Motion analysis system and method based on boot-mounted multi-sensor fusion

Abstract

The invention discloses a motion analysis system and method based on boot-mounted multi-sensor fusion, and relates to the technical field of motion analysis. Comprising a sensor module, the sensor module serves as a data acquisition end, and the sensor module comprises an inertial measurement unit which comprises a three-axis gyroscope and a three-axis accelerometer and is used for measuring the angular velocity and the acceleration of foot equipment in a three-dimensional space in real time; the distance sensing unit adopts an active detection technology, and adopts one or more of flight time laser ranging, ultrasonic ranging or millimeter wave radar to be matched; and the barometer is used for measuring the ambient atmospheric pressure, estimating the relative change of the height through the change of the atmospheric pressure, and providing auxiliary information for height estimation. The device is convenient to mount, is directly fixed to existing foot equipment based on modular design, and does not affect the sports performance; the posture precision is high, foot actions are directly measured, and the posture error is controlled within 3 degrees in combination with distance constraint.

Description

Technical Field

[0001] This invention relates to the field of motion analysis technology, and in particular to a motion analysis system and method based on shoe-mounted multi-sensor fusion. Background Technology

[0002] In the field of sports analytics, particularly in foot-based sports such as skiing, skating, and skateboarding, precise analysis of athletes' postures, trajectories, and interactions with the environment is crucial for skill improvement, training optimization, and sports safety. Existing technological solutions primarily focus on sensor deployment methods, but all suffer from the following shortcomings: Fixed sensor solutions involve pre-embedding or fixing sensors (such as inertial measurement units) inside sports equipment (such as snowboards and ice skates). While this allows for direct measurement of the equipment's motion, it requires structural modifications to the equipment, potentially affecting its mechanical performance and user experience. Furthermore, the fixed sensor placement makes it difficult to flexibly adapt to different athletes' usage habits or various sports scenarios.

[0003] Wearable sensor solutions place sensors on parts of an athlete's body, such as the calves, thighs, or waist. Their advantage is that no modification to sports equipment is required. However, their measurement data reflects the movement of limb segments, requiring modeling to infer foot movements. Errors are amplified during transmission, resulting in low posture measurement accuracy. Secondly, this solution cannot effectively reconstruct the terrain or accurately determine the contact state between the equipment and the ground (such as the edge angle in skiing). Furthermore, movement of other body parts can significantly interfere with sensor data, leading to poor data stability. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing a motion analysis system and method based on shoe-mounted multi-sensor fusion.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: A motion analysis system based on shoe-mounted multi-sensor fusion includes a sensor module, which serves as a data acquisition end. The sensor module includes: An inertial measurement unit, comprising a three-axis gyroscope and a three-axis accelerometer, is used to measure the angular velocity and acceleration of the foot equipment in three-dimensional space in real time; The distance sensing unit employs active detection technology, using one or more of time-of-flight laser ranging, ultrasonic ranging, or millimeter-wave radar in combination. A barometer is used to measure ambient atmospheric pressure and estimate the relative change in altitude by measuring changes in air pressure, thus providing auxiliary information for altitude estimation. The control unit, with a microcontroller at its core, coordinates the synchronous sampling of various sensors within the module, performs preliminary filtering or formatting of the raw data, and manages the module's power supply. A communication unit that transmits data wirelessly or via wired means; The data processing module receives real-time data streams from the sensor module. The data processing module is integrated within the module or performs data processing based on a remote cloud server. The data processing module runs data fusion algorithms and is responsible for completing tasks such as pose calculation, trajectory generation, terrain reconstruction, and motion analysis.

[0006] Preferably, the sensor module is located between the waist and the ankle.

[0007] Preferably, the pointing angle of the sensor module is as follows: Vertical direction angle The angle between the detection signal and the local horizontal plane normal is set between 1° and 89°. Horizontal direction angle The angle between the projection of the detection signal onto the horizontal plane and the forward or backward direction of the equipment; its range is set between 1° and 89°.

[0008] Preferably, the pointing angle of the sensor module is as follows: Vertical direction angle The angle between the detection signal and the local horizontal plane normal is set between 30° and 60°. Horizontal direction angle The angle between the projection of the detection signal onto the horizontal plane and the forward or backward direction of the equipment; its range is set between 10° and 45°.

