Control method for simulating biological social behavior of ai pet hardware

Abstract

The application discloses a control method for simulating biological social behaviors of AI pet hardware, characterized in that a finite state machine architecture is adopted to schedule behaviors of the device, the finite state machine comprising: S0 independent state, an independent standby state when no same device is detected; S1 sensing state, an attention state after a same device signal is detected; S2 excitement state, a response state in which the degree of excitement gradually changes with the distance when the device is close; S3 pairing state, a state in which a pairing process is triggered through near field communication, and device connection is established; S4 synchronization state, a state in which synchronous interaction is performed after pairing is successful; and S5 separation state, a state in which phased emotional expression is performed when the device is far away. The conversion between the states is jointly determined by the wireless signal strength, the near field communication trigger and the duration condition, forming a complete biological social behavior chain covering independence, sensing, excitement, pairing, synchronization and separation, and significantly improving the interaction authenticity.

Description

Technical Field

[0001] This invention relates to the fields of intelligent toys, embedded system state machine design, biomimetic behavior modeling, and IoT device interaction technology. Specifically, it relates to a control method for simulating biological social behavior in AI pet hardware, mapping the complete process of biological social interaction to hardware state transitions. Background Technology

[0002] Most existing smart pet toys adopt a simple trigger-response model, lacking a complete social behavior chain. When multiple devices approach each other, there is either no interaction or only a preset simple response, failing to simulate the complete behavioral chain of real biological social interaction, from discovery, approach, relationship establishment to separation. This results in a monotonous and rigid product interaction experience, failing to create a realistic "biological social" experience.

[0003] Currently, smart pet hardware has accumulated some technical expertise in multi-device interaction, with typical solutions including:

[0004] Patent document 1 (CN105854305A, Toy and Interaction Method of Toy) establishes a connection between toys through NFC touch and maintains a communication list. Its technical solution mainly relies on NFC near-field communication, which requires physical contact between devices to trigger interaction, and the interaction content is limited to data transmission, lacking simulation at the behavioral level.

[0005] Patent document 2 (US12290050B2, Smart Pet Ecosystem) proposes a rule-based interaction triggering mechanism between multiple devices. Its technical solution uses a rule engine to drive device behavior, but the rules are preset fixed conditions-action mappings, lacking gradual behavioral changes and emotional expressions.

[0006] Patent document 3 (US9914062B1, NFC plush toy) implements the playback of preset sound effects after an NFC plush toy is touched. Its technical solution is extremely simple, containing only two states: trigger and play, without any behavior chain or state transition logic.

[0007] Patent document 4 (CN121868880A, A multimodal emotion recognition toy robot interaction system and method) proposes a multimodal emotion recognition interactive toy that identifies user emotions through multiple sensors. Its technical solution focuses on the interaction between a single device and the user, rather than social behavior between devices.

[0008] As can be seen from the above solutions, existing smart pet hardware generally suffers from the following shortcomings in realizing social behavior between devices:

[0009] 1. Incomplete social behavior chain: Most solutions only include two stages, "discovery" and "connection", lacking a complete behavioral loop from perception, excitement, pairing, synchronization to separation.

[0010] 2. Lack of progressiveness in behavioral expression: Device behavior typically adopts a binary triggering mode (no response / response), and cannot express different levels of excitement based on the distance between devices or signal strength.

[0011] 3. Abrupt state transitions and lack of transition: State transitions are mostly abrupt, lacking transition animations or gradual behavior changes, resulting in a mechanical interactive experience and a lack of biological realism.

[0012] 4. Lack of separation behavior: After the device is disconnected, it usually returns directly to the initial state, which cannot simulate the emotional expressions of "searching" and "reluctance" of living organisms during the separation process.

[0013] 5. Insufficient anti-interference capability: In scenarios with large fluctuations in wireless signal strength (RSSI), the state changes frequently, causing behavior jitter.

[0014] 6. Weak anomaly handling mechanism: There is a lack of systematic response strategies for abnormal scenarios such as simultaneous entry of multiple devices, pairing interruption, and low battery.

