A method for individualizing optimization of blade curvature of an ice skate based on human motion posture

Abstract

This invention provides a method for personalized optimization of ice skate blade curvature based on human motion posture. The method uses a multimodal sensing system to collect data on joint angles, center of gravity trajectory, blade pressure distribution, and posture angles of athletes during standard skating movements, extracting key motion feature parameters. A machine learning mapping model is constructed based on a historical athlete database to quantitatively correlate individual characteristics with blade curvature and motion performance. A multi-objective optimization algorithm recommends the optimal curvature value, and high-precision customized machining is achieved using a CNC grinding machine. Field testing and iterative feedback ensure the optimization effect. This invention realizes the transformation of ice skate blade curvature from standardization to precise personalization, significantly improving skating efficiency, motion control, and sports safety. It is applicable to various sports such as figure skating, speed skating, and ice hockey, and has good application prospects and industrialization value.

Description

Technical Field

[0001] This invention relates to the field of ice skate blade curvature optimization technology, and more specifically, to a personalized optimization method for ice skate blade curvature based on human movement posture. Background Technology

[0002] In ice sports such as figure skating, speed skating, short track speed skating, and ice hockey, the ice skate serves as the sole mechanical interface between the athlete and the ice surface. Its structural parameters directly affect skating efficiency, turning stability, takeoff control, and landing cushioning. Among these parameters, the blade curvature (the radius of curvature of the blade's bottom profile, usually expressed in feet, such as 7', 8', 9', etc.) is one of the core geometric parameters determining the skate's handling characteristics. A smaller curvature (e.g., 6'–7') provides greater flexibility and a smaller turning radius, suitable for sports requiring frequent changes of direction and complex footwork (such as figure skating); while a larger curvature (e.g., 8.5'–10') enhances straight-line stability and speed maintenance, and is more commonly used in speed sports.

[0003] However, current ice skates on the market generally use limited standardized curvature specifications, leaving users with only a few fixed options that are difficult to match with the significantly different biomechanical characteristics of individuals. Athletes exhibit high heterogeneity in height, weight, lower limb strength distribution, joint range of motion, center of gravity control habits, and technical movement patterns. A uniform blade curvature often leads to athletes adapting to the blade rather than the blade adapting to the athlete. This not only limits athletic performance but may also increase the risk of sports injuries such as ankle sprains and knee strain due to biomechanical mismatch.

[0004] While some high-end ice skates support manual grinding and adjustment of the blade curvature, this process is highly dependent on the technician's experience, lacks objective data support, and results in poor repeatability and low efficiency. In recent years, although some research has attempted to introduce sports biomechanical analysis to assist in the design of equipment, existing solutions mostly remain at the laboratory simulation or offline evaluation stage, and have not yet formed a closed-loop optimization system integrating real-time posture perception, data-driven modeling, intelligent recommendation, and precision manufacturing. Therefore, we propose an improvement, namely a personalized optimization method for ice skate blade curvature based on human movement posture. Summary of the Invention

[0005] The purpose of this invention is to address the problems raised in the existing background technology. To achieve the above-mentioned objective, this invention provides the following technical solution: a method for personalized optimization of ice skate blade curvature based on human motion posture, comprising the following steps: S1: During the subject's performance of standard skating movements, joint angle data, body center of gravity trajectory data, ice skate-ice surface contact force distribution data, and skating posture angle data are collected through a multimodal sensing system at a sampling frequency of not less than 100Hz; S2: Filter and denoise the collected data and perform time synchronization processing, and extract at least 5 motion feature parameters related to gliding performance; S3: Construct a mapping model based on a historical database containing no less than 100 athletes, wherein the database records the corresponding athletic performance scores, biomechanical indicators and subjective feedback under different blade curvature configurations; S4: Input the motion feature parameters of the current subject into the mapping model, calculate and output the recommended personalized blade curvature value through a multi-objective optimization algorithm. The unit is feet, with a range of 6.0–10.0 feet and an accuracy of 0.1 feet; S5: According to the above Value-controlled CNC grinding machines perform customized grinding of ice skate blades, ensuring the actual formed curvature matches... The deviation is no more than ±0.05 feet, and the optimization effect is verified through field gliding tests.

