Vision recognition coding training method and coding training platform using the same

Abstract

A vision recognition coding education robot according to one embodiment of the present invention is a robot for coding education and includes a coding block recognition unit that recognizes coding blocks, a coding logic generation unit, a coding logic verification unit, and a coding result display unit. The coding result display unit may display a process of proceeding to an operation step when the verification of the coding logic by the coding logic verification unit is successful, and may display hint information for resolving coding logic errors when the verification of the coding logic by the coding logic verification unit fails.

Description

Technology Field

[0001] The present invention relates to a vision recognition coding education robot and a coding education platform using the same. Background Technology

[0002] In order to overcome the difficulties of coding education (including the meaning of learning) through text-based coding, research on block-based coding (hereinafter referred to as block coding) methods is active. This block coding is a method of coding computer programs by dragging and stacking blocks, which allows learners to easily visually check each step of coding, thereby facilitating coding education (including learning).

[0003] Educational tools for such block coding have generally used blocks with built-in electronic circuits. While this method has the advantage of allowing direct verification of the results of coding execution through blocks, it can lead to problems such as power supply inconvenience, limitations on block combinations, instability of wireless connections between blocks and actuators, and frequent breakdowns. The problem to be solved

[0004] The present invention aims to solve the problems described above by providing a vision recognition coding education robot and a coding education platform using the same, which eliminate the need for electronic devices during block coding education, increase the degree of freedom of block combinations by recognizing and processing block combinations using a camera, and enhance ease of use in coding education settings to make coding education more effective and interesting.

[0005] The objectives of the present invention are not limited to those mentioned above, and other unmentioned objectives will be clearly understood by those skilled in the art from the description below. means of solving the problem

[0006] A vision recognition coding education robot according to one embodiment of the present invention for achieving the above technical problem is a robot for coding education and includes a block recognition unit that recognizes coding blocks, a coding logic generation unit, a coding logic verification unit, and a coding result display unit.

[0007] In addition, the block recognition unit may include an artificial intelligence coding logic calling unit that calls the artificial intelligence coding logic.

[0008] In addition, the coding result display unit can display the process of proceeding to the operation step when the verification of the coding logic by the coding logic verification unit is successful, and can display hint information for resolving coding logic errors when the verification of the coding logic by the coding logic verification unit fails.

[0009] In addition, a vision recognition coding education robot according to one embodiment of the present invention may further include a hint information generation unit that generates hint information for resolving coding logic errors based on rules.

[0010] In addition, a vision recognition coding education robot according to one embodiment of the present invention may further include a motion control information generation unit that generates motion control information of a robot based on the result of recognizing the coding logic of a coding block recognized by a coding block recognition unit.

[0011] In addition, a vision recognition coding education robot according to one embodiment of the present invention may further include an object control unit that controls the operation of the robot based on motion control information.

[0012] A coding education platform according to an embodiment of the present invention for achieving the above technical problem includes a data collection unit that receives block coding education information from a robot, an artificial intelligence learning unit that learns the received block coding education information and generates coding learning information, and a learning data generation unit that generates customized learning data for a new user who will use the robot.

[0013] In addition, block coding education information may include at least one of information about the robot user, information about the artificial intelligence coding logic used by the robot user, coding logic verification information, information about coding logic errors that occurred, information about generated error resolution hints, and information about the error resolution process.

[0014] In addition, the AI ​​learning unit can generate key learning information, which is information about the areas where new users need to focus their learning, through AI modeling based on the generated coding learning information.

[0015] In addition, the learning data generation unit can generate customized learning data for new users based on key learning information and information about new users. Effects of the invention

[0016] According to the present invention, the need for electronic devices during block coding education is eliminated, and by recognizing and processing block combinations using a camera, the degree of freedom in block combinations is increased, thereby enhancing usability in the coding education setting and making coding education more effective and interesting. In addition, the efficiency of coding education can be further enhanced by providing user-customized learning data.

[0017] In addition to this, various effects that can be identified directly or indirectly through this document may be provided. Brief explanation of the drawing

[0018] Figure 1 is a configuration diagram of a block-based vision recognition coding education system according to one embodiment of the present invention. FIG. 2 is an exemplary diagram of the operation of a block-based vision recognition coding education system according to another embodiment of the present invention. FIG. 3 is an exemplary diagram of the operation of a block-based vision recognition coding education system according to another embodiment of the present invention. FIG. 4 is a configuration diagram of a vision recognition coding education robot according to one embodiment of the present invention. FIG. 5 is a configuration diagram of a coding education platform according to one embodiment of the present invention. FIG. 6 is a flowchart of a block-based vision recognition coding education method according to one embodiment of the present invention. Specific details for implementing the invention

[0019] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the attached drawings. The advantages and features of the present invention and the methods for achieving them will become clear by referring to the embodiments described below in detail together with the attached drawings. However, the present invention is not limited to the embodiments disclosed below but may be implemented in various different forms. These embodiments are provided merely to ensure that the disclosure of the present invention is complete and to fully inform those skilled in the art of the scope of the invention, and the present invention is defined only by the scope of the claims. In relation to the description of the drawings, the same or corresponding components may be assigned the same reference number.

