Ball dribbling robot simulation bionic control system

Abstract

The invention relates to the technical field of dribbling robots, and one embodiment of the invention provides a dribbling robot simulation bionic control system, which comprises a core chip, a circuit, a controller and a sensor, the core chip is a high-performance microprocessor, has high-speed data processing capability, supports multi-channel ADC acquisition and CAN bus communication, and meets the real-time operation requirement of multi-dimensional data; the circuit comprises a safety protection circuit and a peripheral circuit. The safety protection circuit comprises a power supply, a storage unit and a communication interface; by means of the technical scheme, the problems that in the prior art, an early-stage control system mostly adopts a single microcontroller, the ball grabbing-moving-ball releasing process is completed by presetting a fixed action track code, a driving motor, an air cylinder and other execution components, for example, part of basketball dribbling training robots can only drive moving wheels to move linearly at the fixed rotating speed, and the operation efficiency is poor are solved. Opening and closing of the ball grabbing claws are controlled according to a fixed angle, action parameters cannot be dynamically adjusted, and only mechanical simulation of basic ball conveying action can be achieved.

Description

Technical Field

[0001] The embodiments disclosed herein relate to the field of ball-dribbling robot technology, and more specifically, to a ball-dribbling robot simulated bionic control system. Background Technology

[0002] This background technology involves the intersection of robot control technology, sports equipment technology, and biomechanics, specifically focusing on the development of the core control system for a dribbling posture mimicry robot. It is applicable to scenarios such as sports training and teaching, sports rehabilitation assistance, and demonstration of competitive movements. As sports equipment upgrades towards automation and biomimicry, and as sports rehabilitation increases the demand for standardized movement assistance, dribbling posture mimicry devices are evolving from early mechanical transmission types to intelligent biomimetic control types. As the core of the device, the control system directly determines the accuracy, stability, and scenario adaptability of dribbling posture mimicry. It is the key link connecting mechanical structure and biomimetic function, and its technical level directly affects the practical value and market competitiveness of the device.

[0003] Early control systems often used a single microcontroller, which drove motors, cylinders, and other actuators to complete the ball-grabbing-movement-releasing process by pre-setting fixed motion trajectory codes. For example, some basketball dribbling training robots could only drive the moving wheels to move linearly at a fixed speed and control the opening and closing of the gripper at a fixed angle. The motion parameters could not be dynamically adjusted, and they could only achieve mechanical imitation of basic dribbling actions. In recent years, some devices have introduced single sensor feedback, such as installing pressure sensors on the gripper to stop gripping when the detected pressure reaches a threshold, or installing speed sensors on the moving wheels to correct speed deviations in real time. Although such systems have basic feedback capabilities, they are still limited to linear logic of single parameter detection and single action adjustment, and have not formed a multi-dimensional data collaborative decision-making mechanism. Summary of the Invention

[0004] To overcome the above-mentioned defects, the embodiments of this disclosure provide a dribbling robot simulated bionic control system, which solves the technical problem that early control systems in the prior art mostly use a single microcontroller, which drives motors, cylinders and other actuators to complete the process of grabbing, moving and releasing the ball by preset fixed motion trajectory codes. For example, some basketball dribbling training robots can only drive the moving wheels to move linearly at a fixed speed and control the opening and closing of the grabbing claw at a fixed angle. The motion parameters cannot be dynamically adjusted, and they can only achieve mechanical imitation of basic dribbling actions.

[0005] According to one aspect, at least one embodiment of the present disclosure provides a dribbling robot simulated bionic control system, comprising:

[0006] Core chips, circuits, controllers, and sensors;

[0007] The core chip is a high-performance microprocessor with high-speed data processing capabilities, supporting multi-channel ADC acquisition and CAN bus communication to meet the needs of real-time multi-dimensional data processing.

[0008] The circuit includes a safety protection circuit and peripheral circuits;

[0009] The safety protection circuit includes a power supply, a storage unit, and a communication interface;

[0010] The power supply uses a step-down chip to convert the 12V input to 3.3V to power the chip, and is filtered by tantalum capacitors to ensure stable power supply.

[0011] The storage unit is externally connected to NAND Flash and is used to store biomechanical mapping models, motion parameter libraries and historical data, and supports data retention even when power is off.

