Attitude Estimation of Skis in Ski Jumping Using Low-Cost Inertial Measurement Units

Fang, Xiang; Göttlicher, Christoph; Holzapfel, Florian

doi:10.3390/proceedings2060317

Cited by 6 publications

(5 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The dataset covers a wide range of jump lengths and athletes, which leads to good prediction accuracy over jump lengths and generalization over athletes. Even though the dataset has the largest number of different ski jumping venues and athletes used in a tracking study [ 21 , 31 , 33 , 40 , 41 , 42 , 43 , 44 ], the number of venues is relatively small and unevenly distributed compared to the number of athletes. This could be improved in the future.…”

Section: Discussionmentioning

confidence: 99%

xLength: Predicting Expected Ski Jump Length Shortly after Take-Off Using Deep Learning

Link

Schwinn²,

Pulsmeyer³

et al. 2022

Sensors

View full text Add to dashboard Cite

With tracking systems becoming more widespread in sports research and regular training and competitions, more data are available for sports analytics and performance prediction. We analyzed 2523 ski jumps from 205 athletes on five venues. For every jump, the dataset includes the 3D trajectory, 3D velocity, skis’ orientation, and metadata such as wind, starting gate, and ski jumping hill data. Using this dataset, we aimed to predict the expected jump length (xLength) inspired by the expected goals metric in soccer (xG). We evaluate the performance of a fully connected neural network, a convolutional neural network (CNN), a long short-term memory (LSTM), and a ResNet architecture to estimate the xLength. For the prediction of the jump length one second after take-off, we achieve a mean absolute error (MAE) of 5.3 m for the generalization to new athletes and an MAE of 5.9 m for the generalization to new ski jumping hills using ResNet architectures. Additionally, we investigated the influence of the input time after the take-off on the predictions’ accuracy. As expected, the MAE becomes smaller with longer inputs. Due to the real-time transmission of the sensor’s data, xLength can be updated during the flight phase and used in live TV broadcasting. xLength could also be used as an analysis tool for experts to quantify the quality of the take-off and flight phases.

show abstract

Section: Discussionmentioning

confidence: 99%

xLength: Predicting Expected Ski Jump Length Shortly after Take-Off Using Deep Learning

Link

Schwinn²,

Pulsmeyer³

et al. 2022

Sensors

View full text Add to dashboard Cite

show abstract

“…Regarding the set-up, one of the main advantages was that it was not necessary to perform a calibration of the inertial sensors before doing the data collection. In fact, thanks to the algorithm proposed by Fang and colleagues [30], during the postprocessing, the raw data of the inertial sensors were reconstructed based on the design of the in-run of the ski jumping hill. The advantage of the post-initialization is very important, making the set-up easy to use, in case athletes and coaches would be interested in using the system on their own as feedback during training.…”

Section: Discussionmentioning

confidence: 99%

“…The raw IMU measurement data were postprocessed by using the algorithm presented in [30]. Firstly, the outcomes were low-pass filtered with cut-off frequency of 5 Hz.…”

Section: Methodsmentioning

confidence: 99%

Ski Position during the Flight and Landing Preparation Phases in Ski Jumping Detected with Inertial Sensors

Bessone

Petrat

Schwirtz

2019

Sensors

View full text Add to dashboard Cite

Ski movement plays an important role during landing preparation, as well as in the whole ski jumping performance. Good landing preparation timing and correct ski position increase the jump length and reduce the impact forces. Inertial motion units (IMUs) placed on the skis could constitute a promising technology for analyzing the ski movements during training. During regular summer trainings, 10 elite athletes (17 ± 1 years) performed jumps while wearing IMUs and wireless force insoles. This set-up enabled the analysis of a possible correlation between ski movements and ground reaction force (GRF) during landing impact. The results showed that the pitch during the landing preparation is the most influential movement on the impact kinetic variables since it is related to the angle of attack, which affects the aerodynamics. The ski position at 0.16 s before landing did not influence the kinetics because the athlete was too close to the ground. During the impact, the roll angle did not correlate with GRF. Moreover, each athlete showed a different movement pattern during the flight phase. Concluding, the combination of IMUs and force insoles is a promising set-up to analyze ski jumping performance thanks to the fast placement, low weight, and high reliability.

show abstract

“…On the contrary, the IMUs on the skis can be a cheap solution and easily placed and fixed behind the bindings by not professionals, after being correctly instructed. Finally, they can be used with a post-calibration [11], while the ones placed on the lower body need a pre-calibration before each trial, with a possible introduction of errors and time loss [5].…”

Section: Measuring Ski Jump Landing Performancementioning

confidence: 99%

Landing in Ski Jumping: A Review About its Biomechanics and the Connected Injuries

Bessone

Schwirtz

2021

J. of SCI. IN SPORT AND EXERCISE

View full text Add to dashboard Cite

The present review deals with the current scientific knowledge related with ski jump landing. A specific focus is given on the landing biomechanics, the methods utilized for its analysis and the injuries connected to the landing phase. Despite the demonstrated importance for the safety and the performance of ski jumpers, the landing and its preparation are rarely investigated. In this paper, after having firstly described the execution of landing and its preparation and the reason why is important to analyze it, an overview of the current status of the research related to the landing biomechanics is reported (kinetics, kinematics, electromyographic activation, aerodynamics, computer simulation). The third part describes the methods and technologies utilized in literature to analyze the landing and its preparation (video cameras, inertial sensors, force insoles, wind tunnel and computer simulation). After that, an overview of the injuries related to landing is reported. The final section proposes future research in the field of biomechanics of ski jump landing in different fields, such as computer simulations, kinematic analysis, equipment development and biomechanics of female athletes.

show abstract

Attitude Estimation of Skis in Ski Jumping Using Low-Cost Inertial Measurement Units

Cited by 6 publications

References 6 publications

xLength: Predicting Expected Ski Jump Length Shortly after Take-Off Using Deep Learning

xLength: Predicting Expected Ski Jump Length Shortly after Take-Off Using Deep Learning

Ski Position during the Flight and Landing Preparation Phases in Ski Jumping Detected with Inertial Sensors

Landing in Ski Jumping: A Review About its Biomechanics and the Connected Injuries

Contact Info

Product

Resources

About