & YOLO, and Cloud.

Figure 4 is a diagram referenced to explain a method of providing an artificial intelligence platform for robots according to an embodiment of the present invention.

Referring to the drawing, when the artificial intelligence module 200 is implemented with a GPU, the specifications of the GPU may be determined depending on the robot 100 to which the GPU is installed. GPUs with greater processing power can be used for robots that perform functions requiring more data processing.

The artificial intelligence module 200 may be implemented and installed on the first GPU 410 in the first robot performing the first operation. The artificial intelligence module 200 may be implemented and installed in the second GPU 420 in the second robot performing the second operation. The data processing amount required for the first operation is greater than the data processing amount required for the second operation. The processing power of 1 GPU 410 is greater than that of the second GPU 420. The first GPU 410 may be classified as a high-spec GPU and the second GPU 420 may be classified as a low-spec GPU. .

When the GPUs 410 and 42 are installed in the robot 100 and electrically connected to the control unit 170, the GPUs 410 and 420 and the control unit 170 are linked to control the robot 100. do.

Figure 5 is a diagram referenced for explaining the charging operation of the robot according to an embodiment of the present invention.

Referring to the drawing, the charging station 400 can receive electrical energy from a commercial power source and provide it to the robot 100. The charging station 400 may supply electrical energy to the robot 100 when the robot 100 is docked while connected to a commercial power source. The robot 100 and the charging station 400 can be classified as sub-concepts of a robot system.

The communication unit 125 can exchange signals with the charging station 400.

The control unit 170 may determine whether the battery needs to be charged. For example, the control unit 170 may determine whether the battery needs to be charged based on the amount of remaining electrical energy in the battery. For example, the control unit 170 may determine whether the battery needs to be charged based on the amount of remaining electrical energy compared to the mission of the robot 100. If the amount of energy stored in the battery is less than the amount of energy required to perform the function of the robot 100, the control unit 170 may determine that charging the battery is necessary.

If it is determined that the battery needs to be charged, the control unit 170 may transmit a charging standby signal to the charging station 400 through the communication unit 125.

When the charging station 400 receives a charging standby signal, it may be switched to standby mode. When switched to standby mode, the charging station 400 can wake up the system by supplying power to each unit of the charging station.

If it is determined that the battery needs to be charged, the control unit 170 may control the driving unit 150 to move the robot 100 toward the charging station 400.

The control unit 170 may determine whether the distance between the robot 100 and the charging station 400 is within a reference value. For example, the control unit 170 may calculate the distance between the robot 100 and the charging station 400 based on the strength of the communication signal and determine whether the distance is within a reference value. For example, the control unit 170 may calculate the distance between the robot 100 and the charging station 400 based on data generated by the sensing unit 110 and determine whether the distance is within a reference value.

If it is determined that the distance between the robot 100 and the charging station 400 is within the reference value, the control unit 170 may transmit a charging preparation signal to the charging station 400 through the first communication unit 125. .

The charging station 400 may be switched to the ready mode when receiving a charging ready signal. When switched to the ready mode, the charging station 400 may project a three-dimensional marker.

The control unit 170 may control the driving unit 150 so that the robot 100 docks with the charging station 400 .

The control unit 170 may determine the relative position of the robot 100 compared to the charging station 400.

The control unit 170 may receive image data of a marker of the charging station 400 from a camera. The control unit 170 may determine the relative position of the robot 100 with respect to the charging station based on the state of the three-dimensional marker detected in the image data.

Meanwhile, the marker is an element to help the robot 100 dock with the charging station 400 and is placed on the charging station 400. The marker may include a first marker, a second marker, and a third marker. The first marker may be called a three-dimensional marker. The first marker may move linearly according to the power provided. The first marker may protrude from the surface of the charging station by a first height by the provided power. The first marker can be stored by provided power.

The first marker in a protruding state may be located on the centerline when viewed from above. The second marker is spaced apart from the first marker by a first distance in the first direction. The third marker is spaced apart from the first marker by a first distance in a second direction opposite to the first direction.

The control unit 170 may detect a first marker image, a second marker image, and a third marker image from the received image data.

When the first marker image is closer to the second marker image or closer to the third marker image, the control unit 170 may determine that the robot 100 is not located at the center line.