[0012] The communication interfaces include: 1 CAN bus, 1 USB Type-C port, and 1 Ethernet port.

[0013] Among them, the circuit is responsible for the core calculation of dynamic collaborative decision-making, receives sensor data uploaded by the secondary controller, generates the master-level action command, and synchronously sends the fine-tuning-level action command to the secondary controller.

[0014] As a further technical solution, the controller includes a main controller and a secondary controller;

[0015] The main controller uses a microprocessor dedicated to data acquisition. The main controller is a dedicated information acquisition controller, and its components are as follows:

[0016] Pressure sensor interface: 3-channel high-precision ADC, corresponding to 3 sets of distributed pressure sensors in the gripper, which collect pressure distribution data through differential signals;

[0017] Gyroscope interface: 1-channel SPI communication, connecting the gyroscope and accelerometer to collect ball-catching frame attitude data;

[0018] Friction sensor interface: 1-channel I2C communication, connecting to the friction detection chip, installed on the wheel surface of the mobile wheel, to collect the ground friction coefficient in real time;

[0019] Data transmission: The collected pressure distribution, attitude, and friction coefficient data are packaged and uploaded to the main controller via the CAN bus, and data frames are transmitted periodically.

[0020] The secondary controller is an execution drive controller, and the control components of the secondary controller are as follows:

[0021] Motor drive: It adopts a motor drive chip to control 4 moving wheel motors, adjusts the speed through PWM signal, and supports forward and reverse rotation and braking functions;

[0022] Cylinder drive: A solenoid valve is used to control the extension and retraction of the telescopic cylinder through the IO port, and a pressure sensor is used to detect the cylinder air pressure to ensure accurate ball gripping force;

[0023] Damper control: The current damping value of the elastic damper is acquired by ADC, and the damping coefficient is adjusted by PWM control of the servo motor;

[0024] Functional positioning: Receive fine-tuning action commands from the main controller, drive the execution components to complete the actions, and synchronously feed back the execution status to the main controller.

[0025] As a further technical solution, the sensor uses a binocular camera, combined with an image sensor, to support the recognition of surface features of a sphere;

[0026] Circuit design: Connects to the main controller via an interface, equipped with an independent power supply, and an external infrared fill light (optional, suitable for low-light environments).

[0027] Functionality: Real-time image capture of the sphere's surface; using a surface wear recognition algorithm built into the main controller, the degree of wear on the sphere is determined, providing a basis for adjusting the damping coefficient.

[0028] As a further technical solution, the relevant components of the peripheral circuit are as follows:

[0029] Overcurrent protection: A self-resetting fuse is connected in series in the 24V motor power supply circuit. When the motor stalls and the current becomes too high, the fuse will trip, protecting the motor and the drive chip.

[0030] Overvoltage protection: It adopts a reference voltage chip and an overvoltage detection circuit to cut off the 24V power supply circuit when the input voltage is too high;

[0031] Emergency stop circuit: An independent emergency stop button is provided, which is directly connected to the external interrupt pin of the main controller. When pressed, it immediately triggers the power-off of all actuators to avoid danger.

[0032] As a further technical solution, the sensor signal line and the motor power line are wired separately with a spacing of ≥5mm to avoid electromagnetic interference; the CAN bus uses twisted-pair cable with parallel terminating resistors at both ends to reduce signal reflection; the sensor module and the motor drive module are respectively encapsulated in aluminum shielding boxes with a grounding resistance ≤1Ω to reduce the influence of external electromagnetic radiation; the sensor data uses a 3-sample averaging algorithm to remove outliers; CRC check is added to the CAN bus data frame to ensure error-free data transmission.

[0033] As a further technical solution, the biomechanical data mapping method is as follows:

[0034] S1. Sample Collection: 120 subjects were recruited. Wrist / arm force data were collected using electromyography sensors, dribbling posture was captured by high-speed cameras, and interaction data between different ground surfaces and the ball was recorded simultaneously to form a sample dataset.

[0035] S2. Feature Extraction: The sample data is processed using professional software to extract three core features: force distribution features, posture coordination features, and environment adaptation features.

[0036] S3. Model Training: The BP neural network algorithm is used to train a 3D mapping model with user features and environmental features as inputs and mechanical motion parameters as outputs.

