Using micro-sensor data to quantify macro kinematics of classical cross-country skiing during on-snow training

A Smart Ski Pole for Skiing Pattern Recognition and Quantification Application

In cross-country skiing, ski poles play a crucial role in technique, propulsion, and overall performance. The kinematic parameters of ski poles can provide valuable information about the skier's technique, which is of great significance for coaches and athletes seeking to improve their skiing performance. In this work, a new smart ski pole is proposed, which combines the uniaxial load cell and the inertial measurement unit (IMU), aiming to provide comprehensive data measurement functions more easily and to play an auxiliary role in training. The ski pole can collect data directly related to skiing technical actions, such as the skier's pole force, pole angle, inertia data, etc., and the system's design, based on wireless transmission, makes the system more convenient to provide comprehensive data acquisition functions, in order to achieve a more simple and efficient use experience. In this experiment, the characteristic data obtained from the ski poles during the Double Poling of three skiers were extracted and the sample t-test was conducted. The results showed that the three skiers had significant differences in pole force, pole angle, and pole time. Spearman correlation analysis was used to analyze the sports data of the people with good performance, and the results showed that the pole force and speed (r = 0.71) and pole support angle (r = 0.76) were significantly correlated. In addition, this study adopted the commonly used inertial sensor data for action recognition, combined with the load cell data as the input of the ski technical action recognition algorithm, and the recognition accuracy of five kinds of cross-country skiing technical actions (Diagonal Stride (DS), Double Poling (DP), Kick Double Poling (KDP), Two-stroke Glide (G2) and Five-stroke Glide (G5)) reached 99.5%, and the accuracy was significantly improved compared with similar recognition systems. Therefore, the equipment is expected to be a valuable training tool for coaches and athletes, helping them to better understand and improve their ski maneuver technique.

IMU Data-Driven and PCA-Based Approach to Establish Quantifiable and Practically Applicable Measures for V2 Technique Elements in Cross-Country Skiing

Quantifying movement coordination in cross-country (XC) skiing, specifically the technique with its elemental forms, is challenging. Particularly, this applies when trying to establish a bidirectional transfer between scientific theory and practical experts' knowledge as expressed, for example, in ski instruction curricula. The objective of this study was to translate 14 curricula-informed distinct elements of the V2 ski-skating technique (horizontal and vertical posture, lateral tilt, head position, upper body rotation, arm swing, shoulder abduction, elbow flexion, hand and leg distance, plantar flexion, ski set-down, leg push-off, and gliding phase) into plausible, valid and applicable measures to make the technique training process more quantifiable and scientifically grounded. Inertial measurement unit (IMU) data of 10 highly experienced XC skiers who demonstrated the technique elements by two extreme forms each (e.g., anterior versus posterior positioning for the horizontal posture) were recorded. Element-specific principal component analyses (PCAs)-driven by the variance produced by the technique extremes-resulted in movement components that express quantifiable measures of the underlying technique elements. Ten measures were found to be sensitive in distinguishing between the inputted extreme variations using statistical parametric mapping (SPM), whereas for four elements the SPM did not detect differences (lateral tilt, plantar flexion, ski set-down, and leg push-off). Applicability of the established technique measures was determined based on quantifying individual techniques through them. The study introduces a novel approach to quantitatively assess V2 ski-skating technique, which might help to enhance technique feedback and bridge the communication gap that often exists between practitioners and scientists.

