A spherical companion robot based on self-balancing technology

Abstract

This invention relates to a spherical companion robot based on self-balancing technology. The system includes: a data acquisition unit, a data processing unit, a motion control unit, and an interactive execution unit; acquiring multi-dimensional raw data of the spherical companion robot; extracting the robot's posture feature parameters; decomposing remote control commands into action sequences and parametrically parsing interactive configuration information; obtaining hierarchical encoding of motion parameters using corresponding dynamic adaptation calculations; obtaining self-balancing control parameters and controllable motion drive commands through collaborative matching; adjusting and calibrating the tilt posture to obtain the spherical companion robot's self-balancing state; performing linkage analysis of interactive commands and motion states on the self-balancing posture and controllable rolling motion; triggering peripheral action timing planning based on the analysis results; and driving built-in peripherals to execute corresponding action commands according to the timing sequence; thus realizing the self-balancing motion of the spherical companion robot.

Description

Technical Field

[0001] This invention belongs to the field of intelligent interactive pet toys, specifically relating to a spherical companion robot based on self-balancing technology. Background Technology

[0002] Currently available pet ball toys still have the following areas for improvement: In the field of smart pet companion toys, spherical interactive toys have become the mainstream choice for pet interaction due to their flexible movement. However, the motion control technology of existing pet spherical toys has significant shortcomings. Most of them adopt irregular random rolling and simple vibration motion, lacking precise directional and controllable motion design. They cannot complete movement in a specified direction according to user commands, and the poor motion controllability greatly reduces the fun of pet interaction. At the same time, existing products do not have self-balancing technology, and are prone to tipping over and getting stuck during rolling, making it difficult to maintain a continuous state of motion. This not only reduces the efficiency of the toy's use, but also fails to meet the pet's need for continuous play and interaction, becoming the core problem restricting the improvement of the pet spherical toy experience.

[0003] Existing pet ball toys have limited interactive features, equipped only with basic sound and light effects. These effects are not linked to the pet's movement and often play in a fixed, looping pattern, resulting in short-lived attraction and insufficient playability. Furthermore, these products lack a remote interaction mechanism, supporting only in-person, close-range interaction. Owners cannot remotely control the toy via mobile devices or engage in remote voice interaction with their pets. This makes it difficult to provide remote companionship when owners are away, failing to meet the remote interaction and companionship needs of modern pet owners and falling short of the trend towards intelligent pet ownership.

[0004] The overall design lacks a systematic intelligent control system. The monitoring, control, and interaction modules are independent of each other, lacking a unified data analysis and command scheduling unit. It cannot collect and process data such as the toy's movement posture and cooling status in real time, nor can it dynamically adjust the interaction mode based on the pet's behavioral responses; it can only execute preset simple actions. Furthermore, the product's fixing and driving structure is rudimentary, lacking precise power distribution and posture calibration mechanisms. The rolling trajectory is prone to deviation, and it cannot adapt to different ground contact surfaces. Its movement stability varies greatly on smooth, rough, and other materials, further reducing the product's intelligence and practicality. Summary of the Invention

[0005] To address the aforementioned problems in the existing technology, this invention provides a spherical companion robot based on self-balancing technology; The objective of this invention can be achieved through the following technical solutions: Data acquisition unit, data processing unit, motion control unit, interactive execution unit; The data acquisition unit is used to acquire multi-dimensional raw data of the spherical companion robot; The data processing unit extracts the posture feature parameters of the spherical companion robot based on the multi-dimensional raw data; it decomposes the remote control commands into actions and performs parameterized parsing of the interaction configuration information to obtain the motion control logic and interaction execution requirements; it obtains the hierarchical encoding of motion parameters by performing corresponding dynamic adaptation calculations; and it obtains the self-balancing regulation parameters and controllable motion drive commands by using the coordinated matching of the posture feature parameters and the motion control logic. The motion control unit adjusts and calibrates the tilt posture based on the self-balancing control parameters and controllable motion drive commands to obtain the self-balancing state of the spherical companion robot; at the same time, it controls the power guide wheels to perform differential rotation and directional controllable rolling. The interactive execution unit performs interactive command and motion state linkage analysis on the self-balancing posture and controllable rolling motion, and triggers peripheral action timing planning based on the analysis results, driving the built-in peripheral to execute corresponding action commands according to the timing.

[0006] As a preferred embodiment of the present invention, the specific process of extracting the posture feature parameters of the spherical companion robot includes: Based on the multi-dimensional raw data fusion posture detection and dynamic prediction algorithm, feature recognition is performed to obtain the raw sensor data of the spherical companion robot's tilt angle, rotation speed and dynamic prediction. The original sensing data is subjected to adaptive dimension normalization processing to dynamically match the dimension of the motion state correction data of the spherical companion robot, thereby obtaining standardized posture data. By performing multi-feature extraction on the standardized posture data, posture parameters of the spherical companion robot containing dynamic prediction features are obtained.

[0007] Specifically, the process of decomposing the remote control commands into actions includes: Based on the remote control command data in the multi-dimensional raw data, perform command semantic-action mapping, decode and associate command semantics with the executable actions of the spherical companion robot, and obtain the plaintext remote control command with action priority; The plaintext remote control command is decomposed into hierarchical action logic to obtain a single primitive action command and a combination rule for composite actions.

[0008] Specifically, the process of parametrically parsing the interactive configuration information includes: Based on the interactive configuration information in the multi-dimensional raw data, personalized information extraction and format translation are performed to extract user-defined interactive parameters; Record users' daily behavior data and synchronize it to the robot's accompanying intelligent computing chip for learning and storage, and obtain structured personalized interaction configuration information of user behavior characteristics; The structured personalized interaction configuration information is dynamically decoupled from parameters to obtain independent interaction modes for pet behavior adaptability and user growth adaptability.