[0006] As a preferred technical solution of the present invention, the multimodal sensing system includes: Wearable inertial measurement unit (IMU) with angle measurement accuracy better than ±1°, installed on ankle, knee, and hip joints; The optical motion capture system has a spatial positioning accuracy better than ±2 mm and a frame rate of ≥120 fps. An embedded flexible pressure sensor array is used in the blade, with no fewer than 8 sensing points distributed along the length of the blade, and a pressure resolution of ≤0.5 N / cm. 2 ; High-speed camera equipment is used to assist in attitude verification.

[0007] As a preferred embodiment of the present invention, the motion characteristic parameters include: (a) Maximum tilt angle of the body in the coronal plane during the turning phase; (b) Ankle dorsiflexion angle at the moment of takeoff; (c) The proportion of pressure in the first 1 / 3 of the blade during braking to the total pressure; (d) Standard deviation of acceleration in the vertical direction of the center of mass during gliding; (e) Radius of curvature of a single-step glide trajectory; The parameters are all calculated based on the average value of three or more consecutive standard movements.

[0008] As a preferred technical solution of the present invention, the mapping model is a machine learning regression model, with an input dimension of 5–15-dimensional motion feature vectors and an output of a four-dimensional performance vector [E,T,J,C], representing gliding efficiency E (0–1), turning stability T (0–1), jump control J (0–1), and overall comfort C (out of 1–5 points), respectively. The model training employs cross-validation.2 ≥0.85.

[0009] As a preferred technical solution of the present invention, the multi-objective optimization algorithm in step S4 adopts NSGA-II, with a population size of 50–100, an iteration number ≥100, and the optimization objective function is: Among them, the weighting coefficient satisfy It is automatically set according to the user's preferred mode selected on the mobile terminal, and the preferred mode includes high stability, high flexibility or balanced.

[0010] As a preferred technical solution of the present invention, the blade curvature value The corresponding blade bottom profile is a circular arc segment or a composite curvature curve. When a circular arc segment is used, its theoretical radius of curvature R satisfies... After processing, the measured curvature was verified by a laser profile scanner, and the curvature deviation was ≤±1.5 mm over the entire length.

[0011] As a preferred technical solution of the present invention, the CNC grinding machine mentioned in step S5 is a five-axis linkage precision grinding equipment, the cutting edge grinding path is synthesized by B-spline curve fitting, the surface roughness Ra≤0.4μm, and the cutting edge straightness error≤0.1 mm / m.

[0012] As a preferred technical solution of the present invention, it also includes an iterative optimization mechanism: if any performance index in the field test fails to reach the preset threshold—including gliding efficiency E<0.80, turning stability T<0.75 or athlete subjective score C<4.0, then a second data collection is triggered, and the mapping model is updated based on incremental learning, and the corrected R_opt' is re-output, with a correction step size not exceeding ±0.3 feet.

[0013] As a preferred technical solution of the present invention, the method is deployed on a cloud-based intelligent service platform, which supports uploading encrypted motion data packets via a mobile application. The platform returns recommended radian values ​​and a visual analysis report within ≤5 seconds, and automatically synthesizes CNC machining G-code instructions conforming to the ISO 13485 standard, which are then sent to certified cooperative manufacturing terminals.

[0014] As a preferred technical solution of the present invention, it is applicable to any of the following sports: figure skating, speed skating, ice hockey or short track speed skating, and different standard motion libraries and feature weight templates are preset for different sports.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention accurately characterizes the movement characteristics of athletes by collecting multi-dimensional biomechanical data from their actual skating movements, breaking away from the traditional selection method that relies on experience or trial and error, and enabling the curvature of the ice skate to truly match the individual's physiological structure and technical style.

