A complementary robotic arm system

Abstract

The application discloses a kind of Kang Fu mechanical arm systems, comprising: mechanical arm;The movement end of mechanical arm is provided with gripper;Mechanical arm drives gripper to move in three-dimensional space;Posture sensor and electromyographic signal collector are both arranged on arm;Obtain the posture of arm movement, and solve the posture of arm movement, obtain Euler angle, and after angle conversion, to drive mechanical arm follow-up;While user grip intention generates, electromyographic signal is collected, and according to electromyographic signal, gripper is driven to open, close, cooperates with mechanical arm to complete life auxiliary task.The application deeply fuses electromyographic intention recognition and angle sensor space mapping, breaks through the limitation of traditional single control mode, and reproduces human natural motor nerve conduction logic.

Description

Technical Field

[0001] This invention relates to the field of rehabilitation robotic arm technology, specifically a rehabilitation robotic arm system. Background Technology

[0002] Existing products often pursue high industrial-grade precision, without considering the physiological characteristics of pathological tremors and unstable motor output in patients in the early stages of rehabilitation. This makes them prone to accidental triggering of movement due to limb tremors.

[0003] Limitations of traditional single control modes: Pure electromyography control schemes are easily affected by environmental interference and muscle artifacts, resulting in misjudgment and false triggering; Pure posture control schemes lack grasping flexibility and cannot adapt to the user's personalized grasping action needs, making it difficult for elderly patients and patients in the early stages of rehabilitation to use independently.

[0004] High-end medical rehabilitation equipment costs tens of thousands of yuan and is complex to operate, making it only suitable for hospital settings and unaffordable for ordinary families. Low-end simple assistive devices are inexpensive but lack intelligence and feedback mechanisms, resulting in limited rehabilitation effects and failing to meet patients' dual needs for scientific rehabilitation and daily living assistance. There is a huge market gap for mid-range, intelligent, and full-cycle home rehabilitation equipment priced in the hundreds of yuan.

[0005] Patent application CN117484489A discloses a robotic arm control method, control device, electronic device, and storage medium. The method determines the user's activity intention based on facial electromyography (EMG) signals. Based on the target position of the target object and the activity intention, the robotic arm is controlled to complete a target task, which is to reach the target position and grasp or release the target object. This solution uses eye-tracking signals to locate the object and facial EMG signals to determine the intention. It is suitable for patients with high-level spinal cord injuries who cannot move their upper limbs and have difficulty with daily activities such as eating and picking up objects. However, it is not suitable for hospital-based professional rehabilitation training and home-based daily living assistance for patients with upper limb dysfunction such as stroke sequelae, post-operative upper limb injuries, or Parkinson's disease. Summary of the Invention

[0006] The technical problem to be solved by this invention is that single electromyography control is difficult to express the spatial movement of a robotic arm, and single posture control cannot express the intention to grasp or release. This invention provides a hybrid intuitive control paradigm of electromyography and posture control that is suitable for patients with upper limb dysfunction.

[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution: A rehabilitation robotic arm system includes: a robotic arm 10; a gripper 20 is provided at the end of the robotic arm 10; the robotic arm 10 drives the gripper 20 to move in three-dimensional space; a pose sensor and an electromyography signal acquisition device are both installed on the arm; The arm movement pose is acquired and calculated to obtain Euler angles. After angle transformation, the robotic arm 10 is driven to follow the movement. At the same time, when the user's grasping intention is generated, electromyographic signals are collected, and the gripper 20 is driven to open and close according to the electromyographic signals, so as to complete the life assistance task in coordination with the robotic arm 10. The intensity of electromyographic signals is monitored and visualized for use in rehabilitation training tasks.

[0008] Furthermore, the robotic arm 10 includes a fixed base 11, a rotating base 12 mounted on the fixed base 11, a rocker arm 13, and a large arm 14; the rocker arm 13 and the large arm 14 are mounted on the rotating base 12, and the three are driven by independent servo motors to achieve three-dimensional spatial movement.

