A child programming education accompanying robot system, method and medium based on an emotional adaptive mechanism

Abstract

The present application relates to the technical field of robots, and discloses a child programming education accompanying robot system, method and medium based on an emotion adaptive mechanism, the child programming education accompanying robot system comprising: an accompanying robot body; a reconfigurable semantic programming building block group; a semantic issuing and checking module; a layout identification and program graph construction module; a safe compilation and error correction module; an emotion / engagement estimation and task adaptive arrangement module; an execution control and feedback expression module; wherein the modules are deployed on the robot body or a local terminal / edge device in communication with the robot. The present application improves the instruction expandability, safety and understandable error correction capability of the entity programming toy, and reduces frustration, improves learning persistence and accompanying experience through the emotion adaptive mechanism.

Description

Technical Field

[0001] This invention relates to the field of robotics technology, specifically to a children's programming education companion robot system, method, and medium based on an emotion-adaptive mechanism. Background Technology

[0002] Most existing children's programming educational toys fix instructions such as "forward / turn / loop / condition" onto solid blocks using fixed printing, fixed electronic tags, or fixed functional modules, resulting in:

[0003] 1. Limited instruction set size: Introducing more complex concepts (nested loops, conditional branches, parameterized actions, etc.) usually requires adding new types of building blocks or relying on on-screen programming interfaces, which leads to increased hardware costs and learning difficulties.

[0004] 2. Insufficient error correction and explanation capabilities: When children misplace the blocks in order / connection / parameters, the system struggles to provide interpretable error correction prompts that "point to specific blocks," thus affecting learning efficiency;

[0005] 3. Lack of safety constraints: If the program directly drives the robot's actions, risks such as collisions, boundary crossings, and high-speed movements may occur;

[0006] 4. Lack of an emotionally adaptive support mechanism: Children are prone to frustration and attention shifts during the learning process. Existing products have fixed prompting strategies and fixed level rhythms, making it difficult to maintain sustained engagement.

[0007] Therefore, there is an urgent need for a companion robot solution that can achieve scalable, interpretable, error-correctable, and secure execution of entity programming instructions without significantly increasing hardware complexity, and can dynamically adjust the difficulty and prompts based on engagement. Summary of the Invention

[0008] The purpose of this invention is to provide a children's programming education companion robot system, method, and medium based on an emotion-adaptive mechanism to solve the problems in the prior art.

[0009] The objective of this invention can be achieved through the following technical solutions:

[0010] A children's programming education companion robot system based on an emotion-adaptive mechanism, the children's programming education companion robot system includes: a companion robot body; a reconfigurable semantic programming block group; a semantic distribution and verification module; a layout recognition and program graph construction module; a secure compilation and error correction module; an emotion / engagement estimation and task adaptive orchestration module; and an execution control and feedback expression module;

[0011] The module is deployed on the robot body or on a local terminal / edge device that communicates with the robot.

[0012] Furthermore, the companion robot body includes a processor, memory, camera / depth sensor, microphone, speaker, expression screen / lighting assembly, and motion actuator;

[0013] Each block in the reconfigurable semantic programming block set contains at least a unique identifier (ID) and a refreshable display unit, and is equipped with near-field / short-range communication.

[0014] The semantic delivery and verification module generates a semantic mapping table BSMT and writes the semantic token and parameter field into the building block to complete CRC / signature verification;

[0015] The layout recognition and program diagram construction module identifies the block sequence and connection relationship based on image / base plate data, and constructs the program diagram PG.

[0016] The secure compilation and error correction module performs syntax / type / security constraint checks on PG and outputs action sequence A; when the constraint set Ω is not satisfied, it generates error correction prompt C and correction program PG'.

[0017] The emotion / engagement estimation and task adaptive orchestration module calculates engagement Et, updates difficulty Dt and prompting strategy, and determines the task and semantic reconfiguration for the next round.

[0018] The execution control and feedback expression module executes the action sequence and outputs voice / facial / action feedback, while simultaneously sending back operation monitoring data.

