Flexible finger posture detection glove and method based on posture sensor

Abstract

The invention discloses a flexible finger posture detection glove based on a posture sensor and a method. The flexible finger posture detection glove comprises a glove and a flexible circuit board, the flexible circuit board is arranged on the back face of the glove and comprises attitude sensor chips, a single chip microcomputer, a converter and a data interface, the second section and the third section of each finger are each provided with one attitude sensor chip, the hand back is provided with one attitude sensor chip, the attitude sensor chips communicate with the single chip microcomputer, and the data interface is used for signal output. The finger posture detection method comprises the steps of calibration, joint rotation axis calculation, real-time posture collection and joint rotation angle calculation. According to the invention, the subtle actions of the hand can be accurately captured by using the differential technology, and the detection accuracy and the response speed are effectively improved.

Description

Technical Field

[0001] The present invention relates to a detection device and method, in particular to a flexible finger posture detection glove and method based on a posture sensor, belonging to the technical field of equipment identification. Background Art

[0002] With the advancement of technology, the market has increasingly higher requirements for the accuracy and portability of hand gesture detection. In the field of hand gesture detection technology, traditional gesture recognition systems are often limited by detection accuracy, device size and ease of use. Summary of the invention

[0003] In order to overcome the existing technology, the present invention provides a flexible finger posture detection glove and method based on a posture sensor, which can accurately capture subtle hand movements by using differential technology, effectively improving the detection accuracy and response speed.

[0004] The flexible finger posture detection glove based on posture sensor includes a glove and a flexible circuit board; the flexible circuit board is arranged on the back of the glove, the flexible circuit board includes a posture sensor chip, a single-chip microcomputer, a converter and a data interface, a posture sensor chip is arranged on the second section and the third section of each finger respectively, and a posture sensor chip is arranged on the back of the hand, the posture sensor chip communicates with the single-chip microcomputer, and the data interface is used for signal output.

[0005] Furthermore, the posture sensor chip is an MPU9250 sensor.

[0006] Furthermore, the attitude sensor chip communicates with the single-chip microcomputer via an SPI bus.

[0007] Furthermore, the attitude sensor chip converts the TTL level serial port signal of the microcontroller into an RS485 signal through a 485 converter.

[0008] The present application also provides a flexible finger posture detection method, which is a flexible finger posture detection glove based on a posture sensor, and the method comprises:

[0009] S1. Calibration

[0010] When used for the first time, the attitude sensor chip is calibrated, and the initial attitude angle of each attitude sensor chip in the attitude is obtained by setting a static reference attitude;

[0011] S2. Calculate the joint rotation axis:

[0012] After the calibration is completed, the rotation axis of the joint is determined, and the direction of the rotation axis is determined by the relative position of the two attitude sensors;

[0013] S3, real-time posture collection:

[0014] When the finger moves, the gesture sensor chip collects its angle gesture data in real time;

[0015] S4, joint rotation angle calculation: Use the real-time posture data collected by the two-section posture sensor chip to calculate the finger joint rotation angle, specifically:

[0016] S41. Calculate Euler angles from quaternions

[0017] Use the quaternion output by the attitude sensor chip to calculate the Euler angles (Pitch, Roll, Yaw) of the joints. Assume that the quaternion q = (w, x, y, z), then the Euler angle is calculated by the following formula:

[0018] X-axis rotation: Roll = atan2(2(wx+yz), 1-2(x 2 +y 2 ))

[0019] Y-axis rotation: Pitch = asin(2(wy-zx))

[0020] Z axis rotation: Yaw = atan2(2(wz+xy), 1-2(y 2 +z 2 ))

[0021] S42, joint rotation angle calculation

[0022] Calculate the relative rotation using quaternion multiplication:

[0023]

[0024] Among them, the initial quaternion q 0 Indicates the posture of the sensor when the joint is not initially rotated; the current quaternion q t Indicates the posture of the sensor at the current time point; is the inverse of the quaternion, obtained by conjugation calculation:

[0025]

[0026] Using the Euler angle formula in step S41, the relative rotation quaternion q relative Converted to the rotation angle of the joint.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] 1. Significantly improved accuracy and sensitivity: The present invention achieves accurate capture of hand movements, including small gesture changes and complex hand gestures, by placing gesture sensors at key parts of the hand. The application of differential technology further enhances the sensitivity of detection and ensures the accuracy of the data.