[0009] Furthermore, servo motor one controls the rotating base 12 to achieve XOY plane rotation, servo motor two controls the swing of the large arm 14 to achieve Y-axis movement, servo motor three controls the rocker arm 13 to pull to achieve Z-axis movement, and servo motor four is set on the gripper 20 to control the opening and closing of the gripper 20.

[0010] Furthermore, the Euler angles are obtained by calculating the arm's pose during movement, including: Pitch=asin(-2×q1×q3+2×q0×q2); Roll=atan2(2×q2×q3+2×q0×q1,-2×q1×q1-2×q2×q2+1); Yaw=atan2(2×(q1×q2+q0×q3),q0×q0+q1×q1-q2×q2-q3×q3); In the formula, Pitch, Roll, and Yaw are pitch angle, roll angle, and yaw angle, respectively; asin is the arcsine function; atan2 is the four-quadrant arctangent function; and q0, q1, q2, and q3 are the four data points collected by the pose sensor.

[0011] Furthermore, the angle converted from yaw angle is used to drive the rotating base 12; the angle converted from pitch angle is used to drive the boom 14; and the angle converted from roll angle is used to drive the rocker arm 13 to pull.

[0012] Furthermore, the servo motor 1 drives the rotating base 12 to rotate via a transmission belt, and adopts a speed reduction and torque amplification structure of "small gear driving large gear" to amplify the output torque of the servo motor 1.

[0013] Furthermore, each time a user's grasping intention is generated, an electromyographic signal is counted. The first time an electromyographic signal is received, it triggers the gripper 20 to open. The next time an electromyographic signal is received, it triggers the gripper 20 to close, creating a cycle that allows the gripper 20 to grasp.

[0014] Furthermore, before the electromyographic signal triggers the gripper 20 to grasp, a four-stage filtering and conditioning process is performed, including: capacitive coupling and voltage divider circuit, rectifier circuit, low-pass filter circuit, and dynamic gain adjustment circuit. The system includes a capacitive coupling and voltage divider circuit to isolate the DC component on the surface of human skin, initially attenuate power frequency interference, and adjust the signal amplitude; a rectifier circuit that uses a combination of dual operational amplifiers and diodes to convert bidirectional fluctuating AC electromyographic signals into unidirectional full-wave DC signals, thus processing muscle artifacts caused by strong burst electrical activity in non-target muscles; a low-pass filter circuit that filters out high-frequency noise in the rectified signal and smooths the waveform; and a dynamic gain adjustment circuit that uses a non-inverting amplifier circuit to adjust the gain of the smoothed signal.

[0015] Further rehabilitation training tasks include: visualizing electromyographic signals so that patients can see the relative magnitude of their grasping electromyographic activity; and adaptively adjusting the grasping trigger threshold based on the rehabilitation progress.

[0016] Furthermore, it is applicable to patients with upper limb dysfunction.

[0017] Compared with the prior art, the beneficial effects of the present invention are: Single electromyographic control is insufficient to express the spatial movement of a robotic arm, while single posture control cannot express the intention to grasp or release. This invention decouples the control task into two categories: posture angles control only servos 1 to 3 to complete the spatial movement of the robotic arm, and electromyographic intentions control only servo 4 to open or close the gripper. Thus, the user can control the robotic arm to approach the target using the natural posture of the upper arm or residual limb, and then switch to the grasping state through a single effective muscle contraction, forming a control paradigm that conforms to human intuition. The user's arm posture changes are mapped to the target angles of the robotic arm base, upper arm, and forearm servos, and the effective electromyographic intentions are mapped to the opening or closing state of the gripper, achieving decoupled control of spatial movement and grasping action.

[0018] The four-stage hardware filtering and conditioning circuit filters out the three core noises of electromyography signals through four cascaded processes: "coupling voltage division → precision rectification → low-pass filtering → amplification and adjustment," significantly improving the signal-to-noise ratio.

[0019] With zero-learning-cost operation that allows patients with upper limb dysfunction to independently grasp and deliver daily items, they can regain their ability to take care of themselves.