[0019] Furthermore, the method includes the following steps:

[0020] S101, Level initialization and block ID binding;

[0021] S102. Generate the semantic mapping table BSMT and send it to STP for verification;

[0022] S103. Collect the block layout and identify the order / connection;

[0023] S104, Construct the program diagram PG;

[0024] S105. Safely compile, perform constraint checks, and generate action sequence A;

[0025] S106, Error correction and solution of PG' and generation of interpretable hints C;

[0026] S107. Perform actions and conduct runtime monitoring;

[0027] S108, Calculate the level of engagement Et and update the difficulty Dt+1;

[0028] S109. Generate the next level task and trigger semantic reconfiguration.

[0029] Further, wherein S101 specifically includes:

[0030] S101-1, The robot body reads the currently available set of blocks {IDi}i=1N;

[0031] S101-2. Establish the "block-user-round" binding relationship: record the block ID, user identifier, and round number t of the participants in this round;

[0032] S101-3. Load the teaching objectives and target engagement E∗ for this round from the module, and initialize the difficulty Dt∈[Dmin,Dmax];

[0033] S102 specifically includes:

[0034] S102-1, The module determines the "instruction set size and complexity" for this round based on the current difficulty Dt;

[0035] S102-2, The module generates a record for each building block to form a BSMT;

[0036] S102-3. Encapsulate the module and send STPi={IDi,tokeni,parami,crci} to the corresponding block;

[0037] S102-4. After the CRCI verification at the building block end passes, refresh the display unit to show the current semantics; if the verification fails, request a resend or indicate an error.

[0038] Furthermore, S103 specifically includes:

[0039] S103-1, Module acquisition layout input;

[0040] S103-2. Identify the relationship between the block sequence and the connection ports, and output the connection set L={(IDu,portu,IDv,portv)};

[0041] S103-3. If a branch / loop structure exists, identify the entry / exit point and nesting level, and provide candidate completion for missing connections;

[0042] S104 specifically includes:

[0043] S104-1, Construct a node set V: Each block forms a node vi, and the node attributes are read from BSMT: (tokeni, parami, leveli);

[0044] S104-2, Constructing edge set E: Mapping connection set L to control flow edges;

[0045] S104-3. Obtain PG=(V,E), which can be stored using AST or adjacency list for subsequent compilation and constraint checking.

[0046] Furthermore, S105 specifically includes:

[0047] S105-1, The module performs syntax and type checks on PG;

[0048] S105-2. Compile PG into an action sequence A={ak}, where each ak carries the action type and parameters;

[0049] S105-3. Apply a set of safety constraints Ω to the action sequence. If all constraints are satisfied, proceed to S107; otherwise, proceed to S106.

[0050] S106 specifically includes:

[0051] S106-1 Solve the correction procedure PG' using the principle of minimum modification cost;

[0052] S106-2. Map the editing operation sequence to a set of error correction prompts C=Φ(PG→PG');

[0053] S106-3, Output Prompt:

[0054] Block level: Blocks that are highlighted / flashing and need to be replaced / reconnected;

[0055] At the voice level: provide explanations of "specific building blocks + specific parameters / connections";

[0056] Display aspect: Show the before and after comparison of the correction on the robot screen or the accompanying app;

[0057] S106-4, Wait for the child to correct the error and return to S103 or directly use PG' to enter S107;

[0058] S107 specifically includes:

[0059] S107-1, The module sends execution instructions to the robot body 10 actuator;

[0060] S107-2. Monitor safety indicators during operation and take emergency stop / downgrade if a risk is triggered.

[0061] S107-3. Synchronously output multimodal feedback and record execution logs.

[0062] Furthermore, S108 specifically includes:

[0063] S108-1, The module extracts the multimodal feature vector ft and calculates the engagement degree Et;

[0064] S108-2, Update the difficulty to Dt+1 and calculate the hint level;

[0065] S108-3. Compare Et with the threshold θ to provide transition conditions for the state machine TSM.

[0066] S109 specifically includes:

[0067] S109-1. If Et≥θ, TSM enters challenge mode: increases loop / condition level, expands instruction set; PS reduces prompt level.