[0009] Specifically, the process of obtaining the hierarchical encoding of motion parameters includes: By utilizing dynamic adaptation calculations under self-balancing constraints and combining them with the motion threshold adaptation parameters of the spherical companion robot, adaptive motion control parameters under self-balancing constraints are obtained. Two-dimensional motion control parameters are obtained through multi-dimensional parameter threshold calibration; The two-dimensional motion control parameters are processed by motion priority hierarchical sequencing, and the sequencing is distinguished by combining motion and auxiliary motion to obtain priority-identified motion parameter hierarchical sequencing data.

[0010] Specifically, the process of obtaining the self-balancing control parameters and the controllable motion drive command includes: Based on the hierarchical sequencing data of the motion parameters and the posture parameters, a real-time posture coordination matching calculation is performed, and a motion inertia compensation factor is introduced to obtain the initial self-balancing parameter adjustment with inertia compensation. By combining the initial self-balancing parameter adjustment with the driving parameters, self-balancing correction is performed, and the sensitivity of the control parameters is dynamically adjusted to obtain self-balancing parameter adjustments that adapt to the real-time motion state. By using the self-balancing adjustment and the driving parameters, instruction generation and redundancy configuration are performed to generate the main driving instruction and the backup fault-tolerant instruction, thereby obtaining a controllable driving instruction.

[0011] Specifically, the process of adjusting and calibrating the tilt attitude includes: Based on the self-balancing parameter adjustment combined with real-time sensing data from the electronic gyroscope, deviation anomalies are detected in the real-time tilting posture of the spherical companion robot to obtain posture tilt data. By dynamically adjusting the moving distance and speed of the driving counterweight, the posture of the spherical companion robot with dynamic center of gravity adjustment is obtained. Based on the spherical companion robot's posture combined with the gyroscope closed-loop verification algorithm, real-time posture calibration is performed. The actual posture is compared with the target balanced posture and dynamically corrected to obtain the spherical companion robot's balanced posture.

[0012] Specifically, the process of controlling the power guide wheel to rotate at differential speed includes: Based on the controllable drive command, the independent power distribution of the wheel set is analyzed, and the power distribution parameters of the power guide wheel with force compensation are obtained by combining the force state and movement direction of the guide wheel. By using the power distribution parameters to control the rotational speed and introducing a road surface friction coefficient compensation factor, the target rotational speed parameter value of the power guide wheel adapted to the corresponding contact surface is obtained. By performing frequency conversion independent speed tuning, the speed of the guide wheel is controlled by frequency conversion, thereby obtaining the self-adaptive differential rotation of the power guide wheel.

[0013] Specifically, the directional controllable scrolling is as follows: By using the stable posture of the sphere and the differential rotation state of the power guide wheel to preset the dynamic trajectory, and combining environmental sensing data to avoid obstacles in real time, the directional rolling trajectory parameters of the spherical companion robot with obstacle avoidance strategy are obtained. The directional rolling trajectory parameters are combined with the differentially rotating power guide wheel for trajectory driving, and the wheel speed is adjusted in real time to match the trajectory parameters, so as to obtain the directional rolling driving state of the spherical companion robot; Based on the directional rolling dynamics, a self-balancing and trajectory dual-dimensional real-time verification is performed, and the rolling trajectory is dynamically adjusted to obtain a spherical companion robot that rolls smoothly and controllably along a preset direction without deviation.

[0014] Specifically, the process of analyzing the linkage between interactive commands and motion states includes: Based on the stable posture and the unbiased and controllable rolling, multi-dimensional motion characteristics are extracted, and the motion rate, direction, and posture stability parameters of the spherical companion robot are obtained simultaneously to obtain the real-time comprehensive motion parameters of the spherical companion robot. By establishing a dynamic correlation between motion and interaction, a linkage mapping model between motion state and interactive action is established, resulting in motion-interaction linkage rules with motion adaptability. The interactive execution commands are analyzed in a contextualized manner, and the execution logic of the interactive commands is optimized in combination with the motion scenario to obtain the interactive analysis results that are adapted to the real-time motion status of the spherical companion robot.

[0015] Specifically, the process of triggering peripheral action timing planning based on the parsing results includes: Based on the interaction analysis results, the peripheral action type is identified, the scenario requirements for action execution are identified, and the peripheral type, action requirements and scenario adaptation parameters of the action to be executed are obtained. Simultaneously, the timing sequence of action execution is arranged, and the timing sequence of peripheral actions is matched with the movement rhythm of the spherical companion robot to obtain the timing sequence table of peripheral action execution. Based on the peripheral action execution timing table, the action triggering conditions are dynamically thresholded, the triggering threshold of motion state linkage is preset, and the linkage threshold action triggering command and execution timing parameters of the peripheral are obtained.

[0016] Specifically, the process by which the driver executes corresponding action instructions according to the timing sequence for the built-in peripheral includes: Based on the linkage threshold action trigger command and execution timing parameters, generate peripheral personalized drive signals, adjust the drive signal parameters according to the peripheral type and action requirements, and obtain the drive control signals of the built-in peripherals. Based on the drive control signal, peripheral synchronization adaptation is performed according to the timing table to obtain the real-time execution dynamics of peripheral synchronization. The peripheral device is used to synchronously and in real time to perform dynamic motion verification, and at the same time, the adaptability of the peripheral device's motion effect to the spherical companion robot's motion state is verified, and the peripheral device's motion parameters are dynamically corrected.

[0017] The beneficial effects of this invention are as follows: By incorporating self-balancing technology and a precise differential drive control system, the spherical companion robot achieves real-time posture calibration and self-balancing through the synergistic effect of counterweights and electronic gyroscopes. This effectively solves the problems of easy tipping and jamming in traditional products, ensuring the posture stability of the spherical companion robot during movement. At the same time, through differential rotation control of four power guide wheels, the spherical companion robot achieves directional and controllable rolling in the front, back, left, and right directions, accurately responding to remote control commands. This completely changes the random rolling motion of traditional products, significantly improving the controllability of the spherical companion robot's movement, making pet interaction more fun, and meeting the pet's need for continuous play.