[0037] S4. Model Storage: The trained model is stored in the main controller's storage unit and can be retrieved according to scene type.

[0038] As a further technical solution, its multi-dimensional data collaborative decision-making method is as follows:

[0039] S1. Data preprocessing: The main controller receives four types of data uploaded by the secondary controller 1: pressure distribution, attitude data, friction coefficient, and ball wear level. Noise is removed by Kalman filtering to obtain a clean data matrix.

[0040] S2. Weight Allocation: Dynamically adjust data weights based on scenario type, for example:

[0041] S3, Rehabilitation Training Scenario: Prioritize safety, with the following weights: friction coefficient 30%, pressure distribution 40%, posture data 20%, and ball wear 10%.

[0042] S4. Professional Training Scenario: Prioritize ensuring biomimetic accuracy, with posture data weight set at 40%, pressure distribution weight at 30%, friction coefficient weight at 20%, and sphere wear weight at 10%.

[0043] S5. Instruction Generation: Substitute the weighted data into the biomechanical mapping model to generate two types of instructions:

[0044] S6, Master-level commands: Caster wheel speed, lifting frame height;

[0045] S7, Fine-tuning level command: Claw opening and closing compensation amount, damping coefficient;

[0046] S8. Command Issuance: The main controller synchronously issues commands via the CAN bus. The main level commands directly control the main drive module, while the fine-tuning level commands are sent to the secondary controller.

[0047] As a further technical solution, its hierarchical driving method is as follows:

[0048] S1. After the main controller issues the master level command, the secondary controller first drives the motor drive chip to control the rotation of the moving wheel motor, and at the same time drives the electric push rod to adjust the height of the lifting frame.

[0049] S2. After the main action is started, the secondary controller drives the servo to adjust the opening and closing compensation of the gripper, and simultaneously collects the current value of the damper through the ADC. After comparing with the command value, the servo angle is adjusted to realize the fine-tuning action.

[0050] S3. During the hierarchical drive process, the secondary controller collects motor speed and cylinder position data in real time and feeds them back to the main controller, forming a drive-feedback closed loop.

[0051] As a further technical solution, the dynamic correction method is as follows: if the feedback data shows that the speed deviation of the moving wheel is large, the main controller immediately increases the PWM duty cycle to correct the speed; if the pressure sensor detects that the pressure distribution deviation is large, the main controller issues a fine-tuning command to adjust the pressure on one side of the ball gripper; when the ground friction coefficient suddenly decreases, the main controller simultaneously increases the power of the moving wheel and the gripping force to prevent the ball from slipping.

[0052] The beneficial effects of the embodiments disclosed herein are as follows:

[0053] In this disclosure, by utilizing palm-fingertip pressure distribution data collected by distributed pressure sensors and combining it with the dynamic adjustment of elastic dampers, the fine-tuning force exerted by the human wrist as the ball rotates is simulated. This upgrades the ball-gripping claw from rigid opening and closing to flexible wrapping, improving the matching degree between ball control actions and the actual force exertion logic of humans. Through the linkage of gyroscope and binocular vision sensors, the coordinated relationship between arm rise and fall, movement speed, and ball rotation is captured. The execution components are controlled according to the natural rhythm of human dribbling, avoiding the fragmented feeling of independent actions of each component in existing systems. The visual similarity of the simulated posture is significantly improved, making it a digital coach for standardized teaching demonstrations and solving the problem of inconsistent actions in human demonstrations. Attached Figure Description

[0054] To more clearly illustrate the technical solutions in the embodiments of this disclosure, the accompanying drawings used in the description of the embodiments of this disclosure will be briefly introduced below. Obviously, the drawings described below are merely some exemplary embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on the content of the exemplary embodiments of this disclosure and these drawings without any creative effort.

[0055] Figure 1 This is a schematic diagram of a structure in one embodiment of the present disclosure;

[0056] Figure 2 This is an overview diagram of the overall control method flow of this disclosure;

[0057] Figure 3This is a flowchart of the biomechanical data mapping method disclosed herein;

[0058] Figure 4 This is a flowchart of the multi-dimensional data collaborative decision-making method disclosed herein;

[0059] Figure 5 This is a flowchart of the hierarchical driving and dynamic correction method disclosed herein;

[0060] Figure 6 This is a flowchart of the scenario iteration optimization method disclosed herein; Detailed Implementation

[0061] The present disclosure will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present disclosure and are not intended to limit the scope of the disclosure.