[0015] Therefore, there is an urgent need for an implementation scheme that can map the complete biological social behavior chain into a hardware state machine and support progressive behavior changes, emotional expression and stable state control, so as to improve the interactivity and user experience of AI pet hardware. Summary of the Invention

[0016] The purpose of this invention is to provide a method for controlling simulated biological social behaviors in AI pet hardware, aiming to solve the following core technical problems:

[0017] 1. How to map the complete biological social behavior chain into a hardware state machine, including the entire process of perception, excitement, pairing, synchronization, and separation.

[0018] 2. How to implement excitation gradient mapping based on wireless signal strength so that behavior changes gradually with distance.

[0019] 3. How to design separation behaviors with emotional expression, including the three stages of searching, reluctance, and acceptance.

[0020] 4. How to use anti-jitter mechanisms to avoid frequent state jumps and ensure smooth behavior transitions.

[0021] 5. How to handle abnormal scenarios such as multiple devices entering simultaneously or pairing interruptions.

[0022] The technical solution of this invention is:

[0023] A control method for simulating biological social behavior in AI pet hardware employs a finite state machine architecture to schedule device behavior. The finite state machine includes six states:

[0024] S0 Independent State: Independent standby state when no similar device is detected;

[0025] S1 Sensing Status: Attention status after detecting a signal from a similar device;

[0026] S2 Excitement State: The response state in which the level of excitement gradually changes with distance when the device is brought close;

[0027] S3 Pairing Status: The state in which a device connection is established by triggering the pairing process through near-field communication;

[0028] S4 Synchronization Status: The state of synchronization interaction after successful pairing;

[0029] S5 Separation State: The state in which phased emotional expression is performed when the device is far away;

[0030] The transitions between states are determined by the strength of the wireless signal, the triggering of near-field communication, and the duration of the communication, forming a complete biological social behavior chain covering independence, perception, excitement, pairing, synchronization, and separation.

[0031] Preferably, the S2 excitation state is configured with a three-level excitation gradient mapping mechanism:

[0032] The Received Signal Strength Indicator (RSSI) is divided into three intervals, corresponding to three excitement levels: Micro, ZTE, and High. Each level is matched with different head movements, lighting effects, and sound effects.

[0033] A delayed fallback mechanism is also set up: when the excitement level is upgraded, there is no delay in responding to the change in the RSSI threshold; when the level is downgraded, the switch is only executed after the RSSI falls back to the corresponding range for a certain period of time, so as to avoid frequent level jumps.

[0034] Preferably, the S5 separation state is configured with a three-stage separation behavior:

[0035] The first stage is the search stage: within a few seconds after the device moves away, it will quickly turn its head left and right, flash its lights quickly, and make a short beeping sound at the same frequency as the S4 state;

[0036] The second stage is the reluctant stage: within a few seconds, the head movement slows down and slowly lowers, the brightness of the lights gradually dims, and the output amplitude decreases linearly, accompanied by a low sigh.

[0037] The third stage is the acceptance stage: quiet behavior, faint breathing light, and silence that restore the S0 independent state within a few seconds;

[0038] The range of motion during the reluctant phase Attenuation is achieved using a linear interpolation algorithm:

[0039] ;

[0040] in, The amplitude of the action in the S4 synchronization state. To find the duration of a stage, t represents the current time.

[0041] Preferably, the S5 separation state further includes a reunion interruption mechanism:

[0042] If a device of the same type re-enters the BLE detection range during the separation process, the current separation process is interrupted, and the process jumps directly to the S2 excitation state to execute the reunion response behavior.

[0043] Preferably, a global anti-shake mechanism is also included:

[0044] A dual-threshold design is adopted, with the entry threshold and exit threshold for the same state transition having an interval of ≥10dBm, forming a hysteresis interval;

[0045] All state transitions must be triggered for a specified duration before the transition is executed;

[0046] The S3 pairing status is set with a minimum dwell time to prevent repeated triggering in a short period of time.

