Swimming

Objectives

The performance of swimming is strongly related to the swimmer’s technique. Therefore, a swimmer who wants to improve his swim performance has to devote a substantial proportion of the training to his swim style improvement. The two most important physical principles to swim faster and more efficient are:

• Reduction of the resistance through an improved body balance and body rotation.

• Increasing of the propelling force by improving the arm strokes and hence the stroke efficiency.

General consensus nowadays is that both factors must be optimized. The result of an improved streamlined position together with an increased propelling force is synergetic swimming, meaning that both factors have to be improved likewise for fast swimming

Many recreational swimmers and triathletes often train alone and do not have a trainer aside at all times, neigther theiy have access to a video analysis system. They are non-professional athletes and do not have the needed experience and sense of self-perception.

Therefor we envision a system worn by the swimmer which will help these athletes to keep track of their technique also in the training hours where no trainer is present. In addition this system will support the trainer during his training hours by providing additional continuous swim performance parameters not available so far.

Goal of this project is to show how a wide range of swim parameters can be monitored by a wearable swim assistant, which we call SwimMaster. Perfromance parameters of interest are for example the time per lane, the swimming velocity and the number of strokes per lane. On the other hand, swim style specific factors of interest are for example the body balance, the body rotation and the arm stroke trajectory. Finally online feedback modalities will be tested and evaluated.

Achievements

• We were able to demonstrate the feasibility of an automated swim efficiency evaluation with one wrist sensor. To capture the swim efficiency we detect the wall-push-off (lane start), the single arm strokes and the wall-strike (lane end). With these paramters we are able to calculate the average swimming velocity, the time per arm stroke and the swimming distance per arm stroke [Bächlin et al.; An automatic parameter extraction method for the 7x50m Stroke Efficiency Test; ICPCA 2008].

• We evaluated audio, visual and haptic user feedback interfaces to give feedback to a swimmer while swimming. None of the systems restrict the users in their swim movements. The visual and the haptic interfaces were recognized very well by the swimmer with a detection accuracy of 70%-100% but less than 70% of the audio feedback was recognized by the swimmer [Förster et al.; Non-interrupting user interfaces for electronic body-worn swim devices; PETRA 2009].

• We showed how the body balance and the body rotation can be extracted from a sensors at the back. These are two important swim style specific factors which can be used to evaluate the swim technique of the swimmer [Bächlin et al.; SwimMaster: A wearable assistant for swimmer; Ubicomp 2009].

• We build a SwimModel describing acceleration data of body worn sensors in crawl swimming. The model inputs are the arm and leg forces as well as the body torques. Based on the input forces and torques the model calculates the swimmer’s acceleration, velocity and position as well as the angular accelerations, angular velocities and angles. Internally, the model considers the buoyancy force and the drag forces. The transformation of the body motion into the sensor coordinate system and its sensing modalities is done by a sensor model. We validated our SwimModel by comparing the simulated sensor data with real sensor data recorded during swimming [Bächlin and Tröster; Pervasive computing in swimming: A model describing acceleration data of body worn sensors in crawl swimming; ICPCA 2009].

Student Projects

• Swim style evaluation from body-worn sensors; Urban Suppiger (WS2008)
• Uninterrupting user interfaces for body-worn swim devices; Andrea Moroni Stampa (WS2008)
• Automatic parameter extraction methods for athlete swimming; Dominik Breu, Jürg Germann (SS2008)

Publications

2009

• Non-interrupting user interfaces for electronic body-worn swim devices
  Kilian Förster, Marc Bächlin and Gerhard Tröster
  in: Proceedings of the 2nd International Conference on PErvasive Technologies Related to Assistive Environments (PETRA 2009), 2009
  [ BiBTeX RIS PDF ]
• Pervasive computing in swimming: A model describing acceleration data of body worn sensors in crawl swimming
  Marc Bächlin and Gerhard Tröster
  in: Proceedings of the 4th International Conference on Pervasive Computing and Applications 2009, 2009
  [ BiBTeX RIS URL ]
• SwimMaster: A wearable assistant for swimmer
  Marc Bächlin, Kilian Förster and Gerhard Tröster
  in: Ubicomp '09: Proceedings of the 11th international conference on Ubiquitous computing, Orlando, Florida, USA, pages 215-224, ACM, 2009
  [ BiBTeX RIS PDF DOI ]

2008

• An automatic parameter extraction method for the 7x50m Stroke Efficiency Test
  Marc Bächlin, Kilian Förster, Johannes Schumm, Dominik Breu, Jürg Germann and Gerhard Tröster
  in: Proceedings of 3rd International Conference on Pervasive Computing and Applications 2008, Alexandria, Egypt, pages 442-447, IEEE Conference Publications Management Group, 2008
  [ BiBTeX RIS URL ]