[0021] Although terms such as "first," "second," etc. are used to describe various elements, components, and / or sections, it goes without saying that these elements, components, and / or sections are not limited by these terms. These terms are used merely to distinguish one element, component, or section from another. Accordingly, it goes without saying that the first element, first component, or first section mentioned below may be a second element, second component, or second section within the technical scope of the present invention.

[0022] The terms used herein are for describing embodiments and are not intended to limit the invention. In this specification, the singular form includes the plural form unless specifically stated otherwise in the text. As used herein, "comprises" and / or "made of" do not exclude the presence or addition of one or more other components, steps, actions, and / or elements to the mentioned components, steps, actions, and / or elements.

[0023] Unless otherwise defined, all terms used in this specification (including technical and scientific terms) may be used in a meaning commonly understood by those skilled in the art to which the present invention pertains. Additionally, terms defined in commonly used dictionaries are not to be interpreted ideally or excessively unless explicitly and specifically defined otherwise.

[0024] Hereinafter, the configuration of the present invention will be described in detail with reference to the attached drawings.

[0026] Figure 1 is a configuration diagram of a block-based vision recognition coding education system according to one embodiment of the present invention.

[0027] Referring to FIG. 1, a block-based vision recognition coding education system (100) (hereinafter referred to as the block recognition coding education system (100)) according to one embodiment of the present invention includes a plurality of coding blocks (110), a block recognition unit (120) that recognizes coding logic formed by combining a plurality of coding blocks, and an object control unit (130) that controls the operation of a control target object based on the recognized coding logic.

[0028] Additionally, the block recognition coding education system (100) may further include a feedback analysis execution unit (140) that performs an evaluation of the recognized coding logic based on sensor information received from a controlled object and displays feedback including the analysis result of the performed evaluation.

[0029] A coding block (110) of a block recognition coding education system (100) according to one embodiment provides a physical interface for coding and artificial intelligence education and represents specific coding commands or functions. This coding block (110) can be designed to allow coding by physically assembling code so that lower-grade students who have just started learning coding can easily understand and use it. A coding block (110) according to one embodiment is composed of a plurality (n) of individual coding blocks (110-1, ..., 110-n).

[0030] A coding block (110) according to one embodiment may be implemented in a form that is placed in a tray having a card slot, or may be implemented in the form of a Lego block as shown in FIGS. 2 and FIGS. 3 (however, it is not necessarily limited to this form and may be implemented in various forms depending on the implementation method, such as a magnetic attachment method). Each coding block (110-1, ..., 110-n) may represent a specific coding command or function as follows.

[0031] (1) Program Start / End Block: A block that indicates the start and end of a coding program.

[0032] (2) Number block: Used to display or input data values.

[0033] (3) Variable and Value Assignment Block: Performs the function of assigning values ​​to variables within the program.

[0034] (4) Loop block: Used to set a specific action to repeat.

[0035] (5) Branching block: A conditional block that performs different actions depending on the condition.

[0036] (6) Actuator control block: Used to control the operation of robots or other mechanical devices. For example, it can be implemented to perform commands such as "turn red LED on / off", "move for 3 seconds", and "turn left".

[0037] (7) Artificial Intelligence Block: A block used to train artificial intelligence models such as supervised learning, unsupervised learning, and reinforcement learning.

[0038] A coding block (110) according to one embodiment may be implemented so as not to include an electronic circuit. A coding block (110) according to one embodiment may be implemented with various color processing, various marker display, various geometric shapes, etc., so that a block recognition unit (120) to be described later can easily recognize it.

[0039] To perform image recognition-based processing of coding blocks (110), the block-based vision recognition coding education system (100) can utilize color, marked marker, and implemented shape information expressed in the coding blocks (110), and through the combination of these, the number of data expressions can be increased and the image recognition rate can be improved.

[0040] A coding block according to one embodiment can be recognized by a block recognition unit (120) to be described later. To this end, the coding block (110) according to one embodiment may be implemented in various colors to facilitate recognition by the block recognition unit (120). At this time, the colors used in the coding block (110) according to one embodiment may be implemented using only colors that differ significantly in the HSV color space (examples of such colors include red, yellow, green, blue, purple, etc., but are not necessarily limited thereto and can be determined based on the implementation method of the block recognition unit (120)).