[0016] The optimized blade curvature of this invention can significantly improve gliding smoothness, turning responsiveness, and jump landing stability. In actual tests, it can improve the quality and success rate of action completion, while reducing joint load and risk of injury caused by equipment incompatibility.

[0017] This invention establishes a mapping model based on a large-scale athlete database, transforming subjective feelings into quantifiable performance indicators (such as gliding efficiency, stability scores, etc.), making the arc recommendation scientifically sound and repeatable.

[0018] As athletes age, improve their skills, or recover from injuries, their movement posture may change. This invention supports periodic reassessment and fine-tuning of the arc, enabling the synchronous evolution of equipment and human.

[0019] This invention, by pre-setting motion libraries for different projects and optimizing weight templates, can flexibly adapt to different scenarios such as figure skating, speed skating, and ice hockey, and has good universality and scalability.

[0020] This invention deeply integrates biomechanical analysis, artificial intelligence algorithms, and CNC precision machining, connecting the entire chain of measurement, analysis, design, manufacturing, and verification, and providing a replicable technical paradigm for the personalized customization of high-end sports equipment.

[0021] This invention integrates cloud platforms with mobile terminals, enabling grassroots coaches and amateur enthusiasts to conveniently access professional-grade optimization services, thereby promoting the scientific popularization of ice and snow sports. Attached image description: Figure 1 A flowchart of the method provided by the present invention; Figure 2 This is a data flowchart of the multi-objective optimization algorithm provided by the present invention; Figure 3 This is a block diagram of the multimodal sensing data provided by the present invention; Figure 4 A block diagram of motion feature parameters provided by the present invention. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are specific implementations of the present invention and are not limited to all embodiments.

[0023] Therefore, the following detailed description of embodiments of the present invention is not intended to limit the scope of the claimed invention, but merely illustrates some embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0024] It should be noted that, in the absence of conflict, the embodiments and features and technical solutions in the embodiments of the present invention can be combined with each other. It should be noted that similar reference numerals and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0025] Example: A method for personalized optimization of ice skate blade curvature based on human motion posture, comprising the following steps: S1: During the subject's performance of standard skating movements, joint angle data, body center of gravity trajectory data, ice skate-ice surface contact force distribution data, and skating posture angle data are collected by a multimodal sensing system at a sampling frequency of not less than 100 Hz; S2: Filter and denoise the collected data and perform time synchronization processing, and extract at least 5 motion feature parameters related to gliding performance; S3: Construct a mapping model based on a historical database containing no fewer than 100 athletes. The database records the corresponding athletic performance scores, biomechanical indicators, and subjective feedback under different blade curvature configurations. S4: Input the motion characteristic parameters of the current subject into the mapping model, calculate and output the recommended personalized blade curvature value through a multi-objective optimization algorithm. The unit is feet, with a range of 6.0–10.0 feet and an accuracy of 0.1 feet; S5: According to Value-controlled CNC grinding machines perform customized grinding of ice skate blades, ensuring the actual formed curvature matches... The deviation is no more than ±0.05 feet, and the optimization effect is verified through field gliding tests.

[0026] The multimodal sensing system includes a wearable inertial measurement unit (IMU) with an angle measurement accuracy better than ±1°, which is installed on the ankle, knee, and hip joints; The optical motion capture system has a spatial positioning accuracy better than ±2 mm and a frame rate of ≥120 fps. An embedded flexible pressure sensor array is used in the blade, with no fewer than 8 sensing points distributed along the length of the blade, and a pressure resolution of ≤0.5 N / cm. 2 ; High-speed camera equipment is used to assist in attitude verification.