When the first marker image is located in the center of the second marker image and the third marker image, the control unit 170 may determine that the robot 100 is located at the center line. Here, the center does not mean a mathematically accurate center, but rather means a degree of leaving a certain amount of variance on the left and right, reflecting small numerical errors. Meanwhile, the center line may be defined as an imaginary line formed at a right angle to at least one corner of the charging station while dividing the center between the second marker and the third marker on the floor when viewed from above.

When it is determined that the robot 100 is located at the center line, the control unit 170 may control the driving unit 150 so that the robot 100 moves in a straight line toward the charging station 400. The control unit 170 may attempt docking by moving the robot 100 in a straight line.

The control unit 170 may determine whether the robot 100 has been docked to the charging station 400.

For example, in the ready mode, the charging station 400 may introduce a relatively weak current into the charging terminal of the charging station 400. When it is detected that a weak current is flowing into the charging terminal of the charging unit 190, the control unit 170 may determine that the robot 100 has completed docking to the charging station 400. For this purpose, the charging unit 190 may include a current sensor.

For example, the control unit 170 determines that the center of the robot 100 is located on the center line, and the distance between the robot 100 and the charging station 400 calculated based on the sensing data generated by the sensing unit 110. If it is within the set range, it may be determined that the robot 100 has completed docking to the charging station 400.

For example, the charging station 400 may determine whether the robot 100 has contacted the three-dimensional marker. When the robot 100 touches the three-dimensional marker while moving straight toward the charging station 400 along the center line, the three-dimensional marker moves slightly in the direction of movement of the robot 100. When unintended linear movement of the three-dimensional marker is detected, the charging station 400 may determine that the robot 100 has been docked to the charging station 400. The charging station 400 may transmit docking completion status information to the robot 100. The control unit 170 may determine that the robot 100 has completed docking to the charging station 400, based on information received from the charging station 400.

When it is determined that the robot 100 is docked at the charging station 400, the control unit 170 may transmit a charging request signal to the charging station 400.

The control unit 170 can control the on and off of the relay depending on whether the robot 100 and the charging station 400 are docked. For example, when the robot 100 is fully docked at the charging station 400, the robot 100 may be switched to the charging mode. When switching to the charging mode, the control unit 170 can switch the relay to the on state.

When the robot 100 is completely docked at the charging station 400, the charging station 400 may be switched to charging mode.

In the charging mode, the charging station 400 may apply power to the charging unit 190 of the robot 100 and transmit a charging start signal to the robot 100.

The control unit 170 can charge the battery with electrical energy provided from the charging station 400.

When it is determined that charging of the battery is complete, the control unit 170 may transmit a charging completion signal to the charging station 400.

When a charging completion signal is received, the charging station 400 may stop supplying electrical energy.

The present invention described above can be implemented as computer-readable code on a program-recorded medium. Computer-readable media includes all types of recording devices that store data that can be read by a computer system. Examples of computer-readable media include HDD (Hard Disk Drive), SSD (Solid State Disk), SDD (Silicon Disk Drive), ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, etc. There is. Additionally, the computer may include a processor or control unit. Accordingly, the above detailed description should not be construed as restrictive in all respects and should be considered illustrative. The scope of the present invention should be determined by reasonable interpretation of the appended claims, and all changes within the equivalent scope of the present invention are included in the scope of the present invention.

100: Robot
S300: How to provide an artificial intelligence platform for robots

Claims (5)

An artificial intelligence module performing machine learning using the acquired data as training data;
generating, by the artificial intelligence module, a prediction model for at least one operation of a robot based on the results of the machine learning;
Installing the prediction model into a GPU (Graphics Processing Unit);
A step where the GPU is mounted on a PCB, placed inside the body of a robot, and electrically connected to a sensing unit and a control unit of the robot; and
A method of providing an artificial intelligence model for a robot, including the step of the robot generating a driving path based on the prediction model and autonomously driving along the generated path. According to clause 1,
The electrically connected step is,
A method of providing an artificial intelligence model for a robot comprising: the GPU exchanging signals with the control unit. According to clause 2,
A method of providing an artificial intelligence model for a robot further comprising: receiving, by the GPU, data generated by the sensing unit. According to clause 3,
A method of providing an artificial intelligence model for a robot, further comprising: the GPU repeatedly performing machine learning based on data generated by the sensing unit. According to clause 1,
A method of providing an artificial intelligence model for a robot, further comprising: allowing the robot to perform a preset operation based on the prediction model.