The influence of race tactics for performance in the heats of an international sprint cross-country skiing competition

The purpose of this study was to examine the influence of race tactics for performance in the heats of an international sprint cross-country (XC) skiing competition in the classical style. Thirty elite male XC skiers (age: 24±3 years, sprint International Ski Federation [FIS] points: 61±27) performed a sprint time-trial (STT) followed by one to three 'knock-out' heats on a 1.7 km racecourse. An integrated GNSS/IMU system was used to determine position, sub-technique distribution and kinematics. Positioning was analysed using the television broadcast of the race. STT rank correlated positively with the final rank [(rs (28) = .72, P = .001)]. The top-two finishers in each heat were on average ~3.8% slower in the heats compared to the STT (237.1±3.9 vs. 228.3±4.0 seconds, P = .001). On average, the skiers performed ~10 overtakings per 100 meters from the start to the last uphill segment but only ~3 overtakings per 100 meters in the last two segments in each heat. 93.8% of the top-two finishing skiers positioned themselves at top 2 before approaching the final uphill, in which the top-two finishers and the skiers ranked 3-4 were generally faster than those ranked 5-6 in the heats (both, P = .01). Here, top-four skiers employed 5.3% longer cycle lengths and 3.4% higher cycle rates in the diagonal sub-technique than skiers ranked 5-6 (all, P = .01). The present study demonstrates the importance of race tactics for performance in the heats of sprint XC skiing, in which the main performance-determining factors in the present racecourse were a front position when approaching the final uphill segment combined with the ability to ski fast in that segment. In general, this illustrates how accurate racecourse analyses may help skiers to optimize their race-individual race-strategies in the heats of sprint XC skiing competitions.

Classification of Cross-Country Ski Skating Sub-Technique Can Be Automated Using Carrier-Phase Differential GNSS Measurements of the Head's Position

Position–time tracking of athletes during a race can provide useful information about tactics and performance. However, carrier-phase differential global navigation satellite system (dGNSS)-based tracking, which is accurate to about 5 cm, might also allow for the extraction of variables reflecting an athlete’s technique. Such variables include cycle length, cycle frequency, and choice of sub-technique. The aim of this study was to develop a dGNSS-based method for automated determination of sub-technique and cycle characteristics in cross-country ski skating. Sub-technique classification was achieved using a combination of hard decision rules and a neural network classifier (NNC) on position measurements from a head-mounted dGNSS antenna. The NNC was trained to classify the three main sub-techniques (G2–G4) using optical marker motion data of the head trajectory of six subjects during treadmill skiing. Hard decision rules, based on the head’s sideways and vertical movement, were used to identify phases of turning, tucked position and G5 (skating without poles). Cycle length and duration were derived from the components of the head velocity vector. The classifier’s performance was evaluated on two subjects during an in-field roller skiing test race by comparison with manual classification from video recordings. Classification accuracy was 92–97% for G2–G4, 32% for G5, 75% for turning, and 88% for tucked position. Cycle duration and cycle length had a root mean square (RMS) deviation of 2–3%, which was reduced to <1% when cycle duration and length were averaged over five cycles. In conclusion, accurate dGNSS measurements of the head’s trajectory during cross-country skiing contain sufficient information to classify the three main skating sub-techniques and characterize cycle length and duration.

Comparisons of Macro-Kinematic Strategies During the Rounds of a Cross-Country Skiing Sprint Competition in Classic Technique

This study was designed to examine macro-kinematic parameters of six female cross-country skiers during the qualifying, semi-final and final rounds of a 1.1 km sprint competition in classical technique. During each round these skiers were monitored continuously with a single micro-sensor, and their cycle parameters and relative use of these two sub-techniques calculated. Within each round six sections of the course, during which all skiers employed either double pole (DP) or diagonal stride (DS) sub-technique, were chosen for additional analysis. The mean macro-kinematic cycle parameters and relative usage of sub-techniques over the full course did not differ significantly between rounds. On average 54% of the course was covered employing DP and 13% using DS, while 32% was covered utilizing a non-cyclical or irregular technique. With DP, the mean racing speed and cycle rate (CR) on the starting, middle and finishing sections of the course differed significantly, with no differences in mean cycle length (CL) between the last two sections. At the finish, higher DP speed was achieved by increasing CR. On the three hills, where all athletes utilized DS, mean racing speed and CL, but not mean CR, differed significantly. On these sections DS speed was increased by utilizing longer cycles. The individual skiers utilized a variety of macro-kinematic strategies during different rounds and on different sections of the course, depending on individual strengths, preferences and pacing strategies, as well as the course topography and tactical interactions with other skiers. Such collection of macro-kinematic data during competitions can help to identify an individual skier's strengths and weaknesses, guiding the testing of different cycle rates, and lengths on different terrains during training in order to optimize performance.