[0047] Preferably, it also includes a multi-device interaction processing mechanism:

[0048] When multiple similar devices enter the detection range at the same time, priority is given to establishing interaction with the device with the strongest RSSI signal.

[0049] Preferably, in the S1 sensing state and the S2 excitement state, if the device that triggers near-field communication is a paired device, the S3 pairing process is skipped, and the process directly jumps to the S4 synchronization state and executes the reunion animation.

[0050] Preferably, the S3 pairing state sequentially executes the following four-stage process:

[0051] Surprise phase: Near-field communication triggers pairing interruption, outputting full-body micro-vibration feedback;

[0052] Exchange phase: Exchange device ID, encrypted public key and feature information through near field communication, and output fast light flashing feedback;

[0053] Authentication phase: Perform two-way authentication using the shared key, and output slow pulsed light feedback;

[0054] Setup phase: Add the other party to the friend list and establish a BLE encrypted connection. After outputting the happy beating feedback, enter the S4 synchronization state.

[0055] If pairing fails, a disappointment animation is output, and the system returns to the S2 excited state after a cooldown.

[0056] Preferably, the S4 synchronization state is divided into three behavior patterns according to the emotional value:

[0057] When the emotional value > N1, the same-frequency beating mode is executed, and fast-paced actions and breathing fluctuations are output;

[0058] When the emotional value is in the range of N2–N1, the breathing accompanying mode is executed, and the breathing synchronization actions with a slowed-down rhythm are output;

[0059] When the emotional value < N2, the interactive exploration mode is executed, and independent and minute exploration actions are output while maintaining the connection;

[0060] The emotional value is a quantified value of 0-100 obtained by at least weighted normalization of the distance between devices, the pairing duration, and the number of interactions, where 0 < N2 < N1 < 100.

[0061] An AI pet hardware includes a processor, a memory, a wireless communication module, a near-field communication module, and an actuator, and is characterized in that the memory stores a computer program, and when the program is executed by the processor, the control method for simulating biological social behaviors as described above is implemented.

[0062] Beneficial effects

[0063] Compared with the prior art, the present invention has the following remarkable beneficial effects:

[0064] 1. A complete biological social behavior chain is constructed, significantly improving the interaction fidelity

[0065] The present invention maps the complete process of "independent → perception → excitement → pairing → synchronization → separation" in biological social interaction to a six-state finite state machine, breaking through the limitation of the existing solutions that only support the binary interaction of "discovery - connection", and for the first time realizing the closed-loop social behavior simulation in AI pet hardware, making the device interaction more in line with the behavior logic of real organisms.

[0066] 2. Realize the progressive expression of the excitement level and avoid sudden behavior changes

[0067] By setting three levels of excitement gradients and precisely mapping the RSSI signal strength intervals to different behavior patterns, the device behavior can continuously transition with the change of distance; combined with the hysteresis and fallback mechanism, it effectively avoids the frequent jumping of the excitement level at the threshold boundary, and improves the continuity and stability of the behavior performance.

[0068] 3. Introduce an emotional separation behavior to enhance the emotional companionship experience

[0069] The device is designed with a three-stage behavior of "searching - reluctance - acceptance" in the separation state, and a linear interpolation algorithm is used to achieve a smooth decay of the action amplitude and volume. This allows the device to still show vivid and natural emotional response after disconnection, solving the problem of discontinuous behavior and stiff experience in the separation stage of existing products.

[0070] 4. Multiple anti-shake mechanisms ensure stability during state transitions.

[0071] By employing a dual-threshold design, time-based verification, and minimum state dwell time constraints, false triggering caused by RSSI fluctuations is suppressed from three dimensions: signal layer, time layer, and state layer. This significantly reduces the risk of frequent state transitions and improves the system's robustness in complex wireless environments.

[0072] 5. Supports reunion and multi-device priority management, expanding application scenarios.