[0068] S109-2. If Et < θ, TSM enters companion mode: reduces task complexity and instruction set size; PS increases prompt level and soothes;

[0069] S109-3 When the instruction set size changes or the semantics need to be changed, S102 is triggered to perform semantic reconfiguration and enter the next round.

[0070] A computer-readable storage medium for a children's programming education companion robot system based on an emotion-adaptive mechanism.

[0071] A method for using a children's programming education companion robot system based on an emotion-adaptive mechanism, wherein the method is executed by a processor.

[0072] The beneficial effects of this invention are:

[0073] 1. The present invention is a children's programming education companion robot system based on an emotion adaptive mechanism, which can reconstruct semantic programming blocks and their semantic reconfiguration links of "semantic mapping table BSMT - semantic token packet STP - CRC / signature verification - refresh display";

[0074] 2. The present invention is a children's programming education companion robot system based on an emotion adaptive mechanism, which is a closed loop of "block layout recognition → program graph PG construction → safe compilation → constraint check Ω → error correction and solution PG' → action sequence A execution";

[0075] 3. The present invention is a children's programming education companion robot system based on an emotion adaptive mechanism, wherein the calculation of the engagement index Et is based on the adaptive mechanism of its driven difficulty update Dt+1 and plot / hint strategy (TSM+PS).

[0076] 4. The present invention is a children's programming education companion robot system based on an emotion adaptive mechanism. The error correction prompt C and the interpretable mapping mechanism of block highlighting / blinking / voice prompts enable the companion robot to achieve scalable, interpretable error correction and safe execution of physical programming instructions, and can dynamically adjust the difficulty and prompts based on the level of engagement. Attached Figure Description

[0077] The invention will now be further described with reference to the accompanying drawings.

[0078] Figure 1 This is the overall block diagram of the system of the present invention;

[0079] Figure 2 This is a flowchart of the method of the present invention;

[0080] Figure 3 This is a schematic diagram of the data structure of the semantic mapping table BSMT, the semantic token packet STP, and the program graph PG of this invention;

[0081] Figure 4 This is a schematic diagram of the story state machine (TSM) driven by immersion and the prompting level strategy (PS) of this invention.

[0082] Explanation of reference numerals in the attached figures:

[0083] 10—Companion robot body; 20—Reconfigurable semantic programming blocks; 30—Semantic delivery and verification module; 40—Layout recognition and program graph construction module; 50—Safe compilation and error correction module; 60—Emotion / engagement estimation and task adaptive orchestration module; 70—Execution control and feedback expression module. Detailed Implementation

[0084] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0085] Example:

[0086] A children's programming education companion robot system based on an emotion adaptive mechanism, according to Figure 1 As shown, the system includes:

[0087] Companion robot body 10: includes processor, memory, camera / depth sensor, microphone, speaker, expression screen / lighting assembly, and motion actuator;

[0088] Reconfigurable semantic programming block group 20: Each block contains at least a unique identifier ID and a refreshable display unit, and has near-field / short-range communication capabilities;

[0089] Semantic delivery and verification module 30: generates a semantic mapping table BSMT, writes the semantic token and parameter field into the building block, and completes CRC / signature verification;

[0090] Layout recognition and program diagram construction module 40: Based on image / base plate data, it identifies the block order and connection relationship, and constructs the program diagram PG;

[0091] Safe compilation and error correction module 50: Performs syntax / type / safety constraint checks on PG and outputs action sequence A; generates error correction prompt C and corrector PG' when constraint set Ω is not satisfied;

[0092] Emotion / engagement estimation and task adaptive orchestration module 60: Calculate engagement Et, update difficulty Dt and prompting strategy, and determine the task and semantic reconfiguration for the next round;

[0093] Execution control and feedback expression module 70: Executes action sequences and outputs voice / facial / action feedback, while simultaneously transmitting back operation monitoring data.

[0094] Modules 30-70 can be deployed entirely on the robot body 10, or partially on local terminals / edge devices that communicate with the robot.

[0095] like Figure 3 As shown, this invention introduces a "Semantic Map Table (BSMT)" and a "Semantic Token Packet (STP)" to allow the same physical block to carry different instruction meanings in different rounds / levels.