Ski Skating Race Technique-Effect of Long Distance Cross-Country Ski Racing on Choice of Skating Technique in Moderate Uphill Terrain

The aim of this study was to investigate the effect of prolonged ski racing using skating style on technique choice in a transition section among female and male high-level skiers. Fifty three national-to-elite level skiers (20 females: 26.7 ± 4.8 years, 167.0 ± 6.5 m, 61.0 ± 5.1 kg, and 75.5 ± 68.8 FIS points; 33 males: 25.2 ± 3.5 years, 179.0 ± 5.2 cm, 73.1 ± 5.7 kg, and 73.7 ± 63.2 FIS points) were video recorded along a flat-to-uphill transition section of a course during the 30-km (females) and 50-km (males) races at the 2018 Norwegian National Championships. Across laps, section speeds decreased (P < 0.001) in all skiers, with the best-ranked skiers faster than the lowest-ranked (P < 0.001), and males faster than females in the first and middle laps. Section speed within each lap was associated with race performance (r = 0.76–0.86, P < 0.001 in females and r = 0.87–0.89, P < 0.001 in males). The prevalence of Gear 2 (G2) increased, while Gear 3 (G3) use decreased (both P < 0.001) across the subsequent laps, with females preferring G2 more than males in lap one (P = 0.027). In long-distance skate-style skiing, transition performance is representative of race performance and skiers decrease the use of the often-faster G3 technique while increasing the use of the slower G2 technique due to prolonged exercise. Especially female skiers should consider adding some flat-to-uphill G3 practice into established training, specifically early in the session before fatigue may occur.

Sex-based differences in sub-technique selection during an international classical cross-country skiing competition

The purpose of this study was to compare speed, sub-technique selection and temporal patterns between world-class male and female cross-country (XC) skiers and to examine the combined associations of sex and speed on sub-technique selection. Thirty-three XC skiers performed an international 10-km (women; n = 8) and 15-km (men; n = 25) time-trial competition in the classical style (with the first 10 km of the race being used for analyses). An integrated GNSS/IMU system was used to continuously track position speed and automatically classify sub-techniques and temporal patterns (i.e. cycle length and–rate). When comparing the eight highest ranked men and women, men spent less time than women (29±2 vs. 45±5% of total time) using diagonal stride (DIA), more time (44±4 vs. 31±4%) using double poling (DP) and more time (23±2 vs. 19±3%) using tucking and turning (all P < .01). Here, men and women used these sub-techniques at similar temporal patterns within the same speed-intervals; although men employed all sub-techniques at steeper uphill gradients (all P < .05). In subsequent analyses including all 33 skiers, adjustment for average racing speed did not fully attenuate the observed sex differences in the proportion of time using DIA (CI_95% [-10.7, -1.6]) and DP (CI_95% [0.8, 9.3]). Male world-class XC skiers utilized less DIA and more DP compared to women of equal performance levels. Although these differences coincided with men’s higher speed and their ability to use the various sub-techniques at steeper uphill gradients, sexual dimorphism in the proportional use of DIA and DP also occurred independently of these speed-differences.

Pole Length's Influence on Performance During Classic-Style Snow Skiing in Well-Trained Cross-Country Skiers

PURPOSE

To investigate how self-selected pole length (PL) of ∼84% (PL84%) compared with ∼90% (PL90%) of body height influenced performance during a 700-m time trial with undulating terrain on snow.

METHODS

Twenty-one cross-country skiers, 7 of whom were women, performed 4 trials at a maximal effort in a counterbalanced fashion with PL84% and PL90% separated by 20-minute breaks between trials. In trials I and II, only double poling was allowed, while in trials III and IV, skiers used self-selected classical subtechniques. Continuous speed, cyclic parameters, and heart rate were collected using microsensors in addition to a post-time-trial rating of perceived exertion (RPE).