[0073] The system supports a reunion interruption mechanism during separation, enabling devices to naturally express emotional changes upon reunion. Simultaneously, through RSSI intensity sorting and candidate list management, it achieves orderly interaction in multi-device concurrent scenarios, avoiding resource conflicts and behavioral chaos.

[0074] 6. Covers multiple anomalies, improving system robustness and user experience.

[0075] Existing technical solutions are mostly designed for single scenarios, and often fail to respond or behave erratically when faced with abnormal situations such as interference from multiple devices, pairing interruptions, low battery, or repeated touches. This invention achieves full coverage of the above-mentioned abnormal scenarios through a preset anomaly handling strategy, ensuring that the device can maintain stable and natural behavior output even in complex interactive environments, and avoiding interaction interruptions or user experience disruptions caused by anomalies. Attached Figure Description

[0076] The present invention will be further described below with reference to the accompanying drawings and embodiments:

[0077] Figure 1 This is a state machine transition flowchart for simulating biological social behavior according to the present invention. Detailed Implementation

[0078] Example 1

[0079] like Figure 1 As shown, the control method for simulating biological social behavior in AI pet hardware of the present invention employs a finite state machine architecture to schedule the behavior of the device. The finite state machine includes six states:

[0080] S0 Independent State: Independent standby state when no similar device is detected;

[0081] S1 Sensing Status: Attention status after detecting a signal from a similar device;

[0082] S2 Excitement State: The response state in which the level of excitement gradually changes with distance when the device is brought close;

[0083] S3 Pairing Status: The state in which the device connection is established by triggering the pairing process via NFC touch.

[0084] S4 Synchronization Status: The state of synchronization interaction after successful pairing;

[0085] S5 Separation State: The state in which phased emotional expression is performed when the device is far away;

[0086] The transitions between states are determined by the strength of the BLE signal, the triggering of BLE, and the duration of BLE, forming a complete biological social behavior chain covering independence, perception, excitement, pairing, synchronization, and separation. The state transition rules are shown in Table 1.

[0087] Table 1: Summary of State Transition Rules

[0088]

[0089] The following sections will provide a detailed introduction to the triggering conditions, behavior, and state transitions of each state in the state machine.

[0090] 1. S0 independent state.

[0091] Triggering condition: Enters independent state when no similar device is detected. After the device powers on or returns to independent state from other states, the following behavior is performed.

[0092] BLE scanning strategy:

[0093] (1) Default scan cycle: 500ms;

[0094] (2) When the battery is fully charged: shortened to 300ms;

[0095] (3) When the battery is low: the time is extended to 1000ms;

[0096] (4) When the accelerometer detects movement: shorten to 100ms.

[0097] Behavioral performance:

[0098] (1) Minor random head rotation (±5°);

[0099] (2) Gentle breathing light (brightness 20%-30%, frequency 0.3Hz);

[0100] (3) No sound output;

[0101] (4) Intermittent behavior: The “look around” action is performed randomly every 5-30 seconds.

[0102] State transition: BLE scan detects similar devices and RSSI > -75dBm → transition to S1 sensing state.

[0103] 2. S1 perception state.

[0104] Triggering conditions: The device enters sensing mode when BLE detects a similar device and RSSI > -75dBm.

[0105] RSSI monitoring mechanism:

[0106] (1) FAR threshold: -75dBm (far from the threshold, below this value returns to S0);

[0107] (2) NEAR threshold: -55dBm (near distance threshold, above this value enters S2).

[0108] Direction determination: The target direction is estimated by differential estimation of three consecutive RSSI samples, providing directional information for subsequent head turning.

[0109] Authentication: Parse the manufacturer ID and device type in the broadcast data, and trigger subsequent actions only after confirming that it is a device of the same type.

[0110] Behavioral performance:

[0111] (1) Stop the current action and slowly turn your head toward the target (10° / s);

[0112] (2) Light brightness 40%-50%, frequency 0.8Hz;

[0113] (3) No sound output.