[0041] A coding block (110) according to one embodiment may be implemented in a specific shape so as to be easily recognized by a block recognition unit (120) (for example, a star, triangle, square, circle, etc. are possible, but are not necessarily limited thereto).

[0042] In one embodiment, the coding block (110) can have its borders implemented in a specific color so that the boundaries between the coding blocks (110) are clearly recognized by the block recognition unit (120) (for example, black, but not necessarily limited thereto, and can be determined based on the implementation method of the block recognition unit (120)).

[0043] A block recognition unit (120) of a block recognition coding education system (100) according to one embodiment of the present invention recognizes a coding block (110). A block recognition unit (120) according to one embodiment is implemented as a vision recognition camera and can recognize a coding block (110) by scanning the color of the coding block (110), the expressed marker, the shape of the figure, and the combination form between the coding blocks.

[0044] A block recognition unit (120) according to one embodiment can recognize a coding logic formed by combining a plurality of coding blocks (110), and can be implemented to utilize computer vision and artificial intelligence technologies for this purpose. That is, the block recognition unit (120) according to one embodiment can accurately determine the position, direction, and combination form of the coding blocks (110) by analyzing a recognized image (i.e., an image captured by a camera) using computer vision technology to recognize how the coding blocks are physically arranged.

[0045] Additionally, the block recognition unit (120) according to one embodiment may be implemented to continuously detect environmental changes using a machine learning model for real-time image processing functions and to react immediately to such environmental changes. Through this, the block recognition unit (120) according to one embodiment can immediately recognize and process when the position or direction of the coding block (110) changes (for example, by applying a deep learning algorithm such as TensorFlow, the color, shape, and position changes of the coding block (110) can be quickly recognized).

[0046] A block recognition unit (120) according to one embodiment can identify the color of each coding block (110) using an HSV (Hue, Saturation, Value) color space. As described above, the coding block may be implemented as a specific shape or have a marker expressed on it, and since these markers and shapes can provide additional information about the coding block (110), the block recognition unit (120) according to one embodiment can identify specific functions of the coding block by recognizing shapes such as a star shape, triangle, and square, and markers using an image processing algorithm.

[0047] A block recognition unit (120) according to one embodiment can recognize coding blocks more precisely by using a trained deep learning model. The training model used at this time is trained on a training dataset and can recognize detailed characteristics of blocks by utilizing algorithms such as Convolutional Neural Networks (CNN), Transfer Learning, and You Only Look Once (YOLO).

[0048] Information of a coding block (110) recognized by a block recognition unit (120) according to one embodiment is converted into a programming code to form a programming logic (also referred to as coding logic), and the block recognition unit (120) according to one embodiment can interpret the connection and combination between individual coding blocks (110-1, ..., 110-n) to form a coding command and transmit it to a control target object (200) through an object control unit (130) to perform a desired operation.

[0049] A block recognition unit (120) according to one embodiment may be installed on a control target object (200) as shown in FIGS. 2 and 3, and may be implemented as a block recognition unit (120) using one or more cameras separated from the control target object (200) (depending on the implementation method, it may be installed both inside and outside the control target object (200)).

[0050] The process of recognizing a coding block (110) by a block recognition unit (120) according to one embodiment can be performed in the following procedure.

[0051] (1) The boundaries of the program are determined by searching for the start block and the end block among the coding blocks, and if they are not recognized, the block position adjustment can be notified to the user through the display and / or speaker (not shown) of the block recognition coding education system (100). Meanwhile, the user in the following embodiments includes a coding educator, a coding student, an administrator of the block-based vision recognition coding education system, etc.

[0052] (2) Image correction: If image distortion occurs due to the misalignment of the coding block (110), the coding block (110) or the number recognition rate may decrease. Therefore, the misalignment of the image can be checked using markers present in the start block and end block, and the image can be corrected by rotating it.

[0053] (3) The color of the image of the coding block (110) recognized by the block recognition unit (120) can be converted to gray and HSV.

[0054] (4) The boundaries of the program can be determined by searching for the start and end blocks in the gray image.

[0055] (5) The boundaries of the straight line and the coding block (110) can be derived using the Canny algorithm and the Hough Transform.

[0056] (6) The type of coding block (110) can be determined by recognizing the color, marker, and shape of the coding block (110). At this time, the color can be determined by the HSV hue value, and the marker and shape can be determined by training a deep learning-based image classification model such as CNN, transfer learning, or YOLO.