RESULTS

The 700-m times with only double poling were significantly shorter with PL90% than PL84% (mean ± 95% confidence limits -1.6% ± 1.0%). Segment analyses showed higher speed with PL90% in uphill sections than with PL84% (3.7% ± 2.1%), with the greatest difference found for the female skiers (5.6% ± 2.9%). In contrast, on flat terrain at high skiing speeds, speed was reduced with PL90% compared with PL84% (-1.5% ± 1.4%); this was only significant for the male skiers. During free choice of classical subtechniques, PL did not influence performance in any segments, choice of subtechnique, or cycle rate during the trials. No differences in rating of perceived exertion or heart rate between PLs were found.

CONCLUSIONS

PL90% improved performance in uphills at low speeds when using double poling but hindered performance on flat terrain and at higher speeds compared with self-selected PLs. Choice of PL should, therefore, be based on racecourse topography, preferred subtechniques, and the skier's physiological and technical abilities.

Embedded inertial measurement unit reveals pole lean angle for cross-country skiing

This study introduces an inertial measurement unit-based measurement system for resolving the dynamic lean angle of a ski pole during double poling while cross-country skiing. The measurement system estimates both the pole lean angle and pole–terrain contact events. Reported are results from 20 trials providing validated estimates of ski pole lean angle and the timing of pole plant and pole lift events. The pole lean angle is estimated from a complementary filter that fuses estimates of orientation from the embedded accelerometer and angular rate gyro. Validation follows from comparison with video capture measurements. Bland–Altman analysis showed agreement between the two measurement modalities with less than 5% bias in the mean differences (relative to the lean angle range of motion). Companion correlation analysis confirms strong correlation ($$r = 0.99$$ r = 0.99 ) between the inertial measurement unit and video-estimated lean angles and with mean root-mean-square errors below 4.5$$^{\circ }$$ ∘ .

Cross-Country Skiing Analysis and Ski Technique Detection by High-Precision Kinematic Global Navigation Satellite System

Cross-country skiing (XCS) embraces a broad variety of techniques applied like a gear system according to external conditions, slope topography, and skier-related factors. The continuous detection of applied skiing techniques and cycle characteristics by application of unobtrusive sensor technology can provide useful information to enhance the quality of training and competition. (1) Background: We evaluated the possibility of using a high-precision kinematic global navigation satellite system (GNSS) to detect cross-country skiing classical style technique. (2) Methods: A world-class male XC skier was analyzed during a classical style 5.3-km time trial recorded with a high-precision kinematic GNSS attached to the skier’s head. A video camera was mounted on the lumbar region of the skier to detect the type and number of cycles of each technique used during the entire time trial. Based on the GNSS trajectory, distinct patterns of head displacement (up-down head motion) for each classical technique (e.g., diagonal stride (DIA), double poling (DP), kick double poling (KDP), herringbone (HB), and downhill) were defined. The applied skiing technique, skiing duration, skiing distance, skiing speed, and cycle time within a technique and the number of cycles were visually analyzed using both the GNSS signal and the video data by independent persons. Distinct patterns for each technique were counted by two methods: Head displacement with course inclination and without course inclination (net up-down head motion). (3) Results: Within the time trial, 49.6% (6 min, 46 s) was DP, 18.7% (2 min, 33 s) DIA, 6.1% (50 s) KDP, 3.3% (27 s) HB, and 22.3% (3 min, 03 s) downhill with respect to total skiing time (13 min, 09 s). The %Match for both methods 1 and 2 (net head motion) was high: 99.2% and 102.4%, respectively, for DP; 101.7% and 95.9%, respectively, for DIA; 89.4% and 100.0%, respectively, for KDP; 86.0% and 96.5%, respectively, in HB; and 98.6% and 99.6%, respectively, in total. (4) Conclusions: Based on the results of our study, it is suggested that a high-precision kinematic GNSS can be applied for precise detection of the type of technique, and the number of cycles used, duration, skiing speed, skiing distance, and cycle time for each technique, during a classical style XCS race.