[0018] A smart execution system that deeply integrates movement and interaction has been constructed. This system analyzes and plans the self-balancing posture and controllable rolling motion of the spherical companion robot in conjunction with its sound, light, and voice interaction modules. This ensures that the dazzling light, music, and voice interaction effects are matched to the robot's movement speed and direction, achieving dynamic and scenario-based intelligent interaction. Compared to the fixed sound and light effects of traditional products, this significantly enhances the appeal to pets and enriches the product's playability. Simultaneously, relying on WIFI remote control technology, a remote interaction link between owner and pet has been established. This supports real-time control via mobile terminals, recording and remote playback of owner commands, and two-way voice calls, allowing owners to interact and accompany their pets remotely even when they are away. This builds a communication bridge between owners and pets, improving the product's humanization and intelligence.

[0019] By acquiring multi-dimensional data, the spherical companion robot obtains real-time data on its posture, movement status, and interactive commands. Through data processing unit analysis, adaptation, and computation, precise parameters and commands are provided for motion control and interactive execution, enabling coordinated scheduling and intelligent response across modules. Simultaneously, a road friction coefficient compensation factor is introduced into the power control, enabling adaptive differential rotation of the power guide wheels. This allows for adaptation to different ground contact surfaces, including smooth and rough surfaces, ensuring motion stability in various scenarios. Furthermore, the systematic control system dynamically adjusts the interaction mode and motion parameters based on the pet's behavioral responses, achieving personalized intelligent interaction. Compared to traditional products with independent module designs, this significantly improves the overall intelligence and practicality, providing pets with a superior interactive and companionship experience. Attached Figure Description

[0020] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0021] Figure 1 This is a schematic diagram of the process of a spherical companion robot based on self-balancing technology according to the present invention; Figure 2 This is a schematic diagram of the interactive execution process in this invention; Figure 3 This is a structural diagram of the spherical companion robot of the present invention; The structure of the spherical companion robot includes: ① a spherical soft rubber outer shell, ② an engineering plastic reinforced inner shell, ③ guide wheels, ④ an intelligent control power core, ⑤ a speaker and microphone, ⑥ LED lights, ⑦ a circuit board, and ⑧ a counterweight. Detailed Implementation

[0022] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided.

[0023] Please see Figure 1-3 A spherical companion robot based on self-balancing technology includes: Data acquisition unit, data processing unit, motion control unit, interactive execution unit; The data acquisition unit is used to acquire multi-dimensional raw data of the spherical companion robot; The data processing unit extracts the posture feature parameters of the spherical companion robot based on the multi-dimensional raw data; it decomposes the remote control commands into actions and performs parameterized parsing of the interaction configuration information to obtain the motion control logic and interaction execution requirements; it obtains the hierarchical encoding of motion parameters by performing corresponding dynamic adaptation calculations; and it obtains the self-balancing regulation parameters and controllable motion drive commands by using the coordinated matching of the posture feature parameters and the motion control logic. The motion control unit adjusts and calibrates the tilt posture based on the self-balancing control parameters and controllable motion drive commands to obtain the self-balancing state of the spherical companion robot; at the same time, it controls the power guide wheels to perform differential rotation and directional controllable rolling. The interactive execution unit performs interactive command and motion state linkage analysis on the self-balancing posture and controllable rolling motion, and triggers peripheral action timing planning based on the analysis results, driving the built-in peripheral to execute corresponding action commands according to the timing.

[0024] As a preferred embodiment of the present invention, the specific process of extracting the posture feature parameters of the spherical companion robot includes: Based on the multi-dimensional raw data fusion posture detection and dynamic prediction algorithm, feature recognition is performed to obtain the raw sensor data of the spherical companion robot's tilt angle, rotation speed and dynamic prediction. The original sensing data is subjected to adaptive dimension normalization processing to dynamically match the dimension of the motion state correction data of the spherical companion robot, thereby obtaining standardized posture data. By performing multi-feature extraction on the standardized posture data, posture parameters of the spherical companion robot containing dynamic prediction features are obtained.

[0025] In this embodiment, the six-axis sensing acquisition module integrated by the circuit board is preset with a high-frequency sampling frequency to continuously capture multi-dimensional raw data such as acceleration, angular velocity, and spatial position of the device in all motion states such as rolling, turning, and stationary. The collected real-time data is transmitted to the intelligent control power core without delay through the onboard communication link. The intelligent control power core has a built-in dedicated computing chip that integrates attitude detection algorithm and dynamic prediction algorithm to carry out feature recognition and trend calculation. It accurately captures and obtains the real-time tilt angle and rotation rate of the device within the safe tilt range. At the same time, it predicts the subsequent motion trend based on the previous motion data to obtain raw sensing data including tilt, rotation and trend prediction. To address the issue of inconsistent dimensions among different physical quantities such as tilt angle (°) and rotation rate (rad / s) in the original sensor data, the intelligent control power core automatically performs self-adaptive dimension normalization processing. Based on the real-time motion state of the device, it dynamically calls correction algorithms to uniformly correct all posture data to the standardized dimension range of 0~1, completely eliminating the computational errors caused by differences in data dimensions and forming highly consistent standardized posture data. Finally, through multi-dimensional feature extraction algorithms, the standardized posture data is deeply analyzed, and the dynamic prediction features are deeply fused and feature-enhanced with the real-time posture features of the device to ultimately obtain the posture parameters of the spherical companion robot containing dynamic prediction features.

[0026] Specifically, the process of decomposing the remote control commands into actions includes: Based on the remote control command data in the multi-dimensional raw data, perform command semantic-action mapping, decode and associate command semantics with the executable actions of the spherical companion robot, and obtain the plaintext remote control command with action priority; The plaintext remote control command is decomposed into hierarchical action logic to obtain a single primitive action command and a combination rule for composite actions.