[0096] The Semantic Mapping Table (BSMT) is a mapping table maintained by the robot, where each record corresponds to a specific block ID.

[0097] BSMT={(IDi,tokeni,parami,stylei,crci)}i=1N

[0098] in:

[0099] IDi: The unique identifier of the i-th block;

[0100] tokeni: Semantic token (instruction opcode / instruction category code);

[0101] parami: The parameter domain (such as the range of values ​​or the set of default values ​​for steps, angle, speed, number of loops, condition thresholds, etc.);

[0102] stylei: Display style (icons / text / colors / blinking strategies, etc., used to present semantics in block display units);

[0103] crci: Validation field (cyclic redundancy check value or signature digest).

[0104] Semantic Token Packet (STPi): A packaged data packet distributed to the building blocks, such as... Figure 3 As shown, it includes at least:

[0105] STPi={IDi,tokeni,parami,crci}

[0106] After receiving the data, the module verifies the CRCI. If it passes the verification, it refreshes and displays the visual semantics corresponding to the stylei (the style can be sent from the robot or generated on the module based on the token mapping, which is an optional implementation).

[0107] like Figure 2 , Figure 3 As shown:

[0108] (1) Module 40 identifies from inputs such as camera images / base plate contact matrix:

[0109] Block sequence (execution in sequence);

[0110] Connection relationships (control flow connections, branch inlet / outlet, nested loop levels);

[0111] (2) Construct a flowchart from the recognition results:

[0112] PG=(V,E)

[0113] Where: V: a set of nodes, each node corresponds to a block, and the node attributes include (token, param, level);

[0114] E: Set of control flow edges, representing control relationships such as sequence, branching, and looping (can be stored using an adjacency list or AST);

[0115] (3) Module 50 performs the following on PG: syntax checking, type checking, and safety constraint checking, generating an action sequence:

[0116] A = {a1, a2, ..., aK}

[0117] Where ak is the k-th atomic action (forward, turn, stay, play voice, light expression, etc.), and carries corresponding parameters:

[0118] (4) If the set of safety constraints Ω is not satisfied, a correction procedure and error correction prompts are generated by solving the constraints (specifically, steps S105-S106 below).

[0119] like Figure 4 As shown, this invention organizes tasks using a round-based teaching method. At the end of each round, a state transition is performed based on the engagement level Et and the threshold θ.

[0120] If Et≥θ: Determine high investment, enter challenge mode, increase difficulty and reduce prompt level;

[0121] If Et < θ: Determine low input / frustration, enter companion mode, reduce difficulty and increase prompt level, and trigger semantic reconfiguration to reduce instruction burden.

[0122] Among them, the Story State Machine (TSM) is used to control states such as "challenge / accompany", and the Prompt Strategy (PS) is used to control the intensity of prompts (e.g., progressively increasing from "abstract prompts → providing partial prompts → directly pointing to specific blocks and parameters").

[0123] Explanation of symbols and abbreviations

[0124] t (round / time index); N (number of blocks); IDi (unique identifier of the i-th block); tokeni (semantic token / opcode); parami (parameter field); stylei (display style); crci (verification field, CRC / signature digest); STPi (semantic token packet of the i-th block); BSMT (semantic mapping table); PG=(V,E) program graph; V is the set of nodes, E is the set of control flow edges; A={ak} action sequence; ak is the k-th atomic action; Ω (set of security constraints); ft (multimodal feature vector of round t); Et (engagement index); Dt (difficulty parameter, higher difficulty indicates higher task complexity); E∗ (target engagement, the desired level of engagement); θ (engagement threshold, used for...) Figure 4 Branches); α difficulty update step size; clip(⋅) truncation operator; TSM situation / task state machine; PS hint hierarchy strategy; C error correction hint set (can point to specific blocks / parameters / connections).