Development of a Framework for the Investigation of Speed, Power, and Kinematic Patterns in Para Cross-Country Sit-Skiing: A Case Study of an LW12 Athlete

Objective: To develop a framework for the investigation of speed, power, and kinematic patterns across varying terrain in cross-country (XC) sit-skiing, and to test this framework in a XC sit-skier of the LW12 class during high- (HIT) and low-intensity (LIT) endurance training.

Methods: One XC sit-skiing athlete of the LW12 class with a single above-the-knee amputation was equipped with a GNSS enabled sports watch with integrated barometry and heart rate monitoring (peak heart rate: 195 beats·min⁻¹), and an inertial measurement unit. After a warm-up, he performed two 20-m maximal speed tests on a flat and an uphill section to determine maximal speed and power, followed by skiing 5.75 km at both LIT and HIT in varying terrain.

Results: 51, 28, and 21% of the time during HIT and 53, 28, and 19% of the time during LIT were spent in uphill, flat and downhill terrain, respectively. Maximal speed in the uphill and flat section was 4.0 and 6.2 m·s⁻¹, respectively, and the corresponding maximal power output 342 and 252 W. The % of maximal speed did not differ between the uphill and the flat section (HIT: 66 vs. 67%, LIT: 47 vs. 50%), whereas the % of maximal power output was lower in the uphill than flat section (HIT: 65 and 80%, LIT: 46 and 58%). Still, the absolute power output was slightly higher in the uphill than the flat section (HIT: 222 vs. 201 W, LIT: 156 vs. 145 W). Furthermore, cycle rate was significantly higher during HIT than LIT (60–61 vs. 45–55 cycles·min⁻¹, across all terrains, all p < 0.03), while cycle length was longer in the uphill terrain (3.0 vs. 2.6 m, p < 0.001). Furthermore, the % of peak heart rate was significantly higher in HIT than LIT (90 vs. 78, 85 vs. 67, and 88 vs. 66%, respectively, in the uphill, flat and downhill terrain, all p < 0.001).

Conclusions: Here, we present a new integrative framework for future investigations of performance, technical and physical demands in XC sit-skiing. In this case study, the increase in speed from LIT to HIT was due to increases in cycle rate in all terrains, while cycle length was less affected. Although the absolute power output was slightly higher in the uphill compared to the flat section both for HIT and LIT, the athlete worked closer to his maximum power output in the flat section.

Assessment of Basic Motions and Technique Identification in Classical Cross-Country Skiing

Cross-country skiing is a popular Olympic winter sport, which is also used extensively as a recreational activity. While cross-country skiing primarily is regarded as a demanding endurance activity it is also technically challenging, as it contains two main styles (classical and skating) and many sub-techniques within these styles. To further understand the physiological demands and technical challenges of cross-country skiing it is imperative to identify sub-techniques and basic motion features during training and competitions. Therefore, this paper presents features for identification and assessment of the basic motion patterns used during classical-style cross-country skiing. The main motivation for this work is to contribute to the development of a more detailed platform for comparing and communicating results from technique analysis methods, to prevent unambiguous definitions and to allow more precise discussions and quality assessments of an athlete's technical ability. To achieve this, our paper proposes formal motion components and classical style technique definitions as well as sub-technique classifiers. This structure is general and can be used directly for other cyclic activities with clearly defined and distinguishable sub-techniques, such as the skating style in cross country skiing. The motion component features suggested in our approach are arm synchronization, leg kick, leg kick direction, leg kick rotation, foot/ski orientation and energy like measures of the arm, and leg motion. By direct measurement, estimation, and the combination of these components, the traditional sub-techniques of diagonal stride, double poling, double poling kick, herringbone, as well as turning techniques can be identified. By assuming that the proposed definitions of the classical XC skiing sub-techniques are accepted, the presented classifier is proven to map measures from the motion component definitions to a unique representation of the sub-techniques. This formalization and structure may be used on new motion components, measurement principles, and classifiers, and therefore provides a framework for comparing different methodologies. Pilot data from a group of high-level cross-country skiers employing inertial measurement sensors placed on the athlete's arms and skis are used to demonstrate the approach. The results show how detailed sub-technique information can be coupled with physical, track, and environmental data to analyze the effects of specific motion patterns, to develop useful debriefing tools for coaches and athletes in training and competition settings, and to explore new research hypotheses.

