Cross-country skiing is a highly demanding competitive sport with a strong emphasis on both technique and technology. It is one of the few sports that requires a competitor to use several distinct techniques in one event, chosen based on factors such as gradient, speed and snow conditions. Traditionally, the technological aspect of the sport has revolved around the development of new materials and construction techniques for the skis and other equipment used. However, more recent advancements have come about through using technology to monitor physiological and technical parameters and using the data collected to suggest ways to improve the end result. Typically this monitoring is done indoors in a lab environment, using roller skis on a large treadmill that is about 3 metres square. A skier's technique is analysed using high-speed cameras positioned at various angles around the treadmill, and physiological parameters such as heart rate and oxygen uptake are recorded using sensors attached to static lab equipment. However, this indoor testing has its limitations. Firstly, roller skis on a rubber treadmill behave di erently to standard skis on snow, which a ects the technique of the skier by some degree. Secondly, environmental conditions are vastly di erent outdoors, and can change rapidly, with temperatures typically between -30 and +10 degrees Celsius, compared to a lab at room temperature. Other factors such as humidity and wind speed also become relevant when skiing outdoors. For these reasons, it is desirable for both coaches and skiers if performance data can be collected whilst skiing outdoors on a typical track, using all the standard equipment. Furthermore, the equipment used in the lab tends to be connected by cables to data loggers, meaning it takes a long time to set up the equipment ready for use. With this in mind, it was decided that a new approach to collecting data was needed, one that uses equipment that can be used both indoors and out, and does not require a lot of e ort to set up. The bulky, wired lab equipment would be replaced by a Body-Area Network (BAN) of small wireless sensors attached to the skier. Instead of high-speed cameras, accelerometers in either a smartphone or a sensor node would be used to capture the movement of the skier on three axes, and wired physiological sensors would be replaced by a wireless sensor belt worn by the skier on their chest. This project has been run in conjunction with the Swedish Institute of Computer Science (SICS), who were responsible for developing the software that analyses the skier's technique using accelerometer data. Our task was to develop a complete system, from the network of sensors used for data collection, to the presentation of the data to the user. With respect to the technique analysis part, our system would collect the accelerometer data from a sensor and pass it to the analysis software, which would then produce output that forms part of our presentation interface. This report will rst look at the technical aspects of cross-country skiing, with a focus on what data is important to the coaches and skiers, before moving on to describing the system we have created in detail, and nally looking at how the system performed when tested on elite skiers. Also included is a technical report which uses data collected from this testing to study how di erent skiers perform during a race. Finally the system will be evaluated and future advancements discussed.
© Copyright 2013 All rights reserved.