[0027] In this embodiment, the intelligent control power core accurately extracts remote control command data transmitted via WiFi from multi-dimensional raw data using a command filtering algorithm. It then relies on a built-in pre-trained command semantic-action mapping database to rapidly decode the command semantics. This allows for precise association between the text and voice semantic information of the remote control and the 12 basic actions that the spherical companion robot can perform, such as forward, backward, left turn, right turn, emergency stop, and sound / light triggering. Furthermore, it categorizes actions into three priority levels based on their core importance, urgency, and hardware execution priority, with directional movement and posture calibration receiving a priority of level 1. The highest priority is given to Level 1, followed by Level 2 for audio-visual interaction and voice playback, and Level 3 for parameter fine-tuning and status feedback. Plaintext remote control commands with clear action priority identifiers are generated. Subsequently, the plaintext remote control commands are decomposed into layered action logic. Through the command decomposition algorithm, complex action commands such as "move forward 50cm and simultaneously trigger audio-visual interaction" are accurately decomposed into single primitive action commands such as "move forward" and "audio-visual trigger". At the same time, clear and executable primitive action combination rules are formulated, which clarify the combination execution logic, execution order, linkage time difference and interruption response mechanism of different primitive actions.

[0028] Specifically, the process of parametrically parsing the interactive configuration information includes: Based on the interactive configuration information in the multi-dimensional raw data, personalized information extraction and format translation are performed to extract user-defined interactive parameters; Record users' daily behavior data and synchronize it to the robot's accompanying intelligent computing chip for learning and storage, and obtain structured personalized interaction configuration information of user behavior characteristics; The structured personalized interaction configuration information is dynamically decoupled from parameters to obtain independent interaction modes for pet behavior adaptability and user growth adaptability.

[0029] In this embodiment, the intelligent control power core performs refined processing of user interaction configuration information in multi-dimensional raw data to provide companionship and personalization. Through the personalized information extraction algorithm and the format unification translation module, it accurately extracts core configuration parameters such as customized sound and light combination mode, voice playback frequency, and pet triggering interaction conditions from the information configured on the user's mobile terminal. It transforms unstructured configuration information such as text and button selection into structured personalized interaction configuration information in the form of standardized key-value pairs. At the same time, through the multi-dimensional sensing module, it continuously collects and records the child's daily routine, interaction preferences, interests, and other life behavior data, and synchronizes them in real time to the robot's companion intelligent computing chip for deep learning and long-term storage. It integrates the child's behavioral characteristics into the structured configuration information to form exclusive basic configuration data with personality characteristics.

[0030] The structured personalized interactive configuration information that integrates user behavior characteristics is dynamically decoupled. Each interactive parameter is independently separated into functional modules. At the same time, it is connected to the built-in pet behavior characteristic database and child growth characteristic database. Through big data matching algorithms, it matches pet behavior adaptation parameters according to the pet's breed, living habits and activity level, and matches corresponding interactive function parameters according to the child's age group and growth stage characteristics. This generates an independent interactive mode that combines pet behavior adaptation and child growth adaptation, realizing dynamic adaptation of interactive functions from different stages from toddlers to teenagers, so that the robot's companion function is in sync with the child's growth needs.

[0031] Through parameter standardization calibration and flexible calibration algorithms, the interactive parameters are uniformly standardized and calibrated across the entire range, reserving sufficient dynamic adjustment thresholds for each parameter. At the same time, a dedicated seamless data migration mechanism is built for the accompanying intelligent computing chip, ensuring that all learning and stored life behavior data and interaction preference data are retained and migrated without loss when the chip undergoes functional iteration and upgrades as the child grows. Based on the long-term retained child behavior data, the chip autonomously develops a unique interactive personality that matches the child's character, forming a unique companionship style. It acquires executable interactive parameters that are dynamically optimized in response, support iteration throughout the entire growth cycle, are personality-specific, and allow for seamless data migration, making the robot a truly exclusive companion that accompanies the child's growth, achieving continuous, personalized, and exclusive companionship.

[0032] Specifically, the process of obtaining the hierarchical encoding of motion parameters includes: By utilizing dynamic adaptation calculations under self-balancing constraints and combining them with the motion threshold adaptation parameters of the spherical companion robot, adaptive motion control parameters under self-balancing constraints are obtained. Two-dimensional motion control parameters are obtained through multi-dimensional parameter threshold calibration; The two-dimensional motion control parameters are processed by motion priority hierarchical sequencing, and the sequencing is distinguished by combining motion and auxiliary motion to obtain priority-identified motion parameter hierarchical sequencing data.

[0033] In this embodiment, the intelligent control power core, under the premise of the device's self-balancing physical constraints, conducts multiple rounds of dynamic adaptation calculations. Combining the spherical companion robot's own hardware motion limits and the load-bearing capacity of its core components, it uses motion threshold adaptation parameters, including preset safe rotational speed thresholds for the guide wheels and stable movement speed thresholds for the counterweights. Through iterative calculations, it obtains adaptable motion control parameters that meet both self-balancing control requirements and are fully adapted to the device's hardware physical conditions. The adaptable motion control parameters undergo multi-dimensional parameter threshold verification for speed, attitude, and power output. A threshold filtering algorithm is used to perform threshold checks on the two core control dimensions of speed and attitude. Parameter screening and validity verification eliminate invalid parameters that exceed hardware and control thresholds, obtaining accurate and effective two-dimensional motion control parameters. These parameters are then prioritized and sequenced according to the device's motion control logic. Core actions such as directional rolling and posture calibration are classified as primary motions, while auxiliary actions such as audio-visual interaction and status parameter feedback are classified as auxiliary motions. The sequencing and priority marking are completed based on the principle of prioritizing primary motions and coordinating with auxiliary motions, resulting in clearly prioritized motion parameter sequencing data. This ensures the orderly and efficient execution of motion control commands.