[0125] The engagement index is calculated by assuming the multimodal feature vector extracted within round t is:

[0126] ft=vt,at,rt,pt⊤

[0127] in:

[0128] vt: Speech prosodic features (e.g., aggregated values ​​of energy, speech rate, pitch fluctuations, pause proportions, etc.);

[0129] at: Facial / postural emotional characteristics (e.g., numerical results of pleasure, frustration, attention shift, etc.);

[0130] rt: Interaction rhythm characteristics (e.g., frequency of block placement, pause duration, number of repeated attempts, etc.);

[0131] pt: Task status characteristics (e.g., completion progress, number of errors, number of corrections, timeout flag, etc.).

[0132] The engagement index is defined as:

[0133] Et=σ(w⊤ft+b) (1)

[0134] where: w: weight vector;

[0135] b: bias;

[0136] σ(x)=1 / (1+e−x): Sigmoid function, which makes Et∈(0,1).

[0137] Reasoning explanation: Sigmoid maps the linear fusion result to an engagement characterization in the 0-1 interval, which is convenient for comparison with the threshold θ and used for difficulty update.

[0138] where, for adaptive difficulty update, let the target engagement be E∗∈(0,1), and the difficulty parameter is updated as:

[0139] Dt+1=clip(Dt+α(Et−E∗), Dmin, Dmax) (2)

[0140] where: Dt: the difficulty at round t; α>0: update step size; Dmin, Dmax: the upper and lower bounds of difficulty; clip(x,a,b)=min(max(x,a),b).

[0141] Reasoning explanation:

[0142] If Et<E∗, then (Et−E∗)<0, so that Dt+1<Dt, realizing "low engagement / frustration → difficulty reduction";

[0143] If Et>E∗, then Dt+1>Dt, realizing "high engagement → difficulty increase".

[0144] where, for the safety constraint set Ω and inspection, a safety constraint set Ω is set for the action sequence A={ak}k=1K, which includes constraints such as speed, acceleration, obstacle avoidance distance, and workspace boundary, for example:

[0145] Ω: {speed(ak)≤Vmax, acc(ak)≤Amax, distToObstacle(ak)≥dmin, pos(ak)∈W} (3)

[0146] where: speed(ak), acc(ak): action speed / acceleration;

[0147] Vmax, Amax: upper limits of speed / acceleration;

[0148] distToObstacle(ak): the minimum distance to the obstacle when executing this action;

[0149] dmin: lower limit of safety distance;

[0150] pos(ak): Action execution position / trajectory point;

[0151] W: Permitted workspace / safe area.

[0152] In the error correction and suggestion generation process, when Ω is not satisfied, the corrected procedure graph PG' is obtained through the principle of minimum modification cost.

[0153] PG'=argminPG~cost_edit(PG,PG~)st Ω(PG~) is established (4)

[0154] Where cost_edit(⋅) represents editing the cost function, which can be weighted and summed according to operations such as "replace / insert / delete / reconnect / parameter adjustment":

[0155] cost_edit(PG,PG~)=λ1nrep+λ2nins+λ3ndel+λ4nlink+λ5∑|Δparam|(5)

[0156] in:

[0157] nrep, nins, ndel, and nlink represent the number of replacement, insertion, deletion, and reconnection operations, respectively.

[0158] Δparam represents the parameter adjustment amount;

[0159] λ1~λ5 are the weights for various operations.

[0160] The set of error correction suggestions C can be defined as a sequence mapping of edit operations from PG to PG':

[0161] C=Φ(PG→PG')

[0162] Φ(⋅) is the prompt generation function, which outputs interpretable prompts (such as "Change the block with ID=⋯ from token=loop to token=sequential"; "Change the angle parameter from 90° to 45°"; "Connect the branch exit to port ⋯", etc.), and can be linked with block highlighting / flashing and voice prompts.

[0163] In this embodiment, a "turn-based teaching level" is used as an example. It is executed once per turn. Figure 2 The process shown; Figure 1 Demonstrates the collaboration between various modules; Figure 3 Explain semantic reconfiguration and program graph structure; Figure 4 This explains engagement-driven mode switching and semantic reconfiguration.

[0164] S101, Level Initialization and Block ID Binding (see...) Figure 2 )

[0165] S101-1, Robot body 10 reads the current available block set {IDi}i=1N (which can be obtained through communication handshake or base plate scanning);

[0166] S101-2. Establish the "block-user-round" binding relationship: record the block ID, user identifier, and round number t of the participants in this round;

[0167] S101-3. Load the teaching objectives (sequential / cyclic / conditional / parameterized actions, etc.) and target engagement E∗ of this round from module 60, and initialize the difficulty Dt∈[Dmin,Dmax].