[0020] Visualization of electromyographic signals generates positive stimulation, adapting to rehabilitation training needs throughout the entire cycle, quantifying rehabilitation progress, and optimizing treatment plans.

[0021] The "belt drive + speed reduction and torque amplification" mechanical structure not only provides sufficient torque for daily grasping but also creates a natural mechanical overload protection mechanism. Through the selection of general-purpose components and structural optimization, the overall hardware cost is reduced to the level of hundreds of yuan, breaking the price barrier of high-end rehabilitation equipment and realizing the widespread availability of the product.

[0022] A system architecture of "wireless sensing - wired execution" is constructed, which separates the control end and the execution end through the wireless module, giving users maximum freedom of action, while the wired execution end ensures the stability of the robotic arm's movements.

[0023] The rehabilitation robotic arm is suitable for professional hospital rehabilitation training and home daily life assistance for patients with sequelae of stroke, postoperative upper limb trauma, Parkinson's disease and other upper limb dysfunctions. It has extremely high clinical application value and market promotion prospects. Attached Figure Description

[0024] Figure 1 This is a flowchart of a rehabilitation robotic arm system according to an embodiment of the present invention.

[0025] Figure 2 , 3 These are schematic diagrams of the robotic arm at different angles in embodiments of the present invention.

[0026] Figure 4 This is a schematic diagram of a servo motor according to an embodiment of the present invention.

[0027] Figure 5 This is a schematic diagram of the gripper according to an embodiment of the present invention.

[0028] Figure 6 This is a schematic diagram of the system principle according to an embodiment of the present invention.

[0029] Figure 7 This is a four-stage filtering and conditioning circuit according to an embodiment of the present invention. Detailed Implementation

[0030] To facilitate understanding of the technical solution of the present invention by those skilled in the art, the technical solution of the present invention will now be further described in conjunction with the accompanying drawings.

[0031] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0032] Please see Figure 1As shown, this invention provides a rehabilitation robotic arm system, including: a robotic arm 10, with a gripper 20 at the end of the robotic arm 10. The robotic arm 10 drives the gripper 20 to move in three-dimensional space. A pose sensor and an electromyography (EMG) signal acquisition device are both mounted on the arm. The arm's motion pose is acquired, Euler angles are calculated, and after angle conversion, the robotic arm 10 is driven to follow the pose. Simultaneously, when the user's grasping intention is generated, EMG signals are acquired, and the gripper 20 is driven to open and close based on the EMG signals, coordinating with the robotic arm 10 to complete daily living assistance tasks. The intensity of the EMG signals is monitored and visualized for rehabilitation training tasks. This invention's "EMG-gyroscope" hybrid intuitive control paradigm creatively integrates EMG intention recognition with spatial mapping of angle sensors, breaking through the limitations of traditional single control modes and replicating the natural motor nerve conduction logic of the human body.

[0033] Please see Figures 2 to 5 As shown, in one embodiment of the present invention, the robotic arm 10 is based on the classic industrial ABB IRB460 robotic arm, and is reconstructed and optimized for desktop miniaturization. It adopts a "3+1 degree of freedom" kinematic architecture, that is, the three degrees of freedom of the arm body are precisely mapped to the three-dimensional coordinates of X, Y and Z in space, and the gripper has one independent degree of freedom focused on the grasping action, so as to achieve the decoupling design of position control and action execution.

[0034] In this embodiment, the robotic arm 10 includes a fixed base 11, a rotating base 12 mounted on the fixed base 11, a rocker arm 13, a large arm 14, a small arm 15, a triangular connector 16, and a connecting rod 17. The rocker arm 13 and the large arm 14 are mounted on the rotating base 12, and each of the three is driven by an independent servo motor to achieve three-dimensional spatial movement. The triangular connector 16 is connected to the large arm 14 and moves synchronously with the large arm 14. Furthermore, the rocker arm 13, the small arm 15, the triangular connector 16, and the connecting rod 17 are connected to the gripper 20.