[0034] Specifically, the process of obtaining the self-balancing control parameters and the controllable motion drive command includes: Based on the hierarchical sequencing data of the motion parameters and the posture parameters, a real-time posture coordination matching calculation is performed, and a motion inertia compensation factor is introduced to obtain the initial self-balancing parameter adjustment with inertia compensation. By combining the initial self-balancing parameter adjustment with the driving parameters, self-balancing correction is performed, and the sensitivity of the control parameters is dynamically adjusted to obtain self-balancing parameter adjustments that adapt to the real-time motion state. By using the self-balancing adjustment and the driving parameters, instruction generation and redundancy configuration are performed to generate the main driving instruction and the backup fault-tolerant instruction, thereby obtaining a controllable driving instruction.

[0035] In this embodiment, the intelligent control power core performs high-frequency real-time attitude coordination matching calculations by hierarchically sequencing motion parameters and attitude parameters with a minimum calculation cycle of 20ms. Simultaneously, a dynamic motion inertia compensation factor of 0.05~0.2 is introduced to precisely offset motion deviations caused by frictional resistance between the spherical soft rubber shell and different ground contact surfaces, as well as the inertia of each moving component, during the equipment's movement. This results in the initial self-balancing parameter adjustment with inertia compensation; the formula is: , Note: The results of attitude coordination matching calculations are combined with an inertial compensation factor to offset motion inertia and friction deviations; Symbol definition: P 初平 The initial self-balancing parameter adjustment value is f(P). 编P 姿 ) is the graded and sequenced data of motion parameters P 编 With posture parameter P 姿 The collaborative matching calculus function, K 惯 The inertia compensation factor is 0.05~0.2, dynamically adapted.

[0036] Using the initial self-balancing parameter adjustment results, combined with hardware drive parameters such as the real-time driving force of the guide wheel and the precise movement of the counterweight, multiple rounds of self-balancing correction calculations are performed. The sensitivity of the control parameters is dynamically adjusted according to the real-time movement rate and tilt state of the equipment, and the step size is adjusted to ensure the precision of the control, thereby obtaining self-balancing parameter adjustments that are fully adapted to the real-time movement state of the equipment. Drive commands are generated and redundantly configured through the self-balancing adjustment parameters and hardware drive parameters. The main drive command and the backup fault-tolerant command are generated according to a 1:1 command ratio, and a dual-command redundancy control mechanism is constructed. When the transmission delay of the main command exceeds the threshold or a command is lost, the system automatically and seamlessly switches to the backup command to obtain a controllable drive command.

[0037] Specifically, the process of adjusting and calibrating the tilt attitude includes: Based on the self-balancing parameter adjustment combined with real-time sensing data from the electronic gyroscope, deviation anomalies are detected in the real-time tilting posture of the spherical companion robot to obtain posture tilt data. By dynamically adjusting the moving distance and speed of the driving counterweight, the posture of the spherical companion robot with dynamic center of gravity adjustment is obtained. Based on the spherical companion robot's posture combined with the gyroscope closed-loop verification algorithm, real-time posture calibration is performed. The actual posture is compared with the target balanced posture and dynamically corrected to obtain the spherical companion robot's balanced posture.

[0038] In this embodiment, the self-balancing parameter adjustment results are fused and analyzed with real-time spatial sensing data collected at a preset frequency by a high-precision electronic gyroscope integrated on the circuit board. This allows for precise detection of deviations and anomalies in the real-time tilting posture of the spherical companion robot. When the detected tilt angle deviation exceeds a safety threshold, the system automatically triggers a posture calibration program. Through deviation calculation, complete posture tilt data, such as tilt direction and tilt angle, are accurately obtained. Based on the posture tilt data, a precise drive algorithm controls the counterweight to dynamically adjust its moving distance and speed within a high-precision preset track of an engineering plastic reinforced inner shell. Precision control is applied to the moving distance and speed adjustment. The precise displacement of the counterweight enables dynamic and refined shifting of the device's center of gravity, quickly correcting the device's tilt posture from a physical perspective. Based on the robot's real-time posture after center of gravity adjustment, a high-frequency real-time posture calibration is performed using a gyroscope closed-loop verification algorithm. With a preset time as the verification cycle, the deviation values ​​between the actual posture and the target balanced posture are continuously compared. Dynamic and slight secondary corrections are then performed based on the deviation values, strictly controlling the correction error within a certain range. This obtains the balanced posture of the spherical companion robot, solving the problems of easy tipping and jamming during device rolling.

[0039] Specifically, the process of controlling the power guide wheel to rotate at differential speed includes: Based on the controllable drive command, the independent power distribution of the wheel set is analyzed, and the power distribution parameters of the power guide wheel with force compensation are obtained by combining the force state and movement direction of the guide wheel. By using the power distribution parameters to control the rotational speed and introducing a road surface friction coefficient compensation factor, the target rotational speed parameter value of the power guide wheel adapted to the corresponding contact surface is obtained. By performing frequency conversion independent speed tuning, the speed of the guide wheel is controlled by frequency conversion, thereby obtaining the self-adaptive differential rotation of the power guide wheel.

[0040] In this embodiment, the four sets of power guide wheels are independently analyzed for power distribution according to controllable drive commands. A force analysis algorithm, combined with the real-time force state of each guide wheel and the preset movement direction of the equipment, is used to match force compensation power distribution parameters for each guide wheel, ensuring that the power output of the wheel set matches the force state. The matched power distribution parameters are used to precisely control the rotational speed of the guide wheels. Simultaneously, a road surface friction coefficient compensation factor is introduced. This factor can automatically match different ground materials such as solid wood, ceramic tiles, carpet, and cement that the spherical soft rubber shell contacts, thereby dynamically adjusting the target rotational speed parameter value of the guide wheels. The rotational speed adjustment accuracy is controlled to ensure the stability of the equipment's movement on different ground surfaces. A frequency conversion control module is used for independent frequency conversion speed tuning, allowing independent frequency conversion control of the rotational speed of the four sets of power guide wheels. The speed difference between adjacent guide wheels can be adjusted according to movement requirements, achieving self-adaptive differential rotation of the power guide wheels and meeting the flexible movement requirements of the equipment in all directions (forward, backward, left, and right).