Sex-based differences in speed, sub-technique selection, and kinematic patterns during low- and high-intensity training for classical cross-country skiing

OBJECTIVES

We investigated sex-based differences in speed, sub-technique selection, and kinematic patterns during low- (LIT) and high-intensity training (HIT) for classical cross-country (XC) skiing across varying terrain.

METHODS

Six male and six female elite XC skiers with an approximately 15% differences in VO2max (men: 68.9±2.9 mL·min-1·kg-1, women: 60.1±3.3 mL·min-1·kg-1) were monitored using a multi-sensor system to collect time-synchronised data of heart rate, speed, and multiple tri-axial inertial measurements units while XC skiing on a 5-km competition track.

RESULTS

Men skied 21% faster than women during HIT (5.9±0.3 m·s-1 vs. 4.9±0.2 m·s-1, P < .001), with the greatest difference (26%) while skiing on flat terrain, whereas skiing speed did not significantly differ between men and women during LIT. At similar instructed intensity and rating of perceived effort, women exhibited significantly higher relative heart rate (85±2% vs. 71±3% of maximum) and blood lactate levels (4.0±1.3 vs. 1.2±0.2 mmol/L) during LIT (all P < .001) than men, whereas physiological responses did generally not differ between the sexes during HIT. During both intensities and among both sexes, double poling (DP) was the sub-technique most used relative to distance, followed by miscellaneous sub-techniques (MISC), diagonal stride (DIA), kick double poling (DK) and herringbone (HRB). In relation to distance women used DIA more than men during LIT (22% vs. 17%, P = .009) and HIT (23% vs. 12%, P = .001), whereas men used MISC, including tucking and turning, more than women during LIT (39% vs. 25%, P = .017) and HIT (41% vs. 30%, P = .064). In particular, men used DP more than women while skiing the uphill sections during both LIT (24% vs. 11%, P = .015) and HIT (39% vs. 13%, P = .002).

CONCLUSIONS

Our findings provide novel insights into sex-based differences in speed, sub-technique selection, and kinematic patterns during LIT and HIT for classical skiing.

A Unified Deep-Learning Model for Classifying the Cross-Country Skiing Techniques Using Wearable Gyroscope Sensors

The automatic classification of cross-country (XC) skiing techniques using data from wearable sensors has the potential to provide insights for optimizing the performance of professional skiers. In this paper, we propose a unified deep learning model for classifying eight techniques used in classical and skating styles XC-skiing and optimize this model for the number of gyroscope sensors by analyzing the results for five different configurations of sensors. We collected data of four professional skiers on outdoor flat and natural courses. The model is first trained over the flat course data of two skiers and tested over the flat and natural course data of a third skier in a leave-one-out fashion, resulting in a mean accuracy of ~80% over three combinations. Secondly, the model is trained over the flat course data of three skiers and tested over flat course and natural course data of one new skier, resulting in a mean accuracy of 87.2% and 95.1% respectively, using the optimal sensor configuration (five gyroscope sensors: both hands, both feet, and the pelvis). High classification accuracy obtained using both approaches indicates that this deep learning model has the potential to be deployed for real-time classification of skiing techniques by professional skiers and coaches.

