Personalized route generation for Mountain Biking cycling based on the user’s profile
Personalized route generation for Mountain Biking cycling based on the user’s profile. Advisor: Jugurta Lisboa Filho. The popularity of cycling has been on the rise both as a sustainable transport alter- native and as a leisure activity, through Mountain Biking (MTB). As its popularity increases, so...
| Autor: | |
|---|---|
| Tipo de recurso: | tesis de maestría |
| Estado: | Versión publicada |
| Fecha de publicación: | 2020 |
| País: | Brasil |
| Institución: | Universidade Federal de Viçosa (UFV) |
| Repositorio: | LOCUS Repositório Institucional da UFV |
| Idioma: | inglés |
| OAI Identifier: | oai:locus.ufv.br:123456789/31017 |
| Acceso en línea: | https://locus.ufv.br//handle/123456789/31017 |
| Access Level: | acceso abierto |
| Palabra clave: | Ciclismo Sistemas de informação geográfica Trilhas para mountain bike Ciência da Computação |
| Sumario: | Personalized route generation for Mountain Biking cycling based on the user’s profile. Advisor: Jugurta Lisboa Filho. The popularity of cycling has been on the rise both as a sustainable transport alter- native and as a leisure activity, through Mountain Biking (MTB). As its popularity increases, so does the need for tools to aid in the activity, such as route sharing tools. Most of these tools rely on Volunteered Geographic Information (VGI) both to acquire new trails for their databases and to rank them. While these tools are useful in most places, they can be of little to no help in regions with a smaller number of cyclists. This work proposes using data collected from multiple sources of VGI to automatically generate MTB routes based on user preferences, easing the decision making process of choosing new trails. Due to its vast availability, segments from Strava, a social network for athletes, were chosen as one of the data sources, the other being mapping data from the collaborative mapping tool OpenStreetMap(OSM), which was also used to select Points Of Interest (POI) relevant to the activity. An Integer Linear Programming model was developed to select sets of segments considering user preferences for terrain difficulty, elevation in the trail and total distance of the route, focusing on unpaved streets. The method developed in this work showed that it is possible, through the use of Strava segments, selected POIs and mapping data from OSM, to create pleasant circuits based on user preferences, with the main challenge being the quality of the collaborative data available in OSM. The circuits created with the method proved to be, in most part, pleasant, visiting relevant POIs and avoiding paved streets wherever possible. Keywords: Cycling. MTB. VGI. Route Generation |
|---|