[0035] In this embodiment, servo motor one controls the rotating base 12 to achieve rotation in the X0Y plane, servo motor two controls the swing of the upper arm 14 to achieve movement in the Y-axis direction, and servo motor three controls the pull of the rocker arm 13 to achieve movement in the Z-axis direction. By adjusting the rotation angle of the servo motors, the rocker arm 13 and the upper arm 14 drive the upper arm 14 and the forearm 15 to swing back and forth, realizing multi-dimensional movement of the robotic arm in the vertical plane. The gripper 20 is equipped with servo motor four to control the opening and closing of the gripper 20. Servo motor four drives the drive gear to rotate, which drives the meshing driven structure to achieve stable opening and closing of the gripper 20, completing the grasping and release of the target object.

[0036] In this embodiment, the transmission device of the rotating base 12 is housed within the fixed base 11. Its bearings bear the radial load, reducing rotational friction and ensuring smooth rotation and positioning accuracy of the rotating end. The servo motor 1 achieves efficient torque transmission via a transmission belt, driving the rotating chassis to rotate flexibly. It employs a "small gear driving a large gear" reduction and torque amplification structure to amplify the output torque of the servo motor 1, reduce the force loss of the servo motor under the same load, and extend the service life of the servo motor. The transmission system uses a belt drive scheme, utilizing the flexibility of the belt drive to construct a natural mechanical overload protection mechanism. When the robotic arm accidentally collides with a person or obstacle during operation, the elastic slippage of the belt can effectively absorb the impact energy, preventing damage to the core servo motor and ensuring the personal safety of the user.

[0037] In this embodiment, the STM32 integrated main control board is based on a high-performance STM32 microcontroller and integrates an MP1584EN DC-DC step-down module, an xl4015 step-down module, an MPU6050 angle sensor, an NRF24L01 2.4G wireless module, a multi-channel ADC acquisition unit, and a DMA (Direct Memory Access) module. It is the control center of the entire system.

[0038] The step-down module stably converts the input power supply voltage to the operating voltage required by each module in the system, ensuring the power supply stability of the system under high load concurrent conditions of multiple servos. The MPU6050 angle sensor (pose sensor) integrates a three-axis gyroscope and a three-axis accelerometer, which can capture the user's arm's movement, pitch, and other posture data in space in real time. Through the built-in DMP (Digital Motion Processor) library, it completes sensor fusion calculation and outputs Pitch / Roll / Yaw Euler angles, providing accurate angle basis for the spatial motion control of the robotic arm. The DMP library is a rigorously verified and highly optimized motion processing driver middleware (Embedded Motion Driver) provided by InvenSense. The NRF24L01 wireless module establishes a 2.4GHz bidirectional communication link to realize a "wireless sensing-wired execution" system architecture. It reliably transmits angle control commands, gripper status, power information, and other data between the control end and the robotic arm execution end. It is equipped with an overtime protection mechanism, which triggers a safety stop program if there is no valid data for 1 second, ensuring the safety and stability of wireless control. The ADC acquisition channel, in conjunction with DMA technology, enables high-frequency real-time sampling of electromyographic signals, ensuring real-time response to control commands.

[0039] In this embodiment, the DMP inside the MPU6050 fuses raw data from the three-axis gyroscope and three-axis accelerometer in real time, and directly calculates quaternions using its internal attitude calculation algorithm. 16 bytes of quaternion data are read from the MPU6050's FIFO buffer, with each 4 bytes forming a long integer. Each quaternion component is divided by the constant q30 to convert the fixed-point number back to a unit floating-point quaternion with a modulus of 1. After obtaining the floating-point quaternion, the system performs Euler angle calculation according to the ZYX rotation order (i.e., first around the Z-axis, then around the Y-axis, and finally around the X-axis). Pitch=asin(-2×q1×q3+2×q0×q2); Roll=atan2(2×q2×q3+2×q0×q1,-2×q1×q1-2×q2×q2+1); Yaw=atan2(2×(q1×q2+q0×q3),q0×q0+q1×q1-q2×q2-q3×q3); In the formula, Pitch, Roll, and Yaw are pitch angle, roll angle, and yaw angle, respectively; asin is the arcsine function; atan2 is the four-quadrant arctangent function; and q0, q1, q2, and q3 are the four data points collected by the pose sensor.