[0114] Status transition:

[0115] (1) RSSI > -55dBm for 0.5s → S2 Excitation state;

[0116] (2) RSSI < -75dBm for 2s → S0 independent state.

[0117] 3. S2 excited state.

[0118] The excited state is divided into three intervals by a three-level excitation gradient mapping, corresponding to three excitation levels: slightly excited, moderately excited, and very excited. Each level is matched with differentiated head movements, lighting effects, and sound effects to achieve the effect of behavior gradually changing with distance. The three-level excitation gradient mapping is shown in Table 2.

[0119] Table 2: Excitation Gradient Mapping Table

[0120]

[0121] The switching between sub-levels of excitation gradients employs a hysteresis fallback mechanism to avoid frequent jumps in excitation levels. This hysteresis fallback mechanism includes:

[0122] (1) Upgrade: No delay, upgrade immediately when RSSI reaches the threshold of the previous level;

[0123] (2) Degradation: After a 1s delay, the RSSI must remain at the next threshold for 1s before the degradation is performed.

[0124] Status transition:

[0125] (1) NFC triggers pairing interrupt → S3 pairing status;

[0126] (2) RSSI falls back to the -75 to -55 dBm range → S1 sensing state;

[0127] (3) RSSI < -75dBm for a duration of T lost → S0 Independent state.

[0128] 4. S3 Pairing Status.

[0129] Pairing is triggered by an NFC touch, enabling secure authentication and relationship establishment between devices.

[0130] The pairing process consists of four stages as shown in Table 3.

[0131] Table 3: Four Stages of the Matching Process

[0132]

[0133] The four stages shown in Table 3 include:

[0134] (1) Surprise phase: Near-field communication triggers pairing interruption, outputting whole-body micro-vibration feedback;

[0135] (2) Exchange phase: Exchange device ID, encrypted public key and feature information through near field communication, and output light flashing feedback;

[0136] (3) Authentication phase: Perform two-way authentication with shared key and output slow pulse feedback of light;

[0137] (4) Establishment phase: Add the other party to the friend list and establish a BLE encrypted connection. After outputting the happy beating feedback, enter the S4 synchronization state.

[0138] Failure handling: When pairing fails, a "disappointment" animation is executed, and after a 3-second cooldown, the pairing state returns to the S2 excitement state. The minimum dwell time in the pairing state is 1 second to prevent repeated triggering within a short period of time.

[0139] 5. S4 synchronization state.

[0140] Synchronization state is the main interaction state after successful pairing. It is divided into three behavior modes based on the sentiment value. The sentiment value is a quantitative value of 0-100 obtained by weighting and normalizing at least the distance between devices, pairing time, and number of interactions.

[0141] Table 4: Synchronization State Behavior Patterns

[0142]

[0143] As shown in Table 4, the S4 synchronization state is divided into three behavioral patterns based on the emotional value:

[0144] (1) When the emotional value is > 50, the synchronous jumping mode is executed, and fast-paced movements and breathing-like fluctuations are output;

[0145] (2) When the emotional value is in the range of 25-50, the breathing companion mode is executed, and the breathing synchronization action with a slow rhythm is output;

[0146] (3) When the emotion value is <25, the interactive exploration mode is executed, outputting independent small exploration actions and maintaining the connection.

[0147] Status transition:

[0148] (1) RSSI < -70dBm for 3s → S5 Separation state;

[0149] (2) Changes in emotional value → Switching between three modes;

[0150] (3) New device touch → S3 Re-pair.

[0151] 6. S5 separation state.

[0152] The separation state simulates the emotional expression of organisms through a three-stage separation behavior, reflecting the biological characteristic of "reluctance to part". The three-stage separation behavior is shown in Table 5.

[0153] Table 5: Three Stages of Separation Behavior

[0154]

[0155] The first stage is the search stage: within a few seconds after the device moves away, it will quickly turn its head left and right, flash its lights quickly, and make a short beeping sound at the same frequency as the S4 state;

[0156] The second stage is the reluctant stage: within a few seconds, the head movement slows down and slowly lowers, the brightness of the lights gradually dims, and the output amplitude decreases linearly, accompanied by a low sigh.