[0062] To keep the drawings concise, each drawing only schematically shows the parts relevant to the disclosure; these do not represent the actual structure of the product. Furthermore, for ease of understanding, in some drawings, only one of components with the same structure or function is schematically shown, or only one is labeled. In this document, "one" not only means "only one," but can also mean "more than one," and "several" includes "two" and "more than two."

[0063] In this document, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linkage" should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; they can refer to mechanical connections or electrical connections; they can refer to direct connections or indirect connections through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this disclosure based on the specific circumstances.

[0064] In this disclosure, unless otherwise expressly specified and limited, "above" or "below" the second feature can include direct contact between the first and second features, or contact between the first and second features through another feature between them. Furthermore, "above," "over," and "on top" of the second feature includes the first feature directly above or diagonally above the second feature, or simply indicates that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature includes the first feature directly below or diagonally below the second feature, or simply indicates that the first feature is at a lower horizontal level than the second feature.

[0065] In the description of this embodiment, terms such as "upper," "lower," "left," and "right" are based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of description and simplification of operation, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this disclosure.

[0066] Furthermore, in the description of this application, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0067] like Figures 1-6 As shown, it illustrates a dribbling robot simulated bionic control system of the present disclosure, characterized in that it includes:

[0068] Core chips, circuits, controllers, and sensors;

[0069] The core chip is a high-performance microprocessor with high-speed data processing capabilities, supporting multi-channel ADC acquisition and CAN bus communication to meet the needs of real-time multi-dimensional data processing.

[0070] The circuit includes a safety protection circuit and peripheral circuits;

[0071] The safety protection circuit includes a power supply, a storage unit, and a communication interface;

[0072] The power supply uses a step-down chip to convert the 12V input to 3.3V to power the chip, and is filtered by tantalum capacitors to ensure stable power supply.

[0073] The storage unit is externally connected to NAND Flash and is used to store biomechanical mapping models, motion parameter libraries and historical data, and supports data retention even when power is off.

[0074] The communication interfaces include: 1 CAN bus, 1 USB Type-C port, and 1 Ethernet port.

[0075] Among them, the circuit is responsible for the core calculation of dynamic collaborative decision-making, receives sensor data uploaded by the secondary controller, generates the master-level action command, and synchronously sends the fine-tuning-level action command to the secondary controller.

[0076] like Figures 1-6 As shown in the figure, this embodiment proposes that the controller includes a main controller and a secondary controller;

[0077] The main controller uses a microprocessor dedicated to data acquisition. The main controller is a dedicated information acquisition controller, and its components are as follows:

[0078] Pressure sensor interface: 3-channel high-precision ADC, corresponding to 3 sets of distributed pressure sensors in the gripper, which collect pressure distribution data through differential signals;

[0079] Gyroscope interface: 1-channel SPI communication, connecting the gyroscope and accelerometer to collect ball-catching frame attitude data;

[0080] Friction sensor interface: 1-channel I2C communication, connecting to the friction detection chip, installed on the wheel surface of the mobile wheel, to collect the ground friction coefficient in real time;

[0081] Data transmission: The collected pressure distribution, attitude, and friction coefficient data are packaged and uploaded to the main controller via the CAN bus, and data frames are transmitted periodically.

[0082] The secondary controller is an execution drive controller, and the control components of the secondary controller are as follows:

[0083] Motor drive: It adopts a motor drive chip to control 4 moving wheel motors, adjusts the speed through PWM signal, and supports forward and reverse rotation and braking functions;

[0084] Cylinder drive: A solenoid valve is used to control the extension and retraction of the telescopic cylinder through the IO port, and a pressure sensor is used to detect the cylinder air pressure to ensure accurate ball gripping force;

[0085] Damper control: The current damping value of the elastic damper is acquired by ADC, and the damping coefficient is adjusted by PWM control of the servo motor;

[0086] Functional positioning: Receive fine-tuning action commands from the main controller, drive the execution components to complete the actions, and synchronously feed back the execution status to the main controller.