[0057] An object control unit (130) of a block recognition coding education system (100) according to one embodiment of the present invention can control the operation of a control target object (200) (e.g., a robot, actuator, etc.) based on coding logic recognized by a block recognition unit (120) (i.e., can transmit a command to the control target object (200) using wired or wireless communication so that the control target object (200) performs a specific operation according to the coding logic).

[0058] An object control unit (130) according to one embodiment analyzes coding logic received from a block recognition unit (120). The coding logic includes a combination of each coding block and corresponding programming commands, and the object control unit (120) according to one embodiment can set actual operations based on this.

[0059] At this time, an object control unit (130) according to one embodiment can interpret the recognized coding block (110) sequentially from top to bottom to interpret the code (coding logic) represented by the coding block. At this time, if the configuration of the coding block (110) is incorrect, this can be notified to the user through a display device (not shown, such as a display and / or speaker) of the block-based vision recognition coding education system (100).

[0060] For example, through the display device of the block-based vision recognition coding education system (100), guidance such as "Please connect the start block to the top" or "The loop requires a number of repetitions" can be provided, and the location of the block where the problem occurred can also be highlighted for user-friendly debugging. If it is determined that there are no programmatic errors in the recognized coding block (110), the object control unit (130) according to one embodiment can control the target object (200) according to the interpreted code.

[0061] As described below, various types of actuators may be installed in the control target object (200) (for example, a robot arm equipped with a motor, a rotating wheel, etc.). An object control unit (200) according to one embodiment can control each actuator individually to accurately perform the desired operation.

[0062] For example, the object control unit (130) can convert and execute these commands into actual mechanical and electronic movements according to the coding logic analyzed as commands such as "turn left" or "move for 3 seconds." At this time, the object control unit (130) according to one embodiment can maintain the accuracy of the operation of the controlled object (200) by monitoring sensor data (also referred to as environmental data) obtained through the controlled object (200) or a sensor installed separately externally.

[0063] That is, the control target object (200) can collect environmental data through various sensors attached to the control target object (200) (environmental data may include the location, speed, and surrounding environment state of the control target object (200)), and the object control unit (130) can use the received environmental data to more precisely control the operation of the control target object (200).

[0064] An object control unit (130) according to one embodiment can detect and respond in real time to errors or exception situations that may occur during execution. For example, if there is a risk that a controlled object (200) may collide with an unexpected obstacle, the object control unit (130) according to one embodiment can take emergency measures to immediately stop the operation and maintain a safe state.

[0065] A feedback analysis unit (140) of a block recognition coding education system (100) according to one embodiment of the present invention can evaluate and analyze the execution result of the coding logic of a coding block (110) based on environmental data, which is sensor information received from a control target object (200) (or, if an environmental measurement sensor is installed away from the control target object (200), environmental data can be received from it).

[0066] The result analyzed by the feedback analysis performing unit (140) according to one embodiment is provided to the user as feedback information, so that the user can improve the error of the block recognition coding education system (100) and the coding skills of the coding student and correct the error.

[0067] That is, the feedback analysis performing unit (140) according to one embodiment performs an evaluation of the coding logic recognized by the block recognition unit (120) based on environment data received from the control target object (200), and can display feedback including the analysis result of the performed evaluation.

[0068] Specifically, a feedback analysis unit (140) according to one embodiment can analyze how well an action performed by a controlled object (200) matches the coding logic programmed by the user based on the action performed by the controlled object (200) (for example, it analyzes whether the controlled object (200) accurately followed a set path, whether a designated task was successfully completed, etc.).

[0069] A feedback analysis unit (140) according to one embodiment can identify errors or problems that may occur during the execution of coding logic (e.g., unexpected obstacles, inaccuracies in operation, timing errors, etc.) by comparing them with received environment data. Based on the analysis results of the environment data, the feedback analysis unit (140) according to one embodiment can provide specific and clear feedback information through a user-friendly interface so that the user can improve their coding skills.

[0070] In this case, the feedback may be provided in the form of text messages, graphic displays, voice guidance, etc., and may include instructions on which parts of the coding logic need improvement and how to achieve better results (i.e., instructions on the types or arrangement of coding blocks, etc.).

[0071] The control target object (200) of the block-based vision recognition coding education system (100) according to one embodiment is an object controlled by the block-based vision recognition coding education system (100), and may be a physical device or an object implemented as a program. The control target object (200) according to one embodiment may take various forms such as a robot, motor, light, etc., and may operate according to the commands of the coding logic that is input and analyzed through coding blocks.

[0072] According to one embodiment, a control target object (200) may integrate various sensors to collect environmental data of the control target object (200) and transmit the collected environmental data to a feedback analysis execution unit (140) to be used for evaluating the effectiveness of the coding logic. Specifically, according to one embodiment, the control target object (200) may perform various actions, such as moving along a predetermined path or performing a specific task, according to programming logic composed of block code.