[0009] Preferably, the sensor module specifically includes: The base contains a circuit board, on which a laser rangefinder sensor is mounted. The top cover is detachably mounted on the base via a quick-release button. A laser window adapted to the laser rangefinder sensor is provided on one side of the top cover, and a push-button switch is provided on the top of the top cover. The power supply unit is installed inside the base.

[0010] The motion analysis method based on shoe-mounted multi-sensor fusion includes the following steps: Step S1: Data Synchronization Acquisition The data processing unit synchronously triggers and receives IMU data from the left and right shoe sensor modules, slant range data from the distance sensing unit, and barometer data at a predetermined frequency. Step S2: Geometric correction of distance data based on the pre-calibrated vertical orientation angle of the distance sensing unit. and horizontal direction angle The original slant range L obtained in step S1 is converted into the required geometric parameters, and the slant range information is converted into a form that is easy to fuse with IMU data and constrain the pose. Step S3: Motion state recognition and analysis of the temporal variation characteristics of the dual-boot distance data after geometric correction to identify the current contact state between the motion equipment and the supporting surface; Step S4: Multi-sensor data fusion and pose calculation, transforming the state information identified in step S3 into strong constraints on the pose of the foot equipment, and tightly coupling and fusing it with IMU data; Step S5: Motion trajectory generation and terrain reconstruction; Step S6: Motion parameter analysis and output. Based on the high-precision data obtained in the previous steps, perform analysis.

[0011] Preferably, in step S2, the geometric parameters are calculated in the following way: Vertical height The vertical distance from the sensor to the projection of its detection point on the ground. ; Horizontal projection distance The distance of the detection point relative to the sensor's projection on the horizontal plane. ; in, This is the original slope distance. It is a vertical angle. It is a horizontal angle.

[0012] Preferably, in step S4, the multi-sensor data fusion and pose calculation are performed as follows: When in the flat position, the bottom contact points of both feet are constrained to be on the same height plane, and the angle between the equipment base plate and the local horizontal plane is less than a set value; when in the edge position, the geometric relationship between the edge of the boot and the contact point of the snow surface is constrained. Extended Kalman filtering, unscented Kalman filtering, or graph-based optimization methods are employed. The dynamic model of the IMU is used as the prediction step, while the geometric constraints provided by distance measurement are used as the observation update step. Through optimization, a smooth, continuous, and high-precision six-degree-of-freedom pose estimate is obtained, suppressing the drift error caused by IMU integration.

[0013] Preferably, in step S5, the motion trajectory generation and terrain reconstruction are as follows: The left and right boot pose sequences calculated in step S4 are combined with time information to draw the complete motion trajectory in a unified global coordinate system. By using the distance data and corresponding global coordinates continuously acquired by the two boots during the movement, these discrete three-dimensional spatial points are fused, interpolated, and surface reconstructed to build a three-dimensional terrain model of the area surrounding the movement path.

[0014] Preferably, in step S6, kinematic parameters are calculated, technical movements are analyzed, visualization output is provided, and an analysis report is generated.

[0015] The beneficial effects of this invention are as follows: 1. This invention is easy to install, based on a modular design, and can be directly fixed to existing foot equipment without affecting sports performance; it has high posture accuracy, directly measuring foot movements, and combined with distance constraints, the posture error is controlled within 3°.

[0016] 2. The trajectory of this invention is continuous and stable. The dual-boot inertial data and distance data are fused to maintain trajectory continuity in complex environments. The terrain reconstruction is rich. The tilt detection captures the terrain outside the equipment and generates a high-precision three-dimensional model.

[0017] 3. This invention has strong environmental adaptability, and its active detection is not affected by light or weather conditions; it is cost-effective, using consumer-grade sensors, so the overall cost is controllable; it provides comprehensive data dimensions, offering multi-dimensional outputs such as attitude, trajectory, terrain, and motion analysis. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the sensor module of the motion analysis system based on shoe-mounted multi-sensor fusion proposed in this invention; Figure 2 This is a schematic diagram of the sensor module split state of the motion analysis system based on shoe-mounted multi-sensor fusion proposed in this invention.