[0168] S102. Generate the semantic mapping table BSMT and issue the STP verification (see...). Figure 2 , Figure 3 )

[0169] S102-1, Module 60 determines the "instruction set size and complexity" for this round based on the current difficulty Dt, for example:

[0170] Low Dt: Only sequential actions and a few parameters are enabled;

[0171] High Dt: Enable loops / conditions and increase nesting levels.

[0172] S102-2 and Module 30 generate a record (IDi, tokeni, parami, stylei, crci) for each block, forming a BSMT.

[0173] S102-3, Module 30 encapsulates and sends STPi={IDi,tokeni,parami,crci} to the corresponding building block;

[0174] S102-4. After the block-side verification of CRCI (or signature digest) passes, refresh the display unit to show the current semantics; if the verification fails, request a resend or indicate an error.

[0175] S103, Collect block layout and identify sequence / connection (see...) Figure 2 )

[0176] S103-1, Module 40 Acquisition Layout Input (Image / Base Plate Contact / Magnetic Encoding, etc.);

[0177] S103-2. Identify the relationship between the block sequence and the connection ports, and output the connection set L={(IDu,portu,IDv,portv)};

[0178] S103-3. If a branch / loop structure exists, identify the entry / exit and nesting level, and provide candidate completion for missing connections (optional).

[0179] S104, Construct the program diagram PG (see...) Figure 2 , Figure 3 )

[0180] S104-1, Construct a node set V: Each block forms a node vi, and the node attributes are read from BSMT: (tokeni, parami, leveli);

[0181] S104-2, Constructing edge set E: Mapping connection set L to control flow edges (sequence / branch / loop).

[0182] S104-3. Obtain PG = (V,E), which can be stored using AST or adjacency list for subsequent compilation and constraint checking.

[0183] S105, Secure compilation, constraint checking, and generation of action sequence A (see...) Figure 2 )

[0184] S105-1, Module 50 performs syntax checks (structure validity) and type checks (parameter type / range validity) on PG.

[0185] S105-2. Compile PG into an action sequence A={ak}, where each ak carries the action type and parameters;

[0186] S105-3. Apply a set of safety constraints Ω to the action sequence. If all constraints are satisfied, proceed to S107; otherwise, proceed to S106.

[0187] S106, Error Correction Solving PG' and Generating Interpretable Hints C (see...) Figure 2 )

[0188] S106-1 Solve the correction procedure PG' using the principle of minimum modification cost;

[0189] S106-2. Map the editing operation sequence to a set of error correction prompts C=Φ(PG→PG');

[0190] S106-3, Output Prompt:

[0191] Block level: Blocks that are highlighted / flashing and need to be replaced / reconnected;

[0192] At the voice level: provide explanations of "specific building blocks + specific parameters / connections";

[0193] (Optional) Display: Show a comparison before and after the correction on the robot screen or the accompanying app.

[0194] S106-4. Wait for the child to correct the error and then return to S103 or directly use PG' to enter S107 (both can be used as implementation methods).

[0195] S107, Perform actions and runtime monitoring (see...) Figure 2 , Figure 1 )

[0196] S107-1, Module 70 sends execution instructions to the actuator of the robot body 10;

[0197] S107-2. Monitor safety indicators (speed, distance, boundary crossing, collision, etc.) during operation, and stop or downgrade immediately if a risk is triggered.

[0198] S107-3. Synchronously output multimodal feedback (voice encouragement, facial light effects, action demonstration) and record execution log (optional evidence storage).

[0199] S108, Calculate the input level Et and update the difficulty Dt+1 (see...) Figure 2 , Figure 4 )

[0200] S108-1, Module 60 extracts the multimodal feature vector ft, and calculates the engagement degree Et according to formula (1);

[0201] S108-2. Update the difficulty Dt+1 according to formula (2) and calculate the hint level (for example, the hint level Lt is inversely related to the difficulty).