[0027] Motion characteristic parameters include: (a) Maximum tilt angle of the body in the coronal plane during the turning phase; (b) Ankle dorsiflexion angle at the moment of takeoff; (c) The proportion of pressure in the first 1 / 3 of the blade during braking to the total pressure; (d) Standard deviation of acceleration in the vertical direction of the center of mass during gliding; (e) Radius of curvature of a single-step glide trajectory; All parameters are calculated based on the average value of three or more consecutive standard movements.

[0028] The mapping model is a machine learning regression model. The input dimension is a 5–15 dimensional motion feature vector, and the output is a four-dimensional performance vector [E, T, J, C], representing gliding efficiency E (0–1), cornering stability T (0–1), jump control J (0–1), and overall comfort C (out of 1–5 points), respectively. Model training uses cross-validation. 2 ≥0.85.

[0029] In step S4, the multi-objective optimization algorithm used is NSGA-II, with a population size of 50–100, an iteration number ≥100, and the optimization objective function is: Among them, the weighting coefficient satisfy It is automatically set according to the user's preferred mode selected on the mobile terminal. The preferred modes include high stability, high flexibility, or balanced.

[0030] Blade curvature value The corresponding blade bottom profile is a circular arc segment or a composite curvature curve. When a circular arc segment is used, its theoretical radius of curvature R satisfies... After processing, the measured curvature was verified by a laser profile scanner, and the curvature deviation was ≤±1.5 mm over the entire length.

[0031] In step S5, the CNC grinding machine is a five-axis linkage precision grinding device. The tool grinding path is synthesized by fitting a B-spline curve. The surface roughness Ra≤0.4μm and the cutting edge straightness error≤0.1 mm / m.

[0032] It also includes an iterative optimization mechanism: if any performance index fails to reach the preset threshold during the field test—including gliding efficiency E<0.80, turning stability T<0.75, or athlete subjective score C<4.0—a second data collection is triggered, and the mapping model is updated based on incremental learning, and the corrected R_opt' is re-output, with a correction step size not exceeding ±0.3 feet.

[0033] The method is deployed on a cloud-based intelligent service platform, which supports uploading encrypted motion data packets via a mobile application. The platform returns recommended radian values ​​and a visual analysis report within ≤5 seconds, and automatically synthesizes CNC machining G-code instructions that conform to the ISO 13485 standard, and distributes them to certified partner manufacturing terminals.

[0034] It is applicable to any of the following sports: figure skating, speed skating, ice hockey, or short track speed skating. Different standard motion libraries and feature weight templates are preset for different sports.

[0035] Experimental Example: Personalized Blade Curve Optimization for Figure Skators 1. Basic information of the subjects Gender / Age / Specialty: Female, 18 years old, competitive figure skater Height / Weight: 165 cm / 52 kg Current ice skate model: Jackson Ultima Legacy UB50 Factory default blade curvature: 7.0 feet Key technical requirements: high-difficulty jumps (such as Lutz and Flip), rapid footwork transitions, and stable rotations. 2. Data Acquisition (Step S1) In a standard indoor ice rink (temperature -5℃, ice surface flatness ≤1 mm / m) 2 In this study, participants performed each of the following standard actions 5 times: Continuous large-arc gliding with the outer edge (used to extract gliding efficiency and trajectory curvature) One-legged sharp turn (radius approximately 3 m, used to determine body tilt angle) Simulate Lutz takeoff (including entry into the blade and the moment of takeoff). T-stop braking action The multimodal sensing system is configured as follows:

[0036] All data are aligned with a unified timestamp, and the sampling frequency is uniformly resampled to 100 Hz.

[0037] 3. Feature extraction (step S2) The following five motion characteristic parameters were calculated from the preprocessed data (the average of five valid movements was taken):

[0038] The above parameters constitute a 5-dimensional input vector: x=[32.4,18.7,68.2,0.12,4.1] 4. Mapping Model Inference (Steps S3–S4) Historical database: Contains data on 127 figure skaters, short track speed skaters, and ice hockey players, with each player's four-dimensional performance rating recorded at ≥3 different radians (6.0–10.0 ft).