[0087] like Figures 1-6 As shown in the figure, this embodiment proposes that the sensor be a binocular camera, combined with an image sensor, to support the recognition of surface features of a sphere;

[0088] Circuit design: Connects to the main controller via an interface, equipped with an independent power supply, and an external infrared fill light (optional, suitable for low-light environments).

[0089] Functionality: Real-time image capture of the sphere's surface; using a surface wear recognition algorithm built into the main controller, the degree of wear on the sphere is determined, providing a basis for adjusting the damping coefficient.

[0090] like Figures 1-6 As shown in the figure, the relevant components of the peripheral circuit proposed in this embodiment are as follows:

[0091] Overcurrent protection: A self-resetting fuse is connected in series in the 24V motor power supply circuit. When the motor stalls and the current becomes too high, the fuse will trip, protecting the motor and the drive chip.

[0092] Overvoltage protection: It adopts a reference voltage chip and an overvoltage detection circuit to cut off the 24V power supply circuit when the input voltage is too high;

[0093] Emergency stop circuit: An independent emergency stop button is provided, which is directly connected to the external interrupt pin of the main controller. When pressed, it immediately triggers the power-off of all actuators to avoid danger.

[0094] like Figures 1-6 As shown, this embodiment proposes separate wiring for the sensor signal line and the motor power line, with a spacing of ≥5mm, to avoid electromagnetic interference; the CAN bus uses twisted-pair cable with parallel terminating resistors at both ends to reduce signal reflection; the sensor module and the motor drive module are respectively encapsulated in aluminum shielding boxes with a grounding resistance ≤1Ω to reduce the influence of external electromagnetic radiation; the sensor data uses a 3-sample averaging algorithm to remove outliers; and CRC check is added to the CAN bus data frame to ensure error-free data transmission.

[0095] like Figures 1-6 As shown in the figure, this embodiment proposes a biomechanical data mapping method as follows:

[0096] S1. Sample Collection: 120 subjects were recruited. Wrist / arm force data were collected using electromyography sensors, dribbling posture was captured by high-speed cameras, and interaction data between different ground surfaces and the ball was recorded simultaneously to form a sample dataset.

[0097] S2. Feature Extraction: The sample data is processed using professional software to extract three core features: force distribution features, posture coordination features, and environment adaptation features.

[0098] S3. Model Training: The BP neural network algorithm is used to train a 3D mapping model with user features and environmental features as inputs and mechanical motion parameters as outputs.

[0099] S4. Model Storage: The trained model is stored in the main controller's storage unit and can be retrieved according to scene type.

[0100] like Figures 1-6 As shown in the figure, this embodiment proposes a multi-dimensional data collaborative decision-making method as follows:

[0101] S1. Data preprocessing: The main controller receives four types of data uploaded by the secondary controller 1: pressure distribution, attitude data, friction coefficient, and ball wear level. Noise is removed by Kalman filtering to obtain a clean data matrix.

[0102] S2. Weight Allocation: Dynamically adjust data weights based on scenario type, for example:

[0103] S3, Rehabilitation Training Scenario: Prioritize safety, with the following weights: friction coefficient 30%, pressure distribution 40%, posture data 20%, and ball wear 10%.

[0104] S4. Professional Training Scenario: Prioritize ensuring biomimetic accuracy, with posture data weight set at 40%, pressure distribution weight at 30%, friction coefficient weight at 20%, and sphere wear weight at 10%.

[0105] S5. Instruction Generation: Substitute the weighted data into the biomechanical mapping model to generate two types of instructions:

[0106] S6, Master-level commands: Caster wheel speed, lifting frame height;

[0107] S7, Fine-tuning level command: Claw opening and closing compensation amount, damping coefficient;

[0108] S8. Command Issuance: The main controller synchronously issues commands via the CAN bus. The main level commands directly control the main drive module, while the fine-tuning level commands are sent to the secondary controller.

[0109] like Figures 1-6 As shown, this embodiment proposes a hierarchical driving method as follows:

[0110] S1. After the main controller issues the master level command, the secondary controller first drives the motor drive chip to control the rotation of the moving wheel motor, and at the same time drives the electric push rod to adjust the height of the lifting frame.

[0111] S2. After the main action is started, the secondary controller drives the servo to adjust the opening and closing compensation of the gripper, and simultaneously collects the current value of the damper through the ADC. After comparing with the command value, the servo angle is adjusted to realize the fine-tuning action.