[0202] S108-3. Compare Et with the threshold θ to obtain... Figure 4 The state machine TSM shown provides the transition conditions.

[0203] S109. Generate the next level's task and trigger semantic reconfiguration (see...) Figure 2 , Figure 4 )

[0204] S109-1. If Et≥θ, TSM enters challenge mode: increases loop / condition level, expands instruction set; PS reduces prompt level.

[0205] S109-2. If Et < θ, TSM enters companion mode: reduces task complexity and instruction set size; PS increases prompt level and soothes;

[0206] S109-3 When the instruction set size changes or the semantics need to be changed, S102 is triggered to perform semantic reconfiguration (BSMT / ST new) and enter the next round.

[0207] 3.4 Advantages of the present invention

[0208] (1) Extensible instruction set: Through BSMT and STP mechanisms, the semantics of the same set of building blocks can be reconfigured in different levels, enabling course upgrades without increasing the types of building blocks;

[0209] (2) Explainable error correction: Through PG construction and minimum modification cost error correction, the prompts directly point to "which block / which parameter / which connection" needs to be modified;

[0210] (3) Safety and controllability: Static checks and runtime monitoring are performed through constraint set Ω to reduce the risks of toy scenarios such as collisions, boundary crossings, and high speeds;

[0211] (4) Emotion-adaptive companionship closed loop: Based on the level of engagement (Et), the difficulty (Dt), the prompt level, and the switching of the plot state are driven to reduce frustration and enhance the continuous learning and companionship experience.

[0212] Alternative solution to this embodiment:

[0213] Communication methods: NFC, RFID, BLE, infrared, contact base plate, etc.

[0214] Display methods: electronic paper, dot matrix screen, LED light array, color-changing materials, etc.;

[0215] Layout recognition: visual recognition, magnetic coding, base plate contact matrix, UWB / IMU assistance, etc.;

[0216] Engagement estimation: It can use only voice + interaction rhythm, or add wearable signals such as heart rate / stress (optional);

[0217] Error correction strategies: rule base, search planning, and learning-based error correction based on historical interactions;

[0218] Security constraints: can be expanded to include time budget, energy budget, content security (voice / text), etc.

[0219] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention.

Claims

1. A children's programming education companion robot system based on an emotion adaptive mechanism, characterized in that, The children's programming education companion robot system includes: a companion robot body (10); a reconfigurable semantic programming block group (20); a semantic distribution and verification module (30); a layout recognition and program graph construction module (40); a secure compilation and error correction module (50); an emotion / engagement estimation and task adaptive orchestration module (60); and an execution control and feedback expression module (70). All modules are deployed on the robot body (10) or on a local terminal / edge device that communicates with the robot; The companion robot body (10) includes a processor, memory, camera / depth sensor, microphone, speaker, expression screen / lighting assembly, and motion actuator; Each block in the reconfigurable semantic programming block group (20) contains at least a unique identifier ID and a refreshable display unit, and has near-field / short-range communication capabilities; The semantic delivery and verification module (30) generates a semantic mapping table BSMT and writes the semantic token and parameter field into the building block to complete the CRC / signature verification; The layout recognition and program diagram construction module (40) recognizes the block order and connection relationship based on image / base plate data and constructs the program diagram PG; The secure compilation and error correction module (50) performs syntax / type / security constraint checks on PG and outputs action sequence A; when the constraint set If the conditions are not met, an error message C and a correction procedure PG' are generated; The emotion / engagement estimation and task adaptive orchestration module (60) calculates engagement. Update difficulty The prompting strategy determines the next round's task and semantic reconfiguration; The execution control and feedback expression module (70) executes the action sequence and outputs voice / expression / action feedback, while simultaneously sending back operation monitoring data.

2. The method of using a children's programming education companion robot system based on an emotion adaptive mechanism according to claim 1, characterized in that, The method includes the following steps: S101, Level initialization and block ID binding; S102. Generate the semantic mapping table BSMT and send it to STP for verification; S103. Collect the block layout and identify the order / connection; S104, Constructing the Program Graph ; S105. Safely compile, perform constraint checks, and generate action sequence A; S106, Error Correction and Solution With generating interpretable hints ; S107. Perform actions and conduct runtime monitoring; S108, Calculate the level of commitment And update the difficulty ; S109. Generate the next level task and trigger semantic reconfiguration.