[0040] All calculation results are converted to radians, then multiplied by 57.3 (180 / π) to convert to degrees, resulting in pitch (-90°~+90°), roll (-180°~+180°), and yaw (-180°~+180°). These ranges are then defined by angular correspondence with the actual fixed positions of the servos. The converted yaw angle drives the rotating base 12; the converted pitch angle drives the boom 14; and the converted roll angle drives the joystick 13.

[0041] In this embodiment, the embedded control system adopts a modular architecture design based on the STM32 microcontroller platform. The core components include a sensor data acquisition module, a data processing and filtering module, a wireless communication module, and a servo motor control module.

[0042] Sensor data acquisition module: It acquires acceleration and gyroscope data from MPU6050 via I²C bus (fused and solved by DMP library), and simultaneously acquires filtered electromyographic (EMG) SIG signals and power supply voltage signals via ADC and DMA, so that EMG intention and attitude angle data can be fused into a single input within the same control cycle.

[0043] Data processing and filtering module: Angle mapping + software amplitude limiting, first-order low-pass filter + dynamic dead-zone filtering (anti-tremor), EMG signal threshold conversion + dynamic threshold adjustment (anti-repeated triggering). During initialization, the system acquires the user's resting EMG baseline value V0 and noise fluctuation range ΔV. During runtime, it calculates the difference between the current EMG signal and the baseline, and extracts the average amplitude, peak value, or root mean square feature value within a preset sampling window. A valid capture intent is determined only when the EMG feature exceeds the dynamic threshold determined by V0, ΔV, and rehabilitation stage parameters for N consecutive sampling cycles. After a valid intent is identified, the system enters a time dead zone to prevent the same muscle contraction or pathological tremor from being repeatedly triggered.

[0044] Wireless communication module: NRF24L01 chip + ShockBurst™ protocol, data packets are sent in packets, and the receiver verifies and discards invalid commands to avoid malfunctions. The NRF24L01 module encapsulates attitude angle data, electromyography status, and timestamps for transmission. After verifying the data validity, the receiver outputs four PWM control signals via an STM32 timer. Servos one through three use step interpolation to achieve smooth coordination, while servo four uses a soft start-stop curve to open or close. When communication is interrupted or data is abnormal, the system triggers a safety stop logic.

[0045] Servo control module: The STM32 timer outputs four PWM signals (corresponding to four degrees of freedom), and the interpolation algorithm achieves smooth coordination, enabling the gripper to start and stop gradually and accurately reproduce user intentions. When the user naturally positions the arm, the system records Pitch0, Roll0, and Yaw0 as the zero-position attitude, representing Euler angles describing spatial attitude: pitch (rotation about the lateral axis), roll (rotation about the longitudinal axis), and yaw (rotation about the vertical axis), respectively. During runtime, the system calculates the changes in pitch angle ΔPitch, roll angle ΔRoll, and yaw angle ΔYaw, and after first-order low-pass filtering, dynamic dead zone, direction correction, scaling mapping, and angle limiting, it generates the target angles for servo motors one through three, solving the problem of inconsistency between the human arm coordinate system, the sensor mounting coordinate system, and the robotic arm joint coordinate system.

[0046] Please see Figure 6 As shown, in one embodiment of the present invention, after the system starts up, it first completes hardware initialization, and then enters a dual-mode (life assistance task + rehabilitation training task) working state. The transmitting end continuously collects posture data and electromyographic signals, which are then filtered, packaged, and transmitted wirelessly. After transmission, there is a short delay before new data is collected again in a loop to provide the system with stable user action input. The receiving end continuously listens for wireless data. If valid data is received, the timestamp is verified, and then the servo motor is controlled to perform the action and update the timestamp. If no data is received within 1 second, a safety stop is immediately triggered to avoid malfunction of the device.