[0019] In the picture: 1-Base; 2-Quick release button; 3-Button switch; 4-Top cover; 5-Laser window; 6-Circuit board; 7-Laser rangefinder sensor; 8-Power supply unit. Detailed Implementation

[0020] The technical solution of the present invention will be further described in detail below with reference to specific embodiments.

[0021] Example 1: A motion analysis system based on shoe-mounted multi-sensor fusion, including a sensor module, which serves as a data acquisition end. The sensor module includes: An inertial measurement unit (IMU), which includes a three-axis gyroscope and a three-axis accelerometer, is used to measure the angular velocity and acceleration of the foot equipment in three-dimensional space in real time. The distance sensing unit employs active detection technologies, such as time-of-flight (ToF) laser ranging, ultrasonic ranging, or millimeter-wave radar. In this embodiment, ranging is achieved based on a laser ranging TOF sensor, and its installation orientation is not vertically downward, but rather set with a specific tilt angle. A barometer is used to measure ambient atmospheric pressure and estimate the relative change in altitude by measuring changes in air pressure, providing auxiliary information for altitude estimation, especially when there is significant vertical movement. The control unit, with a microcontroller at its core, coordinates the synchronous sampling of various sensors within the module, performs preliminary filtering or formatting of the raw data, and manages the module's power supply. A communication unit that transmits data wirelessly or via wired means; The data processing module receives real-time data streams from the sensor module. The data processing module is integrated within the module or performs data processing based on a remote cloud server. The data processing module runs the core data fusion algorithm and is responsible for completing the tasks of pose calculation, trajectory generation, terrain reconstruction, and motion analysis. The user display terminal is used to show the analysis results, such as motion trajectory diagrams, real-time posture animations, 3D terrain models, and various motion statistics (speed, distance, slope, etc.).

[0022] The sensor module is located between the waist and the ankle. In this embodiment, it is preferably located near the ankle or on the outside of the boot, and is securely attached to the user's left and right foot sports equipment, respectively. This position can most directly reflect the movement of the feet.

[0023] The pointing angle of the sensor module is as follows: Vertical direction angle The angle between the detection signal and the local horizontal plane normal (i.e., the vertical direction) is set between 1° and 89°, preferably between 30° and 60°; this tilt design avoids the limitation of only obtaining the height of a single point directly below when vertically detecting, so that the detection point is located in front of or behind the side of the equipment. Horizontal direction angle The angle between the projection of the detection signal onto the horizontal plane and the forward (or backward) direction of the equipment; its range is set between 1° and 89°, and it can be tilted forward or backward, preferably between 10° and 45°; this angle allows the detection signal to illuminate the outer support surface of the moving equipment.

[0024] By combining the above angles, the detection signal can illuminate the support surface (snow, ice, etc.) on the outside of the edge of the snowboard, ice skate, or skateboard, thus not only measuring distance, but also effectively capturing the terrain outline and the state of the equipment edge in contact with the ground (edge).

[0025] The sensor module specifically includes: Base 1, with a circuit board 6 fixed inside the base 1, and a laser rangefinder sensor 7 installed on the circuit board 6; Top cover 4 is detachably mounted on base 1 via quick release button 2. A laser window 5 adapted to laser range sensor 7 is provided on one side of top cover 4. A button switch 3 is provided on the top of top cover 4. Power supply unit 8 is installed inside base 1.

[0026] Example 2: A motion analysis method based on shoe-mounted multi-sensor fusion, comprising the following steps: Step S1: Data Synchronization Acquisition The data processing unit synchronously triggers and receives IMU data from the left and right shoe sensor modules, slant range data from the distance sensing unit, and barometer data at a predetermined frequency. Step S2: Geometric correction of distance data based on the pre-calibrated vertical orientation angle of the distance sensing unit. and horizontal direction angle The original slant range L obtained in step S1 is converted into the required geometric parameters, and the slant range information is transformed into a form that is easy to fuse with IMU data and constrain the pose, as follows: Vertical height The vertical distance from the sensor to the projection of its detection point on the ground.