[0157] The third stage is the acceptance stage: quiet behavior, faint breathing light, and silence that restore the S0 independent state within a few seconds;

[0158] The amplitude decay during the non-discarding phase employs a linear interpolation algorithm:

[0159] ;

[0160] in, The amplitude of the action in the S4 synchronization state. To determine the duration of the phase, t represents the current time. In terms of sound, the tone linearly transitions from a "call" to a "whisper," with the volume decreasing synchronously.

[0161] Reunion Mechanism: During the separation process, if the other party re-enters the BLE range (RSSI > -55dBm), the separation behavior is interrupted, and the system directly jumps to the S2 excited state, realizing the emotional expression of "reunion." This design simulates the excited response of real organisms when they meet again after separation.

[0162] Example 2

[0163] To avoid frequent state jumps caused by RSSI signal fluctuations, this embodiment designs a global anti-jitter mechanism based on the finite state machine architecture used in the control method of Embodiment 1, which includes the following four components.

[0164] 1. Dual threshold design:

[0165] This embodiment sets separate "entry threshold" and "exit threshold" for state transitions involving wireless signal strength determination, forming a hysteresis interval to eliminate critical oscillations.

[0166] Take the transition between the S1 perceptual state and the S2 excited state as an example:

[0167] Conditions for entering the S2 excitation state: The system must detect a BLE signal strength RSSI greater than -55dBm from a similar device.

[0168] Conditions for returning to S1 sensing state: The system must detect that RSSI is less than -65dBm.

[0169] The resulting 10dBm hysteresis range (-65dBm to -55dBm) ensures that when RSSI fluctuates slightly near the critical value, the state machine will not frequently switch between two states, thus avoiding repetitive actions by the AI ​​pet.

[0170] 2. Time requirement met:

[0171] In addition to the signal strength threshold, this embodiment also introduces a time dimension as a necessary condition for state transition. The system only performs a state transition after the triggering condition has been continuously met for a specific duration. For example, transitioning from S1 to S2 requires RSSI > -55dBm for 0.5s; conversely, if RSSI falls below -55dBm during this period, the transition is canceled.

[0172] 3. Minimum state dwell time:

[0173] To prevent the state machine from "flashing" due to certain transient events (such as NFC touch interference or sudden noise), this embodiment sets a minimum dwell time constraint for specific states.

[0174] A typical application scenario is the S3 pairing state: once the system enters the S3 pairing state, regardless of whether the pairing process is completed instantly or fails, this state is forcibly maintained for at least 1 second. This mechanism ensures that the entire pairing process of "surprise → exchange → authentication → establishment" has enough time to be fully displayed to the user, avoiding interruptions and ensuring the integrity of the interaction.

[0175] 4. Status History:

[0176] Record the timestamp and triggering event for each state transition for anomaly diagnosis and behavior analysis.

[0177] The threshold values ​​for each state parameter in this embodiment can be set with reference to the summary in Table 6.

[0178] Table 6: Summary of Thresholds for Various State Parameters

[0179]

[0180] Example 3

[0181] To further enhance the robustness and interactive realism of the present invention in complex environments, this embodiment describes in detail the handling strategies for various typical abnormal scenarios.

[0182] Table 7: Comparison Table of Exception Handling Strategies

[0183]

[0184] As shown in Table 6, the strategies for handling abnormal scenarios include:

[0185] 1. Handling multiple devices entering the scene concurrently:

[0186] When the AI ​​pet hardware is in S1 perception state or S2 excitement state, if signals from multiple similar devices are detected simultaneously via BLE broadcast (RSSI all greater than -75dBm), the system will initiate a multi-device arbitration mechanism:

[0187] First, the signals of all similar devices currently scanned are analyzed to extract the RSSI value of each device. Then, a priority queue is established based on RSSI strength, and the device with the largest RSSI value (i.e., the strongest signal and the closest distance) is selected as the priority interaction target, while the remaining devices are added to the candidate list.