[0073] A control target object (200) according to one embodiment may have various sensors embedded to collect environmental data, which is information about the surrounding environment, and transmit it to at least one of an object control unit (130) and a feedback analysis performing unit (140). The environmental data may include the location, speed, and presence of surrounding obstacles of the object, and may be used to more precisely adjust the operation of the control target object (200).

[0074] According to one embodiment, a controlled object (200) reports feedback information, which is the result of an executed operation, to an object control unit (130) and a feedback analysis unit (140) of a block-based vision recognition coding education system (100), and the feedback analysis unit (140) can provide feedback information regarding the accuracy and effectiveness of the operation to a user based on this feedback information.

[0075] A control target object (200) according to one embodiment can immediately respond to the coding logic of a coding block combined by the user, thereby strengthening the link between the user's actions and results. A control target object (200) according to one embodiment can perform more complex tasks through a learned artificial intelligence model, as shown in FIG. 3, beyond simple command execution. Through this, the control target object (200) can respond flexibly in various situations and solve complex problems.

[0076] That is, as shown in FIG. 3, individual coding blocks (110-0, ..., 110-n) of the coding block (110) can be marked to perform specific coding tasks (e.g., command execution, conditional branching, repetition, etc.), and coding logic can be constructed by physically combining these individual coding blocks (110-0, ..., 110-n). As shown in FIG. 2 or FIG. 3, a robot (when equipped with a block-based vision recognition coding education system) which is a controlled object (200) can analyze and interpret the type and arrangement of each individual coding block by photographing the arrangement of coded blocks with a camera of the mounted block recognition unit (120).

[0077] A block recognition coding education system (100) according to one embodiment may be mounted on a robot, which is a controlled object (200), as shown in FIGS. 2 and 3. Alternatively, a block recognition coding education system (100) according to one embodiment may be implemented separately from the controlled object (200).

[0078] In addition, a block recognition coding education system (100) according to one embodiment may include an artificial intelligence-based coding education platform (600) that learns and executes an artificial intelligence model through a computer connected via a wired or wireless communication method such as Bluetooth, Wi-Fi, or serial communication, as shown in FIG. 3. That is, an artificial intelligence model learned in the education platform (200) according to one embodiment may be transmitted to a control target object (200) via a wireless communication method such as Bluetooth or Wi-Fi or a wired communication method such as serial communication.

[0079] A coding education platform (600) according to one embodiment can instruct a robot, which is a controlled object (200), to perform an action corresponding to the result recognized in a block combination recognized by the robot. A coding education platform (600) according to one embodiment can be used to teach coding logic to a user and can provide the user with a combination of digital materials and a practice environment. Through this, the user can proceed with coding theory learning and practice simultaneously.

[0081] FIG. 4 is a configuration diagram of a vision recognition coding education robot according to one embodiment of the present invention.

[0082] Referring to FIG. 4, a vision recognition coding education robot (400) according to one embodiment of the present invention is a robot for block coding education and includes a block recognition unit (410) that recognizes coding blocks, a coding logic generation unit (420), a coding logic verification unit (430), and a coding result display unit (4400).

[0083] A coding block recognition unit (410) according to one embodiment recognizes a coding block (110) as the block recognition unit (120) described above. A coding block recognition unit (410) according to one embodiment may include an artificial intelligence (AI) coding logic calling unit (not shown) that dynamically calls an AI model or algorithm based on information of the recognized coding blocks. That is, an AI coding logic calling unit according to one embodiment may select and call an appropriate AI model based on the result of coding block recognition (for example, if a specific combination of coding blocks is predefined in a logic pattern library, an AI model capable of processing the corresponding logic may be executed to generate or optimize coding logic in real time).

[0084] A coding logic generation unit (420) according to one embodiment generates coding logic from recognized coding blocks. A coding logic generation unit (420) according to one embodiment receives information (e.g., shape, color, marker) of coding blocks recognized by a coding block recognition unit (410) using a camera and image processing technology, and can interpret the type and order of the coding blocks based on this information. A coding logic generation unit (420) according to one embodiment identifies programming commands (e.g., loops, conditional statements, variable assignments, etc.) expressed by each recognized coding block and can combine actual executable coding logic based on this (in this process, the structural accuracy of the logic can be ensured by reviewing the connection structure between coding blocks and programming syntax).

[0085] A coding logic generation unit (420) according to one embodiment can convert a structurally accurate coding logic into a programming language or script suitable for the execution environment of a vision recognition coding education robot (400). A coding logic generation unit (420) according to one embodiment can recognize the visual features of a coding block by utilizing image recognition and processing technology based on a machine learning algorithm.