[0029] Second, wearing comfort and flexibility are greatly improved: Using FPC flexible circuit board technology, the sensor and circuit board design of the present invention can bend and deform freely with the fingers, providing a wearing experience that is perfectly coordinated with hand movements. This design reduces the discomfort that may be caused by traditional rigid sensors, allowing users to use the device naturally and comfortably for a long time.

[0030] 3. High integration and low power consumption advantages: The MPU9250 sensor has high integration and occupies a small space. Combined with the intelligent power management strategy, it achieves low power consumption operation, extends the use time of the equipment, and reduces energy consumption.

[0031] Fourth, the algorithm ensures the accurate capture of hand movements and improves the system's sensitivity to subtle movements through the differential method. At the same time, the high-precision posture sensor provides accurate three-dimensional angle information, further optimizing the calculation of joint rotation angles, making the device show excellent sensitivity and accuracy in application scenarios such as virtual reality and rehabilitation training.

[0032] 5. Wide range of application fields: The innovative design of the present invention has broad application potential in many fields such as virtual reality, game control, sports training, rehabilitation training, health testing, etc., providing users with a variety of usage scenarios.

[0033] 6. Easy maintenance and cost-effectiveness: The modular design simplifies the maintenance process and reduces maintenance costs. The easy replacement of sensors ensures the long-term stable operation of the gloves, while reducing the replacement costs caused by failures.

[0034] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments: BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a schematic diagram of a flexible finger posture detection glove based on a posture sensor of the present invention.

[0036] Among them, A, gloves, B, flexible circuit board, B1, posture sensor chip, B2, single-chip microcomputer, B3, converter, B4, data interface. DETAILED DESCRIPTION

[0037] The embodiments of the technical solution of the present invention will be described in detail below in conjunction with the accompanying drawings. Unless otherwise specified, the technical terms or scientific terms used in this application should have the common meanings understood by those skilled in the art to which the present invention belongs.

[0038] Figure 1A finger posture detection glove is shown, comprising a glove A and a flexible circuit board B; the flexible circuit board B is attached to the back of the glove A, and the flexible circuit board B comprises a posture sensor chip B1, a single-chip computer B2, a 485 converter B3 and a data interface B4. A posture sensor chip B1 is arranged on the second and third sections of each finger, and a posture sensor chip B1 is arranged on the back of the hand. The posture sensor chip B1 communicates with the single-chip computer B2 through the SPI bus, and converts the serial port signal of the single-chip computer TTL level into an RS485 signal through the 485 converter B3. The data interface B4 is used for signal output.

[0039] Optionally, the flexible circuit board B is attached to the back of the glove A. The gesture sensor chip B1 is an MPU9250 sensor, and an innovative hand gesture detection glove based on the MPU9250 sensor. MPU9250 provides new possibilities for the development of gesture recognition technology with its 9-axis motion tracking capability, 3×3×1mm micro package, low power operation, and integrated digital motion processor (DMP). The glove A is a rubber glove, and this embodiment provides an efficient and reliable hand motion capture solution.

[0040] This embodiment places MPU9250 sensors at key locations on each finger, using differential technology to accurately capture subtle hand movements, effectively improving detection accuracy and response speed. In addition, the use of FPC flexible circuit board technology allows the sensor unit to bend and deform freely with the finger, greatly improving the user's wearing comfort and the practicality of the device.

[0041] The innovation of this embodiment lies in its unique sensor layout, which not only solves the accuracy and portability problems in traditional gesture recognition, but also achieves flexible fit and stable operation of the device by introducing advanced FPC technology. In addition, by fusing data from multiple sensors, the overall performance and reliability of hand gesture recognition are improved. The market application prospects are broad, and it is expected to play an important role in multiple fields such as virtual reality, game control, and health monitoring.

[0042] Based on the above ideas, the following feasible design solutions are given:

[0043] Weight: about 50g, ensuring the lightness of wearing and reducing fatigue after long-term use.

[0044] Size: 160mm×80mm, compact design, suitable for most people's hands, easy to wear and operate.

[0045] Sensor accuracy: gyroscope accuracy ±1° / s, accelerometer accuracy ±0.02g, magnetometer accuracy ±0.5μT, which ensures accurate capture of hand movements.