[0047] In this embodiment, the robotic arm 10 supports two functions in parallel: daily living assistance and rehabilitation training. When the user's arm moves, the gyroscope collects posture data, which is then calculated by the STM32 and driven to follow the movement of the robotic arm. At the same time, when the user makes a grasping intention, the electromyographic signal is collected and filtered to drive the gripper 20, together completing daily living assistance tasks such as grasping and delivering everyday items. In the rehabilitation training scenario, after the user makes a grasping intention, the electromyographic signal collector (electrode pads) collects the electromyographic signal, which is then filtered by hardware. On the one hand, it drives the gripper 20 to complete the training action, and on the other hand, it monitors the intensity of the electromyographic signal and provides visual feedback, ultimately completing the rehabilitation training task and providing data basis for adjusting the rehabilitation plan.

[0048] In this embodiment, the electrode pads collect and transmit electromyographic signals. After the signals are processed by the filtering circuit to reduce noise, they are transmitted to the STM32 main control chip. Under the life assistance function, each electromyographic signal is counted as a valid signal. The first time the STM32 receives a valid signal, it triggers the "opening" control of the gripper. The next time it receives a valid signal, it triggers the "closing" control of the gripper, creating a cycle that allows the gripper to grasp the patient, enabling the patient to pick up the medicine box and hold the water cup.

[0049] Under the rehabilitation training function, when a valid signal is received, the electromyographic signal will be visualized and digitized, allowing the patient to see the relative magnitude of their grasping electromyographic signal. This function not only allows the patient to see that they are recovering and gain confidence, but also allows the vendor to record the patient's recovery status and continuously adjust the grasping trigger threshold to avoid the situation where the mechanical gripper is accidentally triggered due to the failure to adjust the control threshold in time as the patient's electromyographic signal recovers.

[0050] Please see Figure 7 As shown in this embodiment, the electromyography (EMG) acquisition and filtering module uses a purchased four-stage hardware filtering and conditioning circuit, which consists of a capacitive coupling and voltage divider circuit, a rectifier circuit, a low-pass filter circuit, and a dynamic gain adjustment circuit cascaded together. It is specifically designed for the weak surface EMG signal (mV level) and its susceptibility to environmental noise interference, and can effectively filter out the three core noises: baseline drift, electric field interference, and muscle artifacts. After the system powers on, the STM32 main control board initializes the electromyography (EMG) acquisition circuit. Electrodes attached to the user's forearm acquire raw EMG signals. The signals are transmitted via leads to the filtering module, where they undergo four stages of processing: The first stage, a capacitive coupling and voltage divider circuit, isolates the DC component from the skin surface, initially attenuating power frequency interference and adjusting the signal amplitude. The second stage, a precision rectifier circuit, uses a combination of dual operational amplifiers and diodes to convert the bidirectional fluctuating AC EMG signal into a unidirectional full-wave DC signal, addressing muscle artifacts caused by strong, sudden electrical activity in non-target muscles. The third stage, a low-pass filter circuit, filters out high-frequency noise from the rectified signal, smooths the waveform, and resolves baseline drift. The fourth stage, a dynamic gain adjustment circuit, uses a non-inverting amplifier circuit to adjust the gain of the smoothed signal, amplifying the weak signal to the standard voltage range that the STM32 ADC can acquire, and outputs a stable SIG signal to the main control board. When the SIG voltage exceeds a set threshold, the main control board determines it as a valid grasping intent from the user.

[0051] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention, and no reference numerals in the claims should be construed as limiting the scope of the claims.

[0052] The above-described embodiments are merely examples of implementation methods of the invention. The scope of protection of the present invention is not limited to the above-described embodiments. For those skilled in the art, several modifications and improvements can be made without departing from the concept of the present invention, and these all fall within the scope of protection of the present invention.