[0039] Mapping model: An XGBoost regression model is used, with input dimension = 5 and output = [E, T, J, C]. After 5-fold cross-validation, R0 is obtained. 2 =0.89.

[0040] User preference settings: Select high flexibility mode via mobile app → Automatically set weights: (w_1=0.2) (Efficiency), (w_2=0.2) (Stability), (w_3=0.4) (Jump Control), (w_4=0.2) (Comfort) Multi-objective optimization: Using the NSGA-II algorithm (population size = 80, 120 generations), solve for the objective function: ( ) Optimization results: Recommended personalized blade curvature value: Ropt = 7.6 feet (0.1 feet precision) 5. Customized grinding and validation (step S5) CNC machining: Equipment: DMG MORI 5-axis precision grinding machine Blade profile: The B-spline grinding path is synthesized based on Ropt = 7.6 ft (i.e., theoretical radius of curvature R = 7.6 × 304.8 ≈ 2316.5 mm). Machining parameters: Surface roughness Ra = 0.35 μm, cutting edge straightness error = 0.08 mm / m Radius verification: The finished blade was scanned along its entire length using a Keyence LJ-V7080 laser profile scanner. The measured average radius of curvature was 2315.2 mm, with a deviation of -1.3 mm (within the allowable range of ±1.5 mm). The corresponding measured arc value was 7.595 feet, meeting the accuracy requirement of ±0.05 feet.

[0041] 6. On-site gliding test and effect evaluation Subjects wore the original sensor system and underwent comparative testing under the same conditions (original 7.0 ft vs. new 7.6 ft):

[0042] Note: Subject feedback: The new arc provides more natural ankle support during takeoff and makes it less likely to "bite" and lose control when turning.

[0043] 7. Conclusion This experimental example successfully applied the described method to recommend and achieve a personalized 7.6-foot blade arc for figure skaters. Field tests showed that, while maintaining skating efficiency, it significantly improved jump control and turning stability, validating the effectiveness and engineering feasibility of the method.

[0044] Note: If the initial test does not meet expectations (e.g., the jump success rate increases by <20%), an iterative optimization process will be triggered—data will be re-collected, feature parameters will be updated, and Ropt will be fine-tuned (e.g., adjusted to 7.7 ft) until the preset performance thresholds are met (e.g., J≥0.85, T≥0.8).

[0045] The above embodiments are only used to illustrate the present invention and are not intended to limit the technical solutions described herein. Although the present invention has been described in detail with reference to the above embodiments, the present invention is not limited to the specific embodiments described above. Therefore, any modifications or equivalent substitutions to the present invention, as well as all technical solutions and improvements that do not depart from the spirit and scope of the invention, are covered within the scope of the claims of the present invention.

Claims

1. A method for personalized optimization of ice skate blade curvature based on human movement posture, characterized in that, Includes the following steps: S1: During the subject's performance of standard skating movements, joint angle data, body center of gravity trajectory data, ice blade-ice surface contact force distribution data, and skating posture angle data were collected using a multimodal sensing system at a sampling frequency of no less than 100 Hz. S2: Filter and denoise the collected data and perform time synchronization processing, and extract at least 5 motion feature parameters related to gliding performance; S3: Construct a mapping model based on a historical database containing no less than 100 athletes, wherein the database records the corresponding athletic performance scores, biomechanical indicators and subjective feedback under different blade curvature configurations; S4: Input the motion feature parameters of the current subject into the mapping model, calculate and output the recommended personalized blade curvature value through a multi-objective optimization algorithm. The unit is feet, with a range of 6.0–10.0 feet and an accuracy of 0.1 feet; S5: According to the above Value-controlled CNC grinding machines perform customized grinding of ice skate blades, ensuring the actual formed curvature matches... The deviation is no more than ±0.05 feet, and the optimization effect is verified through field gliding tests.