[0086] A coding logic verification unit (430) according to one embodiment can verify the coding logic generated by the coding logic generation unit (420) and perform additional optimization. That is, the coding logic verification unit (430) according to one embodiment verifies whether the generated coding logic possesses grammatical and logical accuracy of programming. The coding logic verification unit (430) according to one embodiment can verify whether the block-based coding logic assembled by the learner complies with the grammatical rules of the programming language (for example, verifying whether the condition within a loop is correct, the appropriateness of variable declarations, etc.).

[0087] A coding logic verification unit (430) according to one embodiment can verify logical errors that may be included even in grammatically correct code. A coding logic verification unit (430) according to one embodiment can verify logical consistency so that the combination of blocks can operate as actually intended (for example, verifying whether the code is written so as not to fall into an infinite loop, whether conditional statements are executed in appropriate situations, etc.).

[0088] According to one embodiment, the coding result display unit (440) can display the process of proceeding to the operation step when the verification of the coding logic by the coding logic verification unit (430) is successful, and can display coding logic error resolution hint information when the verification of the coding logic fails.

[0089] A vision recognition coding education robot (400) according to one embodiment may further include a hint information generating unit that generates hint information for resolving coding logic errors based on rules. A hint information generating unit (450) according to one embodiment may generate feedback corresponding to a hint that can resolve errors found in coding logic composed of coding blocks by a user (including various types of coding errors such as grammatical errors, logical errors, and execution errors).

[0090] That is, the hint information generation unit (450) according to one embodiment can identify the cause of the error found by the coding logic verification unit (430) and then generate a hint for an appropriate solution considering the user level. At this time, the hint information generation unit (450) according to the embodiment can generate hints multiple times (the number of times is determined according to the implementation method). In the first hint generation, a hint that is difficult to estimate the correct answer is generated, and as the number of times increases, hints closer to the correct answer can be generated.

[0091] A hint information generation unit (450) according to one embodiment can diagnose errors and provide solutions based on coding patterns and rules. A hint information generation unit (450) according to one embodiment can quickly present a solution (hint) by applying a predefined countermeasure for a specific type of error. A hint information generation unit (450) according to one embodiment can learn the user's coding style and error patterns using artificial intelligence technology and generate more accurate and personalized hints based on this (for example, it can generate error messages using natural language processing technology and generate hints that match the user's level of understanding). A hint information generation unit (450) according to one embodiment can evaluate the user's error after they have resolved it using the hint and determine whether additional support is needed (through this process, user feedback can be collected and the system can be continuously improved).

[0092] A vision recognition coding education robot (400) according to one embodiment may further include a motion control information generating unit that generates motion control information instructing the actual operation of the vision recognition coding education robot (400) based on the logic of a recognized coding block. That is, the motion control information generating unit (460) according to one embodiment may convert a programmed command into a physical operation so that the vision recognition coding education robot (400) can respond appropriately and operate according to the coding block assembled by the user.

[0093] Specifically, the motion control information generation unit (460) according to one embodiment can convert the coding logic information provided by the coding block recognition unit (410) into a motion command (motion sequence) that the vision recognition coding education robot (400) can understand and execute. At this time, the motion sequence determines the order and method of actions that the robot must perform according to the logic, and may include various physical actions such as movement, object manipulation, and sensor response of the vision recognition coding education robot (400) (for example, motion sequences such as 'forward movement', 'object picking', 'rotation').

[0094] A motion control information generation unit (460) according to one embodiment can optimize the generated motion command to be executed efficiently (such as reducing energy consumption or minimizing the time for performing the motion) and can perform a safety review on the motion plan so that the motion of the robot can be performed safely (including a safety review to prevent risks that may occur during the process of interacting with the surrounding environment as well as the safety of the robot itself).

[0095] A vision recognition coding education robot (400) according to one embodiment may further include a motion control information generation unit that generates motion control information instructing the actual operation of the vision recognition coding education robot (400) based on the logic of a recognized coding block. An object control unit (470) according to one embodiment may instruct the robot to perform an appropriate operation according to the motion control information generated by the motion control information generation unit (460) in order to directly control the actual operation of the vision recognition coding education robot (400) (to this end, various driving mechanisms of the robot are adjusted, and the robot's servo motors, actuators, etc. are controlled to enable the execution of various operations such as movement, rotation, object picking, and manipulation).

[0096] At this time, the object control unit (470) according to one embodiment can precisely control each part of the vision recognition coding education robot (400) using control algorithms such as PID (Proportional, Integral, Derivative) control, fuzzy logic control, and neural network-based control, and can use a real-time operating system (RTOS) to efficiently process multiple tasks at the same time and ensure an appropriate response time according to the system requirements.