[0027] Horizontal projection distance The distance of the detection point relative to the sensor's projection on the horizontal plane. (Assuming a horizontal angle here) (Measured relative to the direction of travel) in, This is the original slope distance. It is a vertical angle. It is a horizontal angle; Step S3: Motion state identification and analysis of the temporal variation characteristics of the geometrically corrected dual-boot distance data (H and / or S values) to identify the current contact state between the motion equipment and the supporting surface; for example: Flat position: The vertical height H values ​​measured by both boots are similar and relatively stable, indicating that both feet are flat on the support surface; Edge position: When turning, the distance reading H of the boot on one side (edge ​​side) may increase significantly due to the tilt of the boot (the detection point moves to the snow surface outside the edge of the ski), while the other side may not change much or decrease; the edge position and degree can be determined by threshold comparison or pattern recognition algorithm; Airborne state: When the distance readings of both boots increase sharply at the same time and exceed the threshold, it indicates that the athlete is in the airborne take-off state; Step S4: Multi-sensor data fusion and pose calculation. The state information identified in step S3 is transformed into strong constraints on the pose of the foot equipment, and then tightly coupled and fused with the IMU data; specifically as follows: State constraints: For example, in the flat position, constraints can be applied to assume that the contact points of the bottom of both feet are at the same height plane and that the equipment base plate is approximately parallel to the local horizontal plane; in the edge position, the geometric relationship between the edge of the boot and the contact point with the snow surface can be constrained. Fusion Algorithm: Employs Extended Kalman Filter (EKF), Unscented Kalman Filter (UKF), or graph-based optimization methods (such as SLAM algorithm); the algorithm model uses the IMU's dynamic model as the prediction step, while the geometric constraints provided by distance measurement (and possible GPS signals, magnetometer data, etc.) are used as the observation update step; through optimization, a smooth, continuous, and high-precision six-degree-of-freedom pose estimate (three-dimensional position and three-dimensional attitude) is obtained, effectively suppressing the drift error caused by IMU integration.

[0028] Step S5: Motion trajectory generation and terrain reconstruction; details are as follows: Trajectory generation: Combine the left and right boot pose sequences calculated in step S4 with time information to draw the complete motion trajectory in a unified global coordinate system; Terrain Reconstruction: Using distance data (corrected) and their corresponding global coordinate positions continuously acquired by the dual boots during movement, these discrete 3D spatial points (detection points) are fused, interpolated, and surface reconstructed (e.g., using point cloud processing or mesh generation algorithms) to construct a 3D terrain model of the area surrounding the movement path; the tilt detection of the dual boots provides a wider ground coverage, making the reconstructed terrain richer.

[0029] Step S6: Motion Parameter Analysis and Output Based on the high-precision data obtained in the preceding steps, perform high-level analysis: Calculate kinematic parameters: instantaneous velocity, average velocity, acceleration, deceleration, gliding distance, vertical drop, etc. Analyze technical movements: turning radius, turning angle, edge angle, airtime, landing impact force, etc. Visualization output: Displays motion trajectories, 3D terrain, posture change curves, key action markers, etc. in a graphical manner on the user terminal; Generate analysis reports: Provide comprehensive assessments of athletic performance.

[0030] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A motion analysis system based on shoe-borne multi-sensor fusion, characterized in that, The sensor module includes a sensor module as a data acquisition end, and the sensor module includes: An inertial measurement unit containing a three-axis gyroscope and a three-axis accelerometer for measuring the angular velocity and acceleration of the foot equipment in three-dimensional space in real time; A distance perception unit using active detection technology, using one or more of time-of-flight laser ranging, ultrasonic ranging, or millimeter wave radar; A barometer for measuring the ambient atmospheric pressure and estimating the relative change in height through changes in air pressure to provide auxiliary information for height estimation; A control unit with a microcontroller as the core to coordinate the synchronous sampling of various sensors in the module, perform preliminary filtering or formatting processing on the raw data, and manage the module power supply; A communication unit that sends data wirelessly or through a wired connection; A data processing module that receives real-time data streams from the sensor module, and the data processing module is integrated in the module or implemented based on a remote cloud server for data processing. The data processing module runs data fusion algorithms and is responsible for completing tasks such as pose solving, trajectory generation, terrain reconstruction, and motion analysis.

2. The motion analysis system based on shoe-borne multi-sensor fusion of claim 1, wherein, The sensor module is arranged at a position below the waist and between the ankles.