[0041] Specifically, the directional controllable scrolling is as follows: By using the stable posture of the sphere and the differential rotation state of the power guide wheel to preset the dynamic trajectory, and combining environmental sensing data to avoid obstacles in real time, the directional rolling trajectory parameters of the spherical companion robot with obstacle avoidance strategy are obtained. The directional rolling trajectory parameters are combined with the differentially rotating power guide wheel for trajectory driving, and the wheel speed is adjusted in real time to match the trajectory parameters, so as to obtain the directional rolling driving state of the spherical companion robot; Based on the directional rolling dynamics, a self-balancing and trajectory dual-dimensional real-time verification is performed, and the rolling trajectory is dynamically adjusted to obtain a spherical companion robot that rolls smoothly and controllably along a preset direction without deviation.

[0042] In this embodiment, based on the calibrated stable posture of the device and the real-time differential rotation state of the power guide wheels, a trajectory planning algorithm is used to preset the dynamic rolling trajectory. Simultaneously, environmental sensing data such as obstacles and terrain within a 0-2m detection range collected by the infrared sensing module integrated on the circuit board is used to perform real-time obstacle avoidance path planning. This yields the directional rolling trajectory parameters of the spherical companion robot with a precise obstacle avoidance strategy. The trajectory planning is controlled to ensure trajectory accuracy. The directional rolling trajectory parameters are combined with the differentially rotating power guide wheels for precise trajectory driving. The wheel speed is dynamically adjusted in real-time according to the trajectory parameters, with a preset adjustment period. Precise control is achieved by adjusting the wheel speed step size to ensure the device travels along the preset trajectory, resulting in a stable directional rolling motion for the spherical companion robot. Based on the dynamic self-balancing and trajectory verification of the directional rolling, the system simultaneously monitors the tilt angle balance and the actual rolling trajectory offset through a dual verification module. When the detected trajectory offset exceeds the threshold, the system immediately adjusts the guide wheel speed to dynamically correct the rolling trajectory, ultimately achieving a smooth and controllable rolling motion without offset along the preset direction. Furthermore, the elastic material of the spherical soft rubber shell effectively buffers the impact of contact between the device and obstacles, preventing the device from getting stuck or stopping.

[0043] Specifically, the process of analyzing the linkage between interactive commands and motion states includes: Based on the stable posture and the unbiased and controllable rolling, multi-dimensional motion characteristics are extracted, and the motion rate, direction, and posture stability parameters of the spherical companion robot are obtained simultaneously to obtain the real-time comprehensive motion parameters of the spherical companion robot. By establishing a dynamic correlation between motion and interaction, a linkage mapping model between motion state and interactive action is established, resulting in motion-interaction linkage rules with motion adaptability. The interactive execution commands are analyzed in a contextualized manner, and the execution logic of the interactive commands is optimized in combination with the motion scenario to obtain the interactive analysis results that are adapted to the real-time motion status of the spherical companion robot.

[0044] In this embodiment, based on the stable and balanced posture and unbiased, controllable rolling state of the device, a motion feature extraction algorithm is used to extract multi-dimensional motion characteristics. Real-time motion speed, precise direction, and posture stability parameters (values ​​closer to 1 indicate greater stability) of the spherical companion robot are simultaneously acquired at a preset high-frequency acquisition frequency. These motion parameters are then fused and integrated to obtain the spherical companion robot's real-time comprehensive motion parameters. Through big data correlation analysis, a dynamic correlation between motion and interaction is established, constructing a linkage mapping model between motion state and interactive actions. The motion speed is set to be positively correlated with the operating frequencies of the speaker, microphone, and LED lights. The mapping relationship is established, and the faster the movement speed, the higher the frequency of audio-visual interaction, resulting in motion-interaction linkage rules with precise motion adaptation. The interaction execution commands are analyzed in a scenario-based manner, and the execution logic of the interaction commands is optimized in combination with different motion scenarios such as device forward movement, turning, obstacle avoidance, stationary position, and emergency stop. For example, in the obstacle avoidance scenario, the voice prompt volume of the speaker and microphone is automatically increased to above 80dB; in the turning scenario, the corresponding direction light of the LED light is triggered synchronously; and in the stationary scenario, a low-frequency soft audio-visual effect is triggered. The interaction analysis results are obtained that are fully adapted to the real-time motion state of the spherical companion robot, realizing deep linkage between motion and interaction.

[0045] Specifically, the process of triggering peripheral action timing planning based on the parsing results includes: Based on the interaction analysis results, the peripheral action type is identified, the scenario requirements for action execution are identified, and the peripheral type, action requirements and scenario adaptation parameters of the action to be executed are obtained. Simultaneously, the timing sequence of action execution is arranged, and the timing sequence of peripheral actions is matched with the movement rhythm of the spherical companion robot to obtain the timing sequence table of peripheral action execution. Based on the peripheral action execution timing table, the action triggering conditions are dynamically thresholded, the triggering threshold of motion state linkage is preset, and the linkage threshold action triggering command and execution timing parameters of the peripheral are obtained.