Claims

1. A rehabilitation robotic arm system, characterized in that, include: Robotic arm (10); the end of the robotic arm (10) is equipped with a gripper (20); the robotic arm (10) drives the gripper (20) to move in three-dimensional space; both the pose sensor and the electromyography signal collector are mounted on the arm; The arm movement pose is obtained and the arm movement pose is calculated to obtain Euler angles. After the angle is transformed, the robotic arm (10) is driven to follow the movement. At the same time, when the user's grasping intention is generated, electromyographic signals are collected, and the gripper (20) is driven to open and close according to the electromyographic signals, so as to cooperate with the robotic arm (10) to complete the life assistance task. The intensity of electromyographic signals is monitored and visualized for use in rehabilitation training tasks.

2. The rehabilitation robotic arm system according to claim 1, characterized in that, The robotic arm (10) includes a fixed base (11), a rotating base (12) set on the fixed base (11), a rocker arm (13), and a large arm (14); the rocker arm (13) and the large arm (14) are set on the rotating base (12), and the three are driven by independent servo motors to achieve three-dimensional spatial movement.

3. The rehabilitation robotic arm system according to claim 2, characterized in that, Servo 1 controls the rotating base (12) to achieve XOY plane rotation, servo 2 controls the arm (14) to swing to achieve Y-axis movement, servo 3 controls the rocker arm (13) to pull to achieve Z-axis movement, and servo 4 is set on the gripper (20) to control the opening and closing of the gripper (20).

4. The rehabilitation robotic arm system according to claim 2, characterized in that, The calculation of arm pose and the acquisition of Euler angles include: Pitch=asin(-2×q1×q3+2×q0×q2); Roll=atan2(2×q2×q3+2×q0×q1,-2×q1×q1-2×q2×q2+1); Yaw=atan2(2×(q1×q2+q0×q3),q0×q0+q1×q1-q2×q2-q3×q3); In the formula, Pitch, Roll, and Yaw are pitch angle, roll angle, and yaw angle, respectively; asin is the arcsine function; atan2 is the four-quadrant arctangent function; and q0, q1, q2, and q3 are the four data points collected by the pose sensor.

5. The rehabilitation robotic arm system according to claim 4, characterized in that, The angle converted from yaw angle is used to drive the rotating base (12); the angle converted from pitch angle is used to drive the boom (14); and the angle converted from roll angle is used to drive the rocker arm (13) to pull.

6. The rehabilitation robotic arm system according to claim 3, characterized in that, Servo motor 1 drives the rotating base (12) to rotate via a transmission belt. It adopts a speed reduction and torque amplification structure of "small gear driving large gear" to amplify the output torque of servo motor 1.

7. The rehabilitation robotic arm system according to claim 1, characterized in that, Each time a user's grasping intention is generated, it is counted as one electromyographic signal. The first time an electromyographic signal is received, it triggers the gripper (20) to open. The next time an electromyographic signal is received, it triggers the gripper (20) to close, creating a cycle that allows the gripper (20) to grasp.

8. The rehabilitation robotic arm system according to claim 7, characterized in that, Before the electromyographic signal triggers the gripper (20) to grasp, four levels of filtering and conditioning are performed, including: capacitive coupling and voltage divider circuit, rectifier circuit, low-pass filter circuit, and dynamic gain adjustment circuit. The system includes a capacitive coupling and voltage divider circuit to isolate the DC component on the surface of human skin, initially attenuate power frequency interference, and adjust the signal amplitude; a rectifier circuit that uses a combination of dual operational amplifiers and diodes to convert bidirectional fluctuating AC electromyographic signals into unidirectional full-wave DC signals, thus processing muscle artifacts caused by strong burst electrical activity in non-target muscles; a low-pass filter circuit that filters out high-frequency noise in the rectified signal and smooths the waveform; and a dynamic gain adjustment circuit that uses a non-inverting amplifier circuit to adjust the gain of the smoothed signal.

9. The rehabilitation robotic arm system according to claim 1, characterized in that, The rehabilitation training tasks include: visualizing and digitizing electromyographic signals so that patients can see the relative magnitude of their grasping electromyographic signals; and adaptively adjusting the grasping trigger threshold according to the rehabilitation progress.

10. The rehabilitation robotic arm system according to claim 1, characterized in that, Suitable for patients with upper limb dysfunction.

Images