[0046] Battery life: Under normal conditions of use, the battery life can reach more than 8 hours, meeting the needs of long-term use.

[0047] This embodiment provides a flexible finger posture detection method based on a posture sensor. In order to further improve the accuracy of finger posture detection, posture sensors (such as MPU9250) arranged at both ends of the finger joints are used to obtain the three-axis posture angle information to calculate the rotation angle of the joint. The specific algorithm steps are as follows:

[0048] S1. Calibration

[0049] When used for the first time, the attitude sensor chip is calibrated, and the initial attitude angle of each attitude sensor chip in the attitude is obtained by setting a static reference attitude;

[0050] Specifically: obtain the initial pitch angle, roll angle and yaw angle of each attitude sensor; set the angle of the finger in a static and straight state as the reference zero point;

[0051] S2. Calculate the joint rotation axis:

[0052] After the calibration is completed, the rotation axis of the joint is determined, and the direction of the rotation axis is determined by the relative position of the two attitude sensors;

[0053] S3, real-time posture collection:

[0054] When the finger moves, the gesture sensor chip collects its angle gesture data in real time;

[0055] The angle attitude sensor collected is processed using a Kalman filter to eliminate noise and ensure smooth and reliable data;

[0056] S4, joint rotation angle calculation: Use the real-time posture data collected by the two-section posture sensor chip to calculate the finger joint rotation angle, specifically:

[0057] S41. Calculate Euler angles from quaternions

[0058] Use the quaternion output by the attitude sensor chip to calculate the Euler angles (Pitch, Roll, Yaw) of the joints. Assume that the quaternion q = (w, x, y, z), then the Euler angle is calculated by the following formula:

[0059] X-axis rotation: Roll = atan2(2(wx+yz), 1-2(x 2 +y 2 ))

[0060] Y-axis rotation: Pitch = asin(2(wy-zx))

[0061] Z axis rotation: Yaw = atan2(2(wz+xy), 1-2(y 2 +z 2 ))

[0062] Among them: w is the real part, x, y, z are the imaginary parts, which can represent the axis and angle of rotation

[0063] Pitch, Roll, and Yaw represent the pitch angle, roll angle, and yaw angle, respectively;

[0064] S42, joint rotation angle calculation

[0065] Calculate the relative rotation using quaternion multiplication:

[0066]

[0067] Among them, the initial quaternion q 0 Indicates the posture of the sensor when the joint is not initially rotated; the current quaternion q t Indicates the posture of the sensor at the current time point; is the inverse of the quaternion, obtained by conjugation calculation:

[0068]

[0069] Using the Euler angle formula in step S41, the relative rotation quaternion q relative Converted to the rotation angle of the joint.

[0070] Step S5, real-time update and feedback: By solving the joint rotation angle, the finger posture data is updated in real time, and the result is transmitted to the host computer for posture judgment and feedback. If real-time visualization is required, the algorithm can provide a data stream of angle changes to drive the virtual hand model or for fine control in robot manipulation.

[0071] The algorithm ensures accurate capture of hand movements and improves the system's sensitivity to subtle movements through the differential method. At the same time, the high-precision posture sensor provides accurate three-dimensional angle information, further optimizing the calculation of joint rotation angles, making the device show excellent sensitivity and accuracy in application scenarios such as virtual reality and rehabilitation training.

[0072] Calculation example:

[0073] Assume that the quaternion q of the attitude sensor is in the initial position 0 =(1, 0, 0, 0) (no rotation), the quaternion of the attitude sensor at a certain moment is:

[0074] q t =(0.9239, 0.3827, 0, 0), which means a 45 degree rotation around the X axis (i.e., a Pitch angle of 45 degrees);

[0075]

[0076] The Pitch angle is calculated by the Euler angle formula; Pitch = asin(2(0.9239·0-0·0.3827)) = 45°;

[0077] Finally, the joint's rotation angle is 45 degrees.

[0078] Application Examples

[0079] 1. Medical rehabilitation: This product can be widely used in the field of medical rehabilitation, especially for rehabilitation training of patients with impaired hand motor function. By accurately capturing hand movements and postures, it provides doctors with data support for patients' hand mobility, assists in formulating personalized rehabilitation plans, and monitors rehabilitation progress.

[0080] 2. Virtual Reality: In the virtual reality industry, this product can be used as an interactive device to provide a more natural and intuitive gesture control method. By wearing this glove, users can interact with the virtual environment more realistically and enhance the immersive experience.