[0046] In this embodiment, the peripheral action type is accurately identified based on the interaction analysis results. A peripheral matching algorithm precisely identifies the peripherals requiring actions as one or a combination of two types: speakers and microphones, and LED lights. Simultaneously, the specific action requirements and scene adaptation parameters for each peripheral are defined, such as the volume adjustment range (0-100dB) and voice playback duration (0-10s) for speakers and microphones, and the flashing frequency (0-10Hz), color switching mode, and brightness adjustment range for LED lights. Furthermore, the action execution timing is finely arranged in conjunction with the real-time movement rhythm of the device. A timing matching algorithm precisely synchronizes the peripheral actions with the device movement, with a timing matching accuracy controlled to 10m. The system obtains a detailed and accurate timing table of peripheral actions, clarifying the start time, duration, linkage sequence, and interrupt switching mechanism of various actions of the speaker, microphone, and LED lights. Based on the timing table of peripheral actions, the system dynamically sets thresholds for action triggering conditions. Through a threshold setting algorithm, it presets trigger thresholds for linkage of motion states such as motion rate, posture stability, and direction change. For example, when the motion rate exceeds 0.3 m / s, it automatically triggers high-frequency flashing of the LED lights; when the posture stability parameter is lower than a certain threshold, it pauses the audio-visual interaction. The system obtains the linkage threshold action triggering instructions and precise execution timing parameters of the speaker, microphone, and LED lights, providing a basis for the precise triggering and execution of peripheral actions.

[0047] Specifically, the process by which the driver executes corresponding action instructions according to the timing sequence for the built-in peripheral includes: Based on the linkage threshold action trigger command and execution timing parameters, generate peripheral personalized drive signals, adjust the drive signal parameters according to the peripheral type and action requirements, and obtain the drive control signals of the built-in peripherals. Based on the drive control signal, peripheral synchronization adaptation is performed according to the timing table to obtain the real-time execution dynamics of peripheral synchronization. The peripheral device is used to synchronously and in real time to perform dynamic motion verification, and at the same time, the adaptability of the peripheral device's motion effect to the spherical companion robot's motion state is verified, and the peripheral device's motion parameters are dynamically corrected.

[0048] In this embodiment, based on the linkage threshold action trigger command and execution timing parameters, a signal generation module generates personalized drive signals for each peripheral device, which are then combined with the speaker, microphone, and LED. The system precisely adjusts the core parameters of the drive signals, such as voltage, current, and frequency, based on the different peripheral types and specific action requirements of the lights. For example, the drive voltage of the speaker and microphone is precisely controlled to be 3~5V and the audio signal frequency is 0~20kHz, while the drive frequency of the LED light is 0~20Hz and the working voltage is 3.3V, obtaining precise drive control signals for the built-in peripherals. Based on the drive control signals, the speaker, microphone, and LED light are synchronously adapted and controlled according to the peripheral action execution timing table. Through the synchronization control algorithm, the synchronization error of the peripheral actions is strictly controlled within 5ms, achieving seamless linkage between peripheral actions and device movement, and obtaining the dynamic real-time execution of peripheral synchronization. The dynamic real-time execution of peripheral synchronization is used to verify the action effect in real time, with a verification cycle of 20ms. The feedback module verifies the adaptability of the action effects of the speaker, microphone, and LED light with the real-time movement state of the spherical companion robot. When the adaptability is detected to be lower than 80%, the system automatically and dynamically corrects the peripheral action parameters according to the preset adjustment range, continuously optimizing the interaction effect, ensuring the deep and precise linkage between the peripheral interaction effect and the device movement state, and enhancing the fun of pet interaction.

[0049] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A spherical companion robot based on self-balancing technology, characterized by, include: Data acquisition unit, data processing unit, motion control unit, interactive execution unit; The data acquisition unit is used to acquire multi-dimensional raw data of the spherical companion robot; The data processing unit extracts the posture feature parameters of the spherical companion robot based on the multi-dimensional raw data; it decomposes the remote control commands into actions and performs parameterized parsing of the interaction configuration information to obtain the motion control logic and interaction execution requirements; and it obtains the hierarchical encoding of motion parameters by performing corresponding dynamic adaptation calculations. By utilizing the coordinated matching of the posture feature parameters and the motion control logic, self-balancing adjustment parameters and controllable motion drive commands are obtained; The motion control unit adjusts and calibrates the tilt posture based on the self-balancing control parameters and controllable motion drive commands to obtain the self-balancing state of the spherical companion robot; at the same time, it controls the power guide wheels to perform differential rotation and directional controllable rolling. The interactive execution unit performs interactive command and motion state linkage analysis on the self-balancing posture and controllable rolling motion, and triggers peripheral action timing planning based on the analysis results, driving the built-in peripheral to execute corresponding action commands according to the timing.

2. The system according to claim 1, characterized in that, The specific process for extracting the posture feature parameters of the spherical companion robot includes: Based on the multi-dimensional raw data fusion posture detection and dynamic prediction algorithm, feature recognition is performed to obtain the raw sensor data of the spherical companion robot's tilt angle, rotation speed and dynamic prediction. The original sensing data is subjected to adaptive dimension normalization processing to dynamically match the dimension of the motion state correction data of the spherical companion robot, thereby obtaining standardized posture data. By performing multi-feature extraction on the standardized posture data, posture parameters of the spherical companion robot containing dynamic prediction features are obtained.

3. The system according to claim 1, characterized in that, The specific process of decomposing remote control commands into action sequences includes: Based on the remote control command data in the multi-dimensional raw data, perform command semantic-action mapping, decode and associate command semantics with the executable actions of the spherical companion robot, and obtain the plaintext remote control command with action priority; The plaintext remote control command is decomposed into hierarchical action logic to obtain a single primitive action command and a combination rule for composite actions.

4. The system according to claim 1, characterized in that, The specific process of parametric parsing of the interactive configuration information includes: Based on the interactive configuration information in the multi-dimensional raw data, personalized information extraction and format translation are performed to extract user-defined interactive parameters; Record users' daily behavior data and synchronize it to the robot's accompanying intelligent computing chip for learning and storage, and obtain structured personalized interaction configuration information of user behavior characteristics; The structured personalized interaction configuration information is dynamically decoupled from parameters to obtain independent interaction modes for pet behavior adaptability and user growth adaptability.