[0098] FIG. 5 is a configuration diagram of a coding education platform (500) according to one embodiment of the present invention.

[0099] Referring to FIG. 5, a coding education platform (500) according to one embodiment of the present invention may include a data collection unit (510) that receives block coding education information from a robot, an artificial intelligence learning unit (520) that learns the received block coding education information and generates coding learning information, and a learning data generation unit that generates customized learning data for a new user who will use the robot. The coding education platform (500) according to one embodiment may be implemented in an on-premise manner or a cloud manner.

[0100] A data collection unit (510) of a coding education platform (500) according to one embodiment can collect block coding education information generated during a coding education session conducted through a vision recognition coding education robot (400).

[0101] Block coding education information may include at least one of information about a robot user, information on artificial intelligence coding logic used by the robot user, coding logic verification information, information on coding logic errors that occurred, information on generated error resolution hints, and information on the error resolution process. Specifically, block coding education information may include coding logic created by combining blocks by the user, user interaction, error occurrence information, successful execution information, etc. Additionally, a data collection unit (510) according to one embodiment may collect data such as coding logic created by combining blocks by the user, user input and interaction, user behavior during the session, robot reaction, error occurrence information, and successful execution results.

[0102] A data collection unit (510) according to one embodiment can monitor and record all events occurring during coding education in real time (for example, it may include information on how a user manipulates a specific block, what errors they experienced, and what approach they took to solve the problem). A data collection unit (510) according to one embodiment can transmit this information to other components of the coding education platform (500) to be used as data for analysis and processing.

[0103] A data collection unit (510) according to one embodiment can collect block coding education information through various sensors (the information thus collected can be processed by a logging system that records user interactions and robot reactions). At this time, the data collection unit (510) according to one embodiment can perform preprocessing processes such as data removal, data cleaning, and format conversion on the block coding education information. The data collection unit (510) according to one embodiment may include a database management system for effectively storing and managing the collected data (including functions for security and personal information protection).

[0104] According to one embodiment, the artificial intelligence learning unit (520) can identify the user's learning patterns, error types, success strategies, etc., using a machine learning or deep learning algorithm on block coding education information collected by the data collection unit (510), and analyze and learn the information to generate coding learning information. Through this, assistance can be obtained in improving coding education content and establishing learning support strategies.

[0105] An artificial intelligence learning unit (520) according to one embodiment can identify a user's learning style, error patterns, success strategies, etc. by analyzing block coding education information from a data collection unit (510). To this end, an artificial intelligence learning unit (520) according to one embodiment can generate a prediction model that extracts meaningful information from learning data using various machine learning and deep learning models.

[0106] A learning data generation unit (530) according to one embodiment can generate customized learning data for a new user or an existing user based on coding learning information generated by an artificial intelligence learning unit (520). The generated customized learning data may include educational materials, feedback mechanisms, coding challenges, etc., adjusted to the user's learning style and needs. For example, if a specific user has difficulty using loops, additional learning materials and practice problems on that topic can be provided to strengthen the skill. Alternatively, additional learning materials and practice problems on topics that multiple users repeatedly get wrong or struggle with can be provided.

[0107] In this way, the learning data generation unit (530) according to one embodiment can dynamically adjust the learning path according to the learning progress and performance. That is, it can allocate more resources to areas where the user is having difficulty and allow the user to proceed more quickly in areas where they have already mastered.

[0108] A learning data generation unit (530) according to one embodiment can continuously improve and optimize the generated learning data through continuous feedback from the user. To this end, the learning data generation unit (530) according to one embodiment can increase the efficiency of the learning process by integrating various educational tools such as gamification elements, interactive tutorials, and simulations.

[0109] Hereinafter, a block-based vision recognition coding education method according to an embodiment of the present invention is described based on the above description.

[0111] FIG. 6 is a block-based vision recognition coding education method according to an embodiment of the present invention, and is an example of a block-based vision recognition coding education method using the block-based vision recognition coding education system of FIG. 1 of the present invention.

[0112] Referring to FIG. 6, a block-based vision recognition coding education method according to one embodiment of the present invention recognizes the combined coding blocks (110) using a block recognition unit (120) when coding logic is formed by combining coding blocks (110) (S610). At this time, the coding logic formed through the combined coding blocks (110) can be recognized through the block recognition unit (120). Additionally, the block recognition unit (120) can recognize the combined coding blocks (110) using artificial intelligence-based computer vision technology.

[0113] At this time, the block recognition unit (120) can recognize at least one of the color of the coding block (110), the marker expressed on the coding block (110), the shape of the coding block (110), and the combination of the coding block (110). Additionally, the block recognition unit (120) can recognize at least one of the color of the coding block (110), the marker expressed on the coding block (110), and the shape of the coding block (110) using a deep learning-based image recognition model.