3. The motion analysis system based on shoe-borne multi-sensor fusion of claim 2, wherein, The pointing angle of the sensor module is as follows: vertical direction angle : the angle between the probe signal and the local horizontal plane normal, which is set in the range of 1° to 89°; horizontal direction angle : the angle between the projection of the detection signal on the horizontal plane and the direction of the equipment's advance or retreat; its range is set between 1° and 89°.

4. The motion analysis system based on shoe-borne multi-sensor fusion of claim 3, wherein, The pointing angle of the sensor module is as follows: vertical direction angle : the angle between the probe signal and the local horizontal plane normal, which is set to be between 30° and 60°; horizontal direction angle : the angle between the projection of the detection signal on the horizontal plane and the direction of the equipment's advance or retreat; it is set between 10° and 45°.

5. The motion analysis system based on on-board multi-sensor fusion of claim 1, wherein, The sensor module specifically includes: A base (1) with a circuit board (6) fixed inside, and a laser ranging sensor (7) arranged on the circuit board (6); A top cover (4) that is detachably mounted on the base (1) through a quick-release button (2), and a laser window (5) that is adapted to the laser ranging sensor (7) is formed on one side of the top cover (4), and a button switch (3) is arranged on the top of the top cover (4); A power supply unit (8) installed in the base (1).

6. A method for motion analysis based on shoe-borne multi-sensor fusion, characterized in that, The motion analysis system according to any one of claims 1-5 is implemented, including the following steps: Step S1: Data synchronization acquisition The data processing unit is triggered synchronously at a predetermined frequency and receives IMU data from left and right boot sensor modules, slant distance data from the distance perception unit, and barometer data; Step S2: Distance data geometric correction converts the original slant range L obtained in step S1 into the required geometric parameters, transforming the slant range information into a form that is easy to fuse with IMU data and forms constraints on the pose. and the horizontal direction angle , according to the vertical direction angle and the horizontal direction angle of the distance perception unit pre-calibrated. Step S3: Motion state recognition Analyze the time sequence change characteristics of the double-boot distance data after geometric correction to identify the contact state of the current motion equipment and the support surface; Step S4: Multi-sensor data fusion and pose solving Convert the state information identified in step S3 into a strong constraint condition for the foot equipment pose, and tightly couple and fuse with the IMU data; Step S5: Motion trajectory generation and terrain reconstruction; Step S6: Motion parameter analysis and output Based on the high-precision data obtained in the previous steps, analysis is performed.

7. The sports analysis method based on shoe-borne multi-sensor fusion according to claim 6, characterized in that, In the step S2, the geometric parameters are calculated as follows: vertical height : vertical distance of the sensor to its detection point's projection on the ground, ; horizontal projection distance : the projection distance of the detection point in the horizontal plane relative to the sensor, ; wherein, is the original slant range, is the vertical angle, is the horizontal angle. 8.The sports analysis method based on shoe-borne multi-sensor fusion according to claim 6, wherein, In the step S4, the multi-sensor data fusion and pose solving are as follows: In the flat state, the contact points of the two foot bottoms are constrained to be in the same height plane, and the angle between the equipment bottom plate and the local horizontal plane is less than a set value; in the standing state, the geometric relationship of the contact points of the standing side boot body edge and the snow surface is constrained; The extended Kalman filter, unscented Kalman filter or graph optimization based method is adopted; a dynamic model of the IMU is taken as a prediction step, and geometric constraints provided by distance measurement are taken as an observation update step; through optimization solution, a smooth, continuous and high-precision six-degree-of-freedom pose estimation is obtained, and drift error caused by IMU integration is inhibited. 9.The sports analysis method based on shoe-borne multi-sensor fusion according to claim 6, wherein, In the step S5, the motion trajectory generation and the terrain reconstruction are specifically as follows: The left and right shoe pose sequences solved in the step S4 are combined with time information to draw a complete motion trajectory in a unified global coordinate system; The distance data and the corresponding global coordinate positions continuously acquired by the double shoes during the motion are used to fuse, interpolate and reconstruct surfaces of these discrete three-dimensional space points, so that a three-dimensional terrain model of a region around the motion path is constructed. 10.The sports analysis method based on multi-sensor fusion on board according to claim 6, wherein, In the step S6, kinematic parameters are calculated, technical actions are analyzed, visual output is performed and an analysis report is generated.

Images