5. The system according to claim 1, characterized in that, The specific process for obtaining the hierarchical encoding of motion parameters includes: By utilizing dynamic adaptation calculations under self-balancing constraints and combining them with the motion threshold adaptation parameters of the spherical companion robot, adaptive motion control parameters under self-balancing constraints are obtained. Two-dimensional motion control parameters are obtained through multi-dimensional parameter threshold calibration; The two-dimensional motion control parameters are processed by motion priority hierarchical sequencing, and the sequencing is distinguished by combining motion and auxiliary motion to obtain priority-identified motion parameter hierarchical sequencing data.

6. The system according to claim 1, characterized in that, The specific process of obtaining the self-balancing control parameters and controllable motion drive commands includes: Based on the hierarchical sequencing data of the motion parameters and the posture parameters, a real-time posture coordination matching calculation is performed, and a motion inertia compensation factor is introduced to obtain the initial self-balancing parameter adjustment with inertia compensation. By combining the initial self-balancing parameter adjustment with the driving parameters, self-balancing correction is performed, and the sensitivity of the control parameters is dynamically adjusted to obtain self-balancing parameter adjustments that adapt to the real-time motion state. By using the self-balancing adjustment and the driving parameters, instruction generation and redundancy configuration are performed to generate the main driving instruction and the backup fault-tolerant instruction, thereby obtaining a controllable driving instruction.

7. The system according to claim 1, characterized in that, The specific process of adjusting and calibrating the tilt attitude includes: Based on the self-balancing parameter adjustment combined with real-time sensing data from the electronic gyroscope, deviation anomalies are detected in the real-time tilting posture of the spherical companion robot to obtain posture tilt data. By dynamically adjusting the moving distance and speed of the driving counterweight, the posture of the spherical companion robot with dynamic center of gravity adjustment is obtained. Based on the spherical companion robot's posture combined with the gyroscope closed-loop verification algorithm, real-time posture calibration is performed. The actual posture is compared with the target balanced posture and dynamically corrected to obtain the spherical companion robot's balanced posture.

8. The system according to claim 1, characterized in that, The specific process of controlling the power guide wheel to rotate at differential speed includes: Based on the controllable drive command, the independent power distribution of the wheel set is analyzed, and the power distribution parameters of the power guide wheel with force compensation are obtained by combining the force state and movement direction of the guide wheel. By using the power distribution parameters to control the rotational speed and introducing a road surface friction coefficient compensation factor, the target rotational speed parameter value of the power guide wheel adapted to the corresponding contact surface is obtained. By performing frequency conversion independent speed tuning, the speed of the guide wheel is controlled by frequency conversion, thereby obtaining the self-adaptive differential rotation of the power guide wheel.

9. The system according to claim 1, characterized in that, The directional controllable scrolling is: By using the stable posture of the sphere and the differential rotation state of the power guide wheel to preset the dynamic trajectory, and combining environmental sensing data to avoid obstacles in real time, the directional rolling trajectory parameters of the spherical companion robot with obstacle avoidance strategy are obtained. The directional rolling trajectory parameters are combined with the differentially rotating power guide wheel for trajectory driving, and the wheel speed is adjusted in real time to match the trajectory parameters, so as to obtain the directional rolling driving state of the spherical companion robot; Based on the directional rolling dynamics, a self-balancing and trajectory dual-dimensional real-time verification is performed, and the rolling trajectory is dynamically adjusted to obtain a spherical companion robot that rolls smoothly and controllably along a preset direction without deviation.

10. The system according to claim 1, characterized in that, The specific process of analyzing the linkage between interactive commands and motion states includes: Based on the stable posture and the unbiased and controllable rolling, multi-dimensional motion characteristics are extracted, and the motion rate, direction, and posture stability parameters of the spherical companion robot are obtained simultaneously to obtain the real-time comprehensive motion parameters of the spherical companion robot. By establishing a dynamic correlation between motion and interaction, a linkage mapping model between motion state and interactive action is established, resulting in motion-interaction linkage rules with motion adaptability. The interactive execution commands are analyzed in a contextualized manner, and the execution logic of the interactive commands is optimized in combination with the motion scenario to obtain the interactive analysis results that are adapted to the real-time motion status of the spherical companion robot.

11. The system according to claim 1, characterized in that, The specific process of triggering peripheral action timing planning based on the parsing results includes: Based on the interaction analysis results, the peripheral action type is identified, the scenario requirements for action execution are identified, and the peripheral type, action requirements and scenario adaptation parameters of the action to be executed are obtained. Simultaneously, the timing sequence of action execution is arranged, and the timing sequence of peripheral actions is matched with the movement rhythm of the spherical companion robot to obtain the timing sequence table of peripheral action execution. Based on the peripheral action execution timing table, the action triggering conditions are dynamically thresholded, the triggering threshold of motion state linkage is preset, and the linkage threshold action triggering command and execution timing parameters of the peripheral are obtained.

12. The system according to claim 1, characterized in that, The specific process by which the driver's built-in peripherals execute corresponding action instructions according to the timing sequence includes: Based on the linkage threshold action trigger command and execution timing parameters, generate peripheral personalized drive signals, adjust the drive signal parameters according to the peripheral type and action requirements, and obtain the drive control signals of the built-in peripherals. Based on the drive control signal, peripheral synchronization adaptation is performed according to the timing table to obtain the real-time execution dynamics of peripheral synchronization. The peripheral device is used to synchronously and in real time to perform dynamic motion verification, and at the same time, the adaptability of the peripheral device's motion effect to the spherical companion robot's motion state is verified, and the peripheral device's motion parameters are dynamically corrected.

Images