| ||Format||Pages||Price|| |
|PDF (344K)||20||$25||  ADD TO CART|
|Complete Source PDF (4.1M)||198||$151||  ADD TO CART|
Advances in binding design for alpine skiing have been unable to eliminate injuries to the lower leg and knee. Studies to identify appropriate release settings have concluded that the maximum forces during skiing are not predictable using any measurable anthropometric parameters, and furthermore, that they are variable during skiing due to changes in snow conditions, skiing style, and other unmeasurable variables. This paper considers the feasibility of an alpine sking binding with variable release settings that adapt to changes in the force magnitudes during skiing. Release settings are determined according to difference equations involving previous and current force measurements. To facilitate comparison with contemporary commercial bindings, the learning binding design includes three release modes: external twist and internal twist governed by the lateral force component at the toe and forward lean governed by the vertical force component at the heel. Binding forces measured during more than 36 min of skiing experiments are used to determine values for the design parameters within the adaptive algorithm. Then the performance of the learning binding is judged by comparing the release settings to the settings recommended by ASTM Standard F 939-93.
The release settings of the learning binding algorithms are on average more than 22% less than the ASTM recommended settings for each skier, with a maximum reduction of 50% for one subject. No inadvertent releases are predicted throughout the skiing data. The reduced release settings adapt to temporal changes in forces and yield earlier releases in several falling scenarios experienced in the data. These results suggest that a learning binding can significantly improve the safety of alpine skiing.
Graduate student, University of California, Berkeley, CA
President, University of Maryland, College Park, MD