3. The method of using a children's programming education companion robot system based on an emotion adaptive mechanism according to claim 2, characterized in that, in, S101 specifically includes: S101-1, Companion Robot Body (10) Reads the currently available block set ; S101-2. Establish the "Block-User-Round" binding relationship: Record the blocks participating in this round. User ID, Round Number ; S101-3, Load the teaching objectives and target engagement for this round from the emotion / engagement estimation and task adaptive orchestration module (60). and initialize the difficulty ; S102 specifically includes: S102-1, Emotion / Enthusiasm Estimation and Task Adaptive Orchestration Module (60) based on current difficulty Determine the "instruction set size and complexity" for this round; S102-2, The semantic delivery and verification module (30) generates a record for each block to form a BSMT; S102-3, Semantic delivery and verification module (30) is encapsulated and delivered. To the corresponding block; S102-4, Block End Verification If the verification passes, refresh the display unit to show the current semantics; if the verification fails, request a resend or indicate an error.

4. The method of using a children's programming education companion robot system based on an emotion adaptive mechanism according to claim 3, characterized in that, S103 specifically includes: S103-1, Layout recognition and program graph construction module (40) collects layout input; S103-2, Identify the relationship between the block sequence and the connection ports, and output the connection set. ; S103-3. If a branch / loop structure exists, identify the entry / exit point and nesting level, and provide candidate completion for missing connections; S104 specifically includes: S104-1, Constructing a Node Set Each block forms a node. Node attribute reading BSMT: ; S104-2, Constructing the Edge Set : Connect the set Mapped to control flow edges; S104-3, obtained It can be stored using AST or adjacency list for subsequent compilation and constraint checking.

5. The method of using a children's programming education companion robot system based on an emotion adaptive mechanism according to claim 4, characterized in that, Specifically, S105 includes: S105-1, Secure Compilation and Error Correction Module (50) Perform syntax checking and type checking; S105-2, Compile PG into an action sequence Each Carry action type and parameters; S105-3, Applying a set of safety constraints to the action sequence If all conditions are met, proceed to S107; if any violation occurs, proceed to S106. S106 specifically includes: S106-1 Solving the correction procedure using the principle of minimum modification cost ; S106-2. Map the editing operation sequence to a set of error correction prompts. ; S106-3, Output Prompt: Block level: Blocks that are highlighted / flashing and need to be replaced / reconnected; At the voice level: provide explanations of "specific building blocks + specific parameters / connections"; Display aspect: Show the before and after comparison of the correction on the robot screen or the accompanying app; S106-4, Wait for the child to correct it, then return to S103 or use it directly. Enter S107; Specifically, S107 includes: S107-1, The execution control and feedback expression module (70) sends execution instructions to the execution mechanism of the robot body 10; S107-2. Monitor safety indicators during operation and take emergency stop / downgrade if a risk is triggered. S107-3. Synchronously output multimodal feedback and record execution logs.

6. The method of using a children's programming education companion robot system based on an emotion adaptive mechanism according to claim 5, characterized in that, S108 specifically includes: S108-1, Emotion / Entirety Estimation and Task Adaptive Orchestration Module (60) Extracts Multimodal Feature Vectors Calculate the degree of investment ; S108-2, Update Difficulty And calculate the warning level; S108-3, will With threshold Comparison provides transition conditions for the state machine TSM; S109 specifically includes: S109-1, if TSM enters challenge mode: increases loop / condition levels and expands instruction set; PS reduces prompt level; S109-2, if TSM enters companion mode: reducing task complexity and instruction set size; PS increases prompt level and provides reassurance. S109-3 When the instruction set size changes or the semantics need to be changed, S102 is triggered to perform semantic reconfiguration and enter the next round.

7. A computer-readable storage medium for a children's programming education companion robot system based on an emotion-adaptive mechanism as described in claim 1 or 2.

Images