Macro-Kinematic Differences Between Sprint and Distance Cross-Country Skiing Competitions Using the Classical Technique

We compare the macro-kinematics of six elite female cross-country skiers competing in 1.1-km Sprint and 10.5-km Distance classical technique events on consecutive days under similar weather and track conditions. The relative use of double pole (DP), kick-double pole (KDP), diagonal stride (DS), tucking (Tuck) and turning (Turn) sub-techniques, plus each technique’s respective velocities, cycle lengths and cycle rates were monitored using a single micro-sensor unit worn by each skier during the Sprint qualification, semi-final and finals, and multiple laps of the Distance race. Over a 1.0-km section of track common to both Sprint and Distance events, the mean race velocity, cyclical sub-technique velocities, and cycle rates were higher during the Sprint race, while Tuck and Turn velocities were similar. Velocities with KDP and DS on the common terrain were higher in the Sprint (KDP +12%, DS +23%) due to faster cycle rates (KDP +8%, DS +11%) and longer cycle lengths (KDP +5%, DS +10%), while the DP velocity was higher (+8%) with faster cycle rate (+16%) despite a shorter cycle length (-9%). During the Sprint the percentage of total distance covered using DP was greater (+15%), with less use of Tuck (-19%). Across all events and rounds, DP was the most used sub-technique in terms of distance, followed by Tuck, DS, Turn and KDP. KDP was employed relatively little, and during the Sprint by only half the participants. Tuck was the fastest sub-technique followed by Turn, DP, KDP, and DS. These findings reveal differences in the macro-kinematic characteristics and strategies utilized during Sprint and Distance events, confirm the use of higher cycle rates in the Sprint, and increase our understanding of the performance demands of cross-country skiing competition.

Automatic Classification of Sub-Techniques in Classical Cross-Country Skiing Using a Machine Learning Algorithm on Micro-Sensor Data

The automatic classification of sub-techniques in classical cross-country skiing provides unique possibilities for analyzing the biomechanical aspects of outdoor skiing. This is currently possible due to the miniaturization and flexibility of wearable inertial measurement units (IMUs) that allow researchers to bring the laboratory to the field. In this study, we aimed to optimize the accuracy of the automatic classification of classical cross-country skiing sub-techniques by using two IMUs attached to the skier's arm and chest together with a machine learning algorithm. The novelty of our approach is the reliable detection of individual cycles using a gyroscope on the skier's arm, while a neural network machine learning algorithm robustly classifies each cycle to a sub-technique using sensor data from an accelerometer on the chest. In this study, 24 datasets from 10 different participants were separated into the categories training-, validation- and test-data. Overall, we achieved a classification accuracy of 93.9% on the test-data. Furthermore, we illustrate how an accurate classification of sub-techniques can be combined with data from standard sports equipment including position, altitude, speed and heart rate measuring systems. Combining this information has the potential to provide novel insight into physiological and biomechanical aspects valuable to coaches, athletes and researchers.

Full course macro-kinematic analysis of a 10 km classical cross-country skiing competition

In this study micro-sensors were employed to analyse macro-kinematic parameters during a classical cross-country skiing competition (10 km, 2-lap). Data were collected from eight male participants during the Australian championship competition wearing a single micro-sensor unit (MinimaxX™, S4) positioned on their upper back. Algorithms and visual classification were used to identify skiing sub-techniques and calculate velocities, cycle lengths (CL) and cycle rates (CR) over the entire course. Double poling (DP) was the predominant cyclical sub-technique utilised (43 ± 5% of total distance), followed by diagonal stride (DS, 16 ± 4%) and kick double poling (KDP, 5 ± 4%), with the non-propulsive Tuck technique accounting for 24 ± 4% of the course. Large within-athlete variances in CL and CR occurred, particularly for DS (CV% = 25 ± 2% and CV% = 15 ± 2%, respectively). For all sub-techniques the mean CR on both laps and for the slower and faster skiers were similar, while there was a trend for the mean velocities in all sub-techniques by the faster athletes to be higher. Overall velocity and mean DP-CL were significantly higher on Lap 1, with no significant change in KDP-CL or DS-CL between laps. Distinct individual velocity thresholds for transitions between sub-techniques were observed. Clearly, valuable insights into cross-country skiing performance can be gained through continuous macro-kinematic monitoring during competition.