[0112] S3. During the hierarchical drive process, the secondary controller collects motor speed and cylinder position data in real time and feeds them back to the main controller, forming a drive-feedback closed loop.

[0113] like Figures 1-6 As shown, this embodiment proposes the following dynamic correction method: If the feedback data shows a large deviation in the rotational speed of the moving wheel, the main controller immediately increases the PWM duty cycle to correct the rotational speed; if the pressure sensor detects a large deviation in pressure distribution, the main controller issues a fine-tuning command to adjust the pressure on one side of the gripping claw; when the ground friction coefficient suddenly decreases, the main controller simultaneously increases the power of the moving wheel and the gripping force to prevent the ball from slipping.

[0114] When in use, the system automatically starts after power-on. The main controller, two secondary controllers, and all sensors and actuators complete self-checks to confirm normal hardware connections. If there are no faults, it proceeds to the next step; otherwise, a fault triggers a buzzer alarm. The main controller loads the pre-stored biomechanical mapping model and, based on the scene type and sphere parameters input by the user through the accompanying terminal, calls the corresponding action reference parameters. The secondary controller calibrates each sensor, completing zero-point calibration and accuracy initialization. The secondary controller initializes the actuators such as motors, cylinders, and servos, resetting the ball gripper, lifting frame, and moving wheels to their initial positions. The power supply system achieves stable dual-voltage output, the lithium battery pack enters working mode, and the system is ready.

[0115] After the ball collection component transports the simulated ball to the preset gripping position, the position sensor sends a signal to the secondary controller. The secondary controller then initiates multi-dimensional data acquisition. The pressure sensor collects the pressure distribution data of the ball in initial contact with the gripping claw, the gyroscope collects the current attitude angle of the gripping frame, the friction sensor collects the current ground friction coefficient, and the binocular vision sensor captures the surface of the ball to identify the wear level. The secondary controller packages all the collected data and uploads it to the main controller in real time via the CAN bus to form an initial data matrix.

[0116] The main controller preprocesses the received initial data, removes noise through Kalman filtering to obtain clean data, dynamically assigns data weights according to preset scenario types, substitutes the weighted data into the biomechanical mapping model, and generates dominant-level instructions and fine-tuning-level instructions. The main controller synchronously sends instructions through the CAN bus. Dominant-level instructions are directly transmitted to the main drive module, and fine-tuning-level instructions are sent to the slave controller.

[0117] After receiving the command, the secondary controller first executes the master-level action, driving the motor drive chip to control the moving wheel to rotate at the target speed, and simultaneously driving the electric push rod to adjust the lifting frame to the target height, achieving synchronous start. After the master-level action is started, the fine-tuning action is executed after a delay, driving the servo to adjust the opening and closing compensation of the ball gripper, so that the ball gripper and the ball form a close fit; simultaneously, the current value of the damper is collected through the ADC, and the servo angle is adjusted after comparing it with the target value to ensure that the damping coefficient is adapted. During the ball handling process, the secondary controller collects the execution status data in real time and continuously feeds it back to the main controller, forming a closed loop.

[0118] The main controller compares the feedback data with the target command value in real time to determine if there is a deviation. If the deviation of the moving wheel speed exceeds the threshold, it immediately adjusts the PWM duty cycle to correct the speed. If the pressure sensor detects that the pressure distribution deviation is too large, it sends a fine-tuning command to adjust the pressure on one side of the ball gripper to ensure that the ball is subjected to uniform force. If the ground friction coefficient changes suddenly, it synchronously increases the power output of the moving wheel and the gripping force to prevent the ball from slipping or deviating. If the binocular vision sensor detects that the ball deviates beyond the safe range, it adjusts the height of the lifting frame and the angle of the gripper to pull the ball back to the preset trajectory. All correction commands are quickly sent through the CAN bus, and the secondary controller responds in milliseconds to ensure that the ball's posture is stable and the movement trajectory is accurate during ball handling.