[0081] 3. Robot remote control: This product can be used to remotely control robots, especially in situations where delicate operations are required. The operator's hand movements can be converted into robot movements in real time, achieving precise control of the robot.

[0082] The present invention has been disclosed as above with preferred implementation cases, but it is not used to limit the present invention. Any technician familiar with the profession can make slight changes or modifications to equivalent implementation cases with equivalent changes by using the above-disclosed structures and technical contents without departing from the scope of the technical solution of the present invention, which still fall within the scope of the technical solution of the present invention.

Claims

1. A flexible finger posture detection glove based on a posture sensor, characterized in that: The invention comprises a glove (A) and a flexible circuit board (B); the flexible circuit board (B) is arranged on the back of the glove (A); the flexible circuit board (B) comprises a posture sensor chip (B1), a single chip computer (B2), a converter (B3) and a data interface (B4); a posture sensor chip (B1) is arranged on the second section and the third section of each finger respectively, and a posture sensor chip (B1) is arranged on the back of the hand; the posture sensor chip (B1) communicates with the single chip computer (B2); and the data interface (B4) is used for signal output.

2. The flexible finger posture detection glove based on posture sensor according to claim 1, characterized in that: The posture sensor chip (B1) is an MPU9250 sensor.

3. The flexible finger posture detection glove based on posture sensor according to claim 1, characterized in that: The posture sensor chip (B1) communicates with the single-chip computer (B2) via an SPI bus.

4. The flexible finger posture detection glove based on posture sensor according to claim 1, characterized in that: The gloves (A) are rubber gloves.

5. The flexible finger posture detection glove based on posture sensor according to claim 1, characterized in that: The flexible circuit board (B) is attached to the back of the glove (A).

6. The flexible finger posture detection glove based on posture sensor according to claim 1, characterized in that: The attitude sensor chip (B1) converts the TTL level serial port signal of the microcontroller into an RS485 signal through a converter (B3).

7. A flexible finger posture detection method, characterized in that: Based on the flexible finger gesture detection glove as claimed in claim 1, the method comprises: S1. Calibration When used for the first time, the attitude sensor chip (B1) is calibrated, and by setting a static reference attitude, the initial attitude angle of each attitude sensor chip (B1) in the attitude is obtained; S2. Calculate the joint rotation axis: After the calibration is completed, the rotation axis of the joint is determined, and the direction of the rotation axis is determined by the relative posture of the two attitude sensor chips (B1); S3, real-time posture collection: When the finger moves, the posture sensor chip (B1) collects its angle posture data in real time; S4, joint rotation angle calculation: Use the real-time posture data collected by the two-section posture sensor chip (B1) to calculate the finger joint rotation angle, specifically: S41. Calculate Euler angles from quaternions Use the quaternion output by the attitude sensor chip (B1) to calculate the Euler angles (Pitch, Roll, Yaw) of the joints. Assume that the quaternion q = (w, x, y, z), then the Euler angles are calculated using the following formula: X-axis rotation: Roll = atan2(2(wx+yz), 1-2(x 2 +y 2 )) Y-axis rotation: Pitch = asin(2(wy-zx)) Z axis rotation: Yaw = atan2(2(wz+xy), 1-2(y 2 +z 2 )); S42, joint rotation angle calculation Calculate the relative rotation using quaternion multiplication: The initial quaternion q0 represents the posture of the sensor when the joint is not initially rotated; the current quaternion q t Indicates the posture of the sensor at the current time point; is the inverse of the quaternion, obtained by conjugation calculation: Using the Euler angle formula in step S41, the relative rotation quaternion q relative Converted to the rotation angle of the joint.

8. The flexible finger posture detection method according to claim 7, characterized in that: The specific process of step S1 is: obtaining the initial pitch angle, roll angle and heading angle of each posture sensor; setting the angle of the finger in a static and straight state as the reference zero point.

9. The flexible finger posture detection method according to claim 7, characterized in that: The step S3 also includes using a filter to process the collected angle attitude sensor to eliminate noise.

10. The flexible finger posture detection method according to claim 7, characterized in that: The method further comprises step S5, real-time updating and feedback: by solving the joint rotation angle, updating the finger posture data in real time, and transmitting the result to the host computer for posture judgment and feedback.

Images