[0188] In terms of behavior, the AI ​​pet hardware will only execute the three-level gradient response of the S2 excitement state (such as head turning, light flashing, and sound feedback) for the priority interaction object, while temporarily not responding to devices in the candidate list. This effectively avoids resource conflicts and chaotic actions caused by simultaneous interaction of multiple devices, ensuring the orderly nature of the interaction.

[0189] 2. Handling pairing interruption scenarios:

[0190] In S3 pairing mode, if pairing is interrupted due to NFC communication timeout, data verification error, or encryption authentication failure, the system will execute the following fault-tolerant procedure:

[0191] First, the current pairing process is immediately terminated, triggering a "confused" animation (e.g., slight head movement, irregular light flashing). Then, the system enters a 2-second waiting period, during which NFC detection is paused. After the waiting period, the system automatically initiates a retry request, allowing a maximum of 3 retries. If a connection cannot be established after 3 attempts, a "disappointed" animation is forcibly executed (e.g., head tilting, light dimming), and after a 3-second cooldown, it automatically reverts to the S2 excited state, waiting for the next trigger. This mechanism prevents frequent invalid pairing attempts due to momentary interference.

[0192] 3. Handling low battery scenarios:

[0193] The system monitors the battery management unit (BMU) power signal in real time. Once the power level is detected to be below a preset threshold (e.g., 20%), the system will trigger a forced power-saving rollback mechanism.

[0194] Regardless of the current state, the system will forcibly clear the current state and jump to the S0 independent state. In this state, the system will shut down unnecessary peripherals (such as high-brightness LEDs and speech synthesis modules) and extend the BLE scan cycle from the default 500ms to 1000ms to reduce power consumption. At this time, the AI ​​pet hardware only maintains a faint breathing light effect (brightness 20%-30%) and random "looking around" actions to ensure that it can still maintain a basic standby presence when the battery is low, while maximizing the battery life.

[0195] 4. Handling repeated contact between paired devices:

[0196] When a device is in S0 standalone state or S2 excited state, if a paired "old friend" device is detected to be touched again via NFC, the system will skip the standard S3 pairing process (including data exchange and encryption authentication stages) and directly recognize it as a "reunion" event:

[0197] The system reads the device's ID from memory, confirms its existence in the "friends list," and then triggers a special "reunion" animation (such as a high-frequency cheerful chirping sound and rapid tail wagging), before directly entering S4 synchronization mode. This approach not only shortens the interaction latency but also significantly enhances the product's anthropomorphic emotional experience by simulating a biological response to the memory of familiar people.

[0198] Example 4

[0199] This embodiment also proposes an AI pet hardware, including a processor, a memory, a wireless communication module, a near-field communication module, and an actuator. The memory stores a computer program, which, when executed by the processor, implements the control method for simulating biological social behavior.

[0200] The above embodiments are only for illustrating the technical concept and features of the present invention, and are intended to enable those skilled in the art to understand the content of the present invention and implement it accordingly. They should not be construed as limiting the scope of protection of the present invention. All modifications made according to the spirit and essence of the main technical solution of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A control method for simulating biological social behavior of an AI pet hardware, characterized in that, A finite state machine architecture is used to schedule device behavior. The finite state machine includes six states: S0 Independent State: Independent standby state when no similar device is detected; S1 Sensing Status: Attention status after detecting a signal from a similar device; S2 Excitement State: The response state in which the level of excitement gradually changes with distance when the device is brought close; S3 Pairing Status: The state in which a device connection is established by triggering the pairing process through near-field communication; S4 Synchronization Status: The state of synchronization interaction after successful pairing; S5 Separation State: The state in which phased emotional expression is performed when the device is far away; The transitions between states are determined by the strength of the wireless signal, the triggering of near-field communication, and the duration of the communication, forming a complete biological social behavior chain covering independence, perception, excitement, pairing, synchronization, and separation.