[0114] The object control unit (130) controls the operation of the object to be controlled (200) based on the coding logic recognized from the combination of recognized individual coding blocks (110-0, ..., 110-n) (S630). The object control unit (130) and / or the feedback analysis unit (140) receive environment data from the object to be controlled (200) (S640), perform an evaluation of the coding logic recognized based on the received environment data, and may display a feedback result including the analysis result of the performed evaluation through a display device such as a display and / or a speaker (S650).

[0116] As explained above, the present invention eliminates the need for electronic devices during block coding education and allows a robot to recognize and process block combinations using a camera, thereby increasing the degree of freedom in block combinations and enhancing usability in the educational setting, making coding education more effective and interesting. Furthermore, it can further increase the efficiency of coding education by providing user-customized learning data.

[0118] The above embodiments may be implemented using various forms of computing means including one or more processors, memory, and storage means. Additionally, a network interface connected to a wired or wireless network may be included. The processor may be a central processing unit or a semiconductor device that executes processing instructions stored in memory and / or storage units. The memory and storage units may include volatile storage media or non-volatile storage media. For example, the memory may include ROM and RAM. Accordingly, embodiments of the present invention may be implemented as a method implemented by a computer or as a non-transient computer-readable medium having computer-executable instructions stored on said computer. In one embodiment of the present invention, when executed by a processor, the computer-readable instructions may perform a method according to at least one aspect of the present invention.

[0119] Although the present invention has been described with reference to illustrated embodiments as above, these are merely exemplary, and it will be obvious to those skilled in the art that various modifications, changes, and equivalent alternative embodiments are possible without departing from the gist and scope of the present invention. For example, the block recognition unit (120) and the object control unit (130) may be implemented as a single integrated module or divided into two or more devices. Accordingly, the true technical scope of protection of the present invention should be determined by the technical concept of the appended claims. Explanation of the symbols

[0120] 100: Block-based vision recognition coding education system 110: Complete set of coding blocks 110-1: First Coding Block 110-n: nth coding block 120: Block recognition unit 130: Object Control Unit 140: Feedback Analysis Execution Unit 200: Controlled object 400: Vision Recognition Coding Education Robot 410: Coding block recognition unit 420: Coding logic generation section 430: Coding Logic Verification Unit 440: Coding result display section 450: Hint Information Generation Section 460: Operation control information generation unit 470: Object Control Unit 500: Coding Education Platform 510: Data Collection Unit 520: Artificial Intelligence Learning Unit 530: Training Data Generation Unit

Claims

Claim 1 A vision recognition coding education robot for coding education, comprising: a coding block recognition unit that recognizes coding blocks; a coding logic generation unit; a coding logic verification unit; and a coding result display unit. Claim 2 A vision recognition coding education robot according to claim 1, wherein the block recognition unit includes an artificial intelligence coding logic calling unit that calls artificial intelligence coding logic. Claim 3 A vision recognition coding education robot according to claim 1, wherein the coding result display unit displays a process of proceeding to an operation step when the verification of the coding logic by the coding logic verification unit is successful, and displays hint information for resolving coding logic errors when the verification of the coding logic by the coding logic verification unit fails. Claim 4 A vision recognition coding education robot according to claim 1, further comprising a hint information generating unit that generates hint information for resolving coding logic errors based on rules. Claim 5 A vision recognition coding education robot according to claim 1, further comprising a motion control information generation unit that generates motion control information of the robot based on the result of recognizing the coding logic of a coding block recognized by the coding block recognition unit. Claim 6 A vision recognition coding education robot according to claim 5, further comprising an object control unit that controls the operation of the robot based on the operation control information. Claim 7 A coding education platform comprising: a data collection unit that receives block coding education information from a robot; an artificial intelligence learning unit that learns the received block coding education information and generates coding learning information; and a learning data generation unit that generates customized learning data for a new user who will use the robot. Claim 8 A coding education platform according to claim 7, wherein the block coding education information comprises at least one of information about a robot user, information on an artificial intelligence coding logic used by the robot user, information on coding logic verification, information on a coding logic error that occurred, information on a generated error resolution hint, and information on an error resolution process. Claim 9 A coding education platform according to claim 7, wherein the artificial intelligence learning unit generates, through artificial intelligence modeling, focused learning information, which is information about the parts that the new user needs to focus on learning, based on the generated coding learning information. Claim 10 A coding education platform according to claim 9, wherein the learning data generation unit generates customized learning data for the new user based on the focus learning information and the information about the new user.

Images