[0119] After a single preset dribbling action is completed, the system automatically records the action parameters, ball status, and user feedback. The recorded data is stored in the main controller's storage unit. Once the accumulated data reaches a certain threshold, the system initiates iterative optimization. It uses algorithms to identify high-frequency deviation data, analyzes the causes of deviations, and adjusts the parameters of the biomechanical mapping model based on the causes. The optimized model automatically overwrites the old model, providing more accurate instructions for the next run. If the user chooses to continue, the system returns to the ball positioning and sensing stage, repeats the above process, and executes the next dribbling action. If the user chooses to stop, the system drives all execution components to reset, cuts off unnecessary power supply circuits, and enters standby mode.

[0120] If the safety protection circuit detects overcurrent, overvoltage, or if the emergency stop button is pressed, it immediately triggers an emergency stop command. The main controller instantly cuts off the power supply to all actuators, the moving wheels brake, the ball gripper releases, the lifting frame resets to its lowest height, the buzzer sounds an alarm to indicate the abnormality, and fault data is recorded for subsequent troubleshooting to ensure the safety of the equipment and the user.

[0121] It should be noted that the above embodiments are only used to illustrate the technical solutions of this disclosure and are not intended to limit it. Although this disclosure has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of this disclosure without departing from the spirit and scope of the technical solutions of this disclosure, and all such modifications and substitutions should be covered within the scope of the claims of this disclosure.

Claims

1. A dribbling robot simulated bionic control system, characterized in that, include: Core chips, circuits, controllers, and sensors; The core chip is a high-performance microprocessor with high-speed data processing capabilities, supporting multi-channel ADC acquisition and CAN bus communication to meet the needs of real-time multi-dimensional data processing. The circuit includes a safety protection circuit and peripheral circuits; The safety protection circuit includes a power supply, a storage unit, and a communication interface; The power supply uses a step-down chip to convert the 12V input to 3.3V to power the chip, and is filtered by tantalum capacitors to ensure stable power supply. The storage unit is externally connected to NAND Flash and is used to store biomechanical mapping models, motion parameter libraries and historical data, and supports data retention even when power is off. The communication interfaces include: 1 CAN bus, 1 USB Type-C port, and 1 Ethernet port. Among them, the circuit is responsible for the core calculation of dynamic collaborative decision-making, receives sensor data uploaded by the secondary controller, generates the master-level action command, and synchronously sends the fine-tuning-level action command to the secondary controller.

2. The controller according to claim 1, characterized in that, The controller includes a main controller and a secondary controller; The main controller uses a microprocessor dedicated to data acquisition. The main controller is a dedicated information acquisition controller, and its components are as follows: Pressure sensor interface: 3-channel high-precision ADC, corresponding to 3 sets of distributed pressure sensors in the gripper, which collect pressure distribution data through differential signals; Gyroscope interface: 1-channel SPI communication, connecting the gyroscope and accelerometer to collect ball-catching frame attitude data; Friction sensor interface: 1-channel I2C communication, connecting to the friction detection chip, installed on the wheel surface of the mobile wheel, to collect the ground friction coefficient in real time; Data transmission: The collected pressure distribution, attitude, and friction coefficient data are packaged and uploaded to the main controller via the CAN bus, and data frames are transmitted periodically. The secondary controller is an execution drive controller, and the control components of the secondary controller are as follows: Motor drive: It adopts a motor drive chip to control 4 moving wheel motors, adjusts the speed through PWM signal, and supports forward and reverse rotation and braking functions; Cylinder drive: A solenoid valve is used to control the extension and retraction of the telescopic cylinder through the IO port, and a pressure sensor is used to detect the cylinder air pressure to ensure accurate ball gripping force; Damper control: The current damping value of the elastic damper is acquired by ADC, and the damping coefficient is adjusted by PWM control of the servo motor; Functional positioning: Receive fine-tuning action commands from the main controller, drive the execution components to complete the actions, and synchronously feed back the execution status to the main controller.

3. The sensor according to claim 1, characterized in that, The sensor uses a binocular camera, combined with an image sensor, to support the recognition of surface features of a sphere. Circuit design: Connects to the main controller via an interface, equipped with an independent power supply, and an external infrared fill light (optional, suitable for low-light environments). Functionality: Real-time image capture of the sphere's surface; using a surface wear recognition algorithm built into the main controller, the degree of wear on the sphere is determined, providing a basis for adjusting the damping coefficient.