2. The method for personalized optimization of ice skate blade curvature based on human motion posture according to claim 1, characterized in that, The multimodal sensing system includes: Wearable inertial measurement unit (IMU) with angle measurement accuracy better than ±1°, installed on ankle, knee, and hip joints; The optical motion capture system has a spatial positioning accuracy better than ±2 mm and a frame rate of ≥120 fps. An embedded flexible pressure sensor array is used in the blade, with no fewer than 8 sensing points distributed along the length of the blade, and a pressure resolution of ≤0.5 N / cm. 2 ; High-speed camera equipment is used to assist in attitude verification.

3. The method for personalized optimization of ice skate blade curvature based on human motion posture according to claim 1, characterized in that, The motion characteristic parameters include: (a) Maximum tilt angle of the body in the coronal plane during the turning phase; (b) Ankle dorsiflexion angle at the moment of takeoff; (c) The proportion of pressure in the first 1 / 3 of the blade during braking to the total pressure; (d) Standard deviation of acceleration in the vertical direction of the center of mass during gliding; (e) Radius of curvature of a single-step glide trajectory; The parameters are all calculated based on the average value of three or more consecutive standard movements.

4. The method for personalized optimization of ice skate blade curvature based on human motion posture according to claim 1, characterized in that, The mapping model is a machine learning regression model. The input dimension is a 5–15 dimensional motion feature vector, and the output is a four-dimensional performance vector [E, T, J, C], representing gliding efficiency E (0–1), cornering stability T (0–1), jump control J (0–1), and overall comfort C (out of 1–5 points), respectively. Model training uses cross-validation. 2 ≥0.

85.

5. The method for personalized optimization of ice skate blade curvature based on human motion posture according to claim 1, characterized in that, The multi-objective optimization algorithm described in step S4 uses NSGA-II, with a population size of 50–100, an iteration count ≥100, and the optimization objective function is: Among them, the weighting coefficient satisfy It is automatically set according to the user's preferred mode selected on the mobile terminal, and the preferred mode includes high stability, high flexibility or balanced.

6. The method for personalized optimization of ice skate blade curvature based on human motion posture according to claim 1, characterized in that, The blade curvature value The corresponding blade bottom profile is a circular arc segment or a composite curvature curve. When a circular arc segment is used, its theoretical radius of curvature R satisfies... After processing, the measured curvature was verified by a laser profile scanner, and the curvature deviation was ≤±1.5 mm over the entire length.

7. The method for personalized optimization of ice skate blade curvature based on human motion posture according to claim 1, characterized in that, The CNC grinding machine mentioned in step S5 is a five-axis linkage precision grinding equipment. The tool grinding path is synthesized by fitting a B-spline curve. The surface roughness Ra≤0.4μm and the cutting edge straightness error≤0.1 mm / m.

8. The method for personalized optimization of ice skate blade curvature based on human motion posture according to claim 1, characterized in that, It also includes the S6 iterative optimization mechanism: if any performance index fails to reach the preset threshold during the field test—including gliding efficiency E<0.80, turning stability T<0.75 or athlete subjective score C<4.0—a second data collection is triggered, and the mapping model is updated based on incremental learning, and the corrected R_opt' is re-output, with a correction step size not exceeding ±0.3 feet.

9. The method for personalized optimization of ice skate blade curvature based on human motion posture according to claim 1, characterized in that, The method is deployed on a cloud-based intelligent service platform, which supports uploading encrypted motion data packets via a mobile application. The platform returns recommended radian values ​​and a visual analysis report within ≤5 seconds, and automatically synthesizes CNC machining G-code instructions that conform to the ISO 13485 standard, and distributes them to certified cooperative manufacturing terminals.

10. A method for personalized optimization of ice skate blade curvature based on human motion posture according to any one of claims 1 to 9, characterized in that, It is applicable to any of the following sports: figure skating, speed skating, ice hockey, or short track speed skating, and different standard motion libraries and feature weight templates are preset for different sports.

Images