2. The control method according to claim 1, characterized by, The S2 excitation state is configured with a three-level excitation gradient mapping mechanism: The Received Signal Strength Indicator (RSSI) is divided into three intervals, corresponding to three excitement levels: Micro, ZTE, and High. Each level is matched with different head movements, lighting effects, and sound effects. A delayed fallback mechanism is also set up: when the excitement level is upgraded, there is no delay in responding to the change in the RSSI threshold; when the level is downgraded, the switch is only executed after the RSSI falls back to the corresponding range for a certain period of time, so as to avoid frequent level jumps.

3. The control method according to claim 1, characterized by, The S5 separation state is configured with a three-stage separation behavior: The first stage is the search stage: within a few seconds after the device moves away, it will quickly turn its head left and right, flash its lights quickly, and make a short beeping sound at the same frequency as the S4 state; The second stage is the reluctant stage: within a few seconds, the head movement slows down and slowly lowers, the brightness of the lights gradually dims, and the output amplitude decreases linearly, accompanied by a low sigh. The third stage is the acceptance stage: quiet behavior, faint breathing light, and silence that restore the S0 independent state within a few seconds; where the action amplitude of the non-sustained phase Decay with linear interpolation algorithm: ； in, The amplitude of the action in the S4 synchronization state. To find the duration of a stage, t represents the current time.

4. The control method according to claim 3, characterized in that, The S5 separation state also includes a reunion interruption mechanism: If a device of the same type re-enters the BLE detection range during the separation process, the current separation process is interrupted, and the process jumps directly to the S2 excitation state to execute the reunion response behavior.

5. The control method according to claim 1, characterized in that, It also includes a global anti-shake mechanism: A dual-threshold design is adopted, with the entry threshold and exit threshold for the same state transition having an interval of ≥10dBm, forming a hysteresis interval; All state transitions must be triggered for a specified duration before the transition is executed; The S3 pairing status is set with a minimum dwell time to prevent repeated triggering in a short period of time.

6. The control method according to claim 1, characterized in that, It also includes a multi-device interaction processing mechanism: When multiple similar devices enter the detection range at the same time, priority is given to establishing interaction with the device with the strongest RSSI signal.

7. The control method according to claim 1, characterized in that, In the S1 sensing state and S2 excitement state, if the device that triggers near-field communication is a paired device, skip the S3 pairing process and directly jump to the S4 synchronization state and execute the reunion animation.

8. The control method according to claim 1, characterized in that, The S3 pairing state executes the following four-stage process sequentially: Surprise phase: Near-field communication triggers pairing interruption, outputting full-body micro-vibration feedback; Exchange phase: Exchange device ID, encrypted public key and feature information through near field communication, and output fast light flashing feedback; Authentication phase: Perform two-way authentication using the shared key, and output slow pulsed light feedback; Setup phase: Add the other party to the friend list and establish a BLE encrypted connection. After outputting the happy beating feedback, enter the S4 synchronization state. If the pairing fails, a disappointed animation is output, and after cooling, it returns to the excited state in S2.

9. The control method according to claim 1, characterized in that, The S4 synchronization state is divided into three behavior patterns according to the emotion value: When the emotion value > N1, the same-frequency beating mode is executed, and fast-paced actions and breathing-like fluctuations are output. When the emotion value is in the range of N2–N1, the breathing accompaniment mode is executed, and the breathing-like synchronization actions with a slowed-down rhythm are output. When the emotion value < N2, the interactive exploration mode is executed, and independent and tiny exploration actions are output while maintaining the connection. The emotion value is a quantified value of 0-100 obtained by weighted normalization of at least the distance between devices, the pairing duration, and the number of interactions, where 0 < N2 < N1 < 100.

10. An AI pet hardware, comprising a processor, a memory, a wireless communication module, a near-field communication module, and an actuator, characterized in that, The memory stores a computer program, and when the program is executed by a processor, it implements the control method for simulating biological social behaviors described in any one of claims 1 to 9.

Images