Ski mountaineering is a sport in which uphill walking depends on physical fitness, movement technique, and the equipment used. One important piece of equipment is ski poles, but their effect on temporal gait parameters in ski mountaineering remains relatively poorly studied. In this thesis, we developed an algorithm for the automatic analysis of temporal gait parameters in ski mountaineering using accelerometers mounted on the skis. The algorithm determines the beginnings and ends of contact times, which correspond to the loading and unloading of the ski, and uses this information to calculate contact time, gait cycle, and double support time. The performance of the algorithm was checked by comparing the contact times with manual readings from a high-speed camera. In addition, the ground reaction force was estimated from the signals of an accelerometer mounted on the pelvis and expressed as a multiple of body weight. The measurements were performed on a treadmill at a constant incline, with participants walking with primary, shorter, and longer pole lengths. Using the obtained data, the effect of pole length on the considered gait parameters was statistically evaluated with a linear mixed model. The results showed that pole length had a statistically significant effect on contact time and gait cycle. With shorter and longer poles, the predicted values of contact time and gait cycle were lower than with primary poles, with the effect being more pronounced for shorter poles. For double support time, the differences were statistically significant, but the effect of pole length was small, making its practical relevance questionable. For ground reaction force, a statistically significant difference was found between shorter and primary poles, while no statistically significant difference was found between longer and primary poles. The developed algorithm proved to be a useful tool for the automatic processing of a large number of steps and for further biomechanical analysis of gait in ski mountaineering.
|