4. The circuit according to claim 1, characterized in that, The relevant components of the peripheral circuit are as follows: Overcurrent protection: A self-resetting fuse is connected in series in the 24V motor power supply circuit. When the motor stalls and the current becomes too high, the fuse will trip, protecting the motor and the drive chip. Overvoltage protection: It adopts a reference voltage chip and an overvoltage detection circuit to cut off the 24V power supply circuit when the input voltage is too high; Emergency stop circuit: An independent emergency stop button is provided, which is directly connected to the external interrupt pin of the main controller. When pressed, it immediately triggers the power-off of all actuators to avoid danger.

5. The ball-driving robot simulated bionic control system according to claim 1, characterized in that, The sensor signal lines and motor power lines are wired separately with a spacing of ≥5mm to avoid electromagnetic interference. The CAN bus uses twisted-pair cables with parallel terminating resistors at both ends to reduce signal reflection. The sensor module and motor drive module are respectively encapsulated in aluminum shielding boxes with a grounding resistance ≤1Ω to reduce the influence of external electromagnetic radiation. The sensor data uses a 3-sample averaging algorithm to eliminate outliers. CRC checks are added to the CAN bus data frames to ensure error-free data transmission.

6. The ball-driving robot simulated bionic control system according to claim 1, characterized in that, The biomechanical data mapping method is as follows: S1. Sample Collection: 120 subjects were recruited. Wrist / arm force data were collected using electromyography sensors, dribbling posture was captured by high-speed cameras, and interaction data between different ground surfaces and the ball was recorded simultaneously to form a sample dataset. S2. Feature Extraction: The sample data is processed using professional software to extract three core features: force distribution features, posture coordination features, and environment adaptation features. S3. Model Training: The BP neural network algorithm is used to train a 3D mapping model with user features and environmental features as inputs and mechanical motion parameters as outputs. S4. Model Storage: The trained model is stored in the main controller's storage unit and can be retrieved according to scene type.

7. The ball-driving robot simulated bionic control system according to claim 1, characterized in that, Its multi-dimensional data collaborative decision-making method is as follows: S1. Data preprocessing: The main controller receives four types of data uploaded by the secondary controller 1: pressure distribution, attitude data, friction coefficient, and ball wear level. Noise is removed by Kalman filtering to obtain a clean data matrix. S2. Weight Allocation: Dynamically adjust data weights based on scenario type, for example: S3, Rehabilitation Training Scenario: Prioritize safety, with the following weights: friction coefficient 30%, pressure distribution 40%, posture data 20%, and ball wear 10%. S4. Professional Training Scenario: Prioritize ensuring biomimetic accuracy, with posture data weight set at 40%, pressure distribution weight at 30%, friction coefficient weight at 20%, and sphere wear weight at 10%. S5. Instruction Generation: Substitute the weighted data into the biomechanical mapping model to generate two types of instructions: S6, Master-level commands: Caster wheel speed, lifting frame height; S7, Fine-tuning level command: Claw opening and closing compensation amount, damping coefficient; S8. Command Issuance: The main controller synchronously issues commands via the CAN bus. The main level commands directly control the main drive module, while the fine-tuning level commands are sent to the secondary controller.

8. The ball-driving robot simulated bionic control system according to claim 1, characterized in that, Its hierarchical driving method is as follows: S1. After the main controller issues the master level command, the secondary controller first drives the motor drive chip to control the rotation of the moving wheel motor, and at the same time drives the electric push rod to adjust the height of the lifting frame. S2. After the main action is started, the secondary controller drives the servo to adjust the opening and closing compensation of the gripper, and simultaneously collects the current value of the damper through the ADC. After comparing with the command value, the servo angle is adjusted to realize the fine-tuning action. S3. During the hierarchical drive process, the secondary controller collects motor speed and cylinder position data in real time and feeds them back to the main controller, forming a drive-feedback closed loop.

9. The hierarchical driving method according to claim 8, characterized in that, The dynamic correction method is as follows: if the feedback data shows that the speed deviation of the moving wheel is large, the main controller immediately increases the PWM duty cycle to correct the speed; if the pressure sensor detects that the pressure distribution deviation is large, the main controller issues a fine-tuning command to adjust the pressure on one side of the ball gripper; when the ground friction coefficient suddenly decreases, the main controller simultaneously increases the power of the moving wheel and the gripping force to prevent the ball from slipping.

Images