Automated Route Generation for Bicycle Tours of the United States

Abstract

Planning a multi-day bicycle tour is a time intensive and difficult task. To enable cyclists to construct their own multi-day tour routes, we propose an automated system which, given an origin, destination, and points of interest, generates detailed bicycle tour routes of minimal perceived exertion. This dissertation is comprised of three bodies of work contained in chapters 2, 3, and 4, respectively. In the first work, chapter 2, we build and test a model of the perceived exertion of different categories of cyclists on a daily path within a long bicycle tour. We first propose an additive formula for calculating the perceived exertion of cyclists on component parts of a tour and then present the results of a survey designed to verify the accuracy of the model. In the second work, chapter 3, we describe an algorithmic procedure for transforming a traditional road graph into a topographical graph. To that end we use polyline data from the OpenStreetMap (OSM), elevation data from the U.S. Geological Survey, and a model of perceived exertion for bicyclists of different levels of expertise. Our topographical graph allows for the calculation of the perceived exertion over any sequence of road segments (a path) for bicyclists of different levels of expertise and serves as input for generating paths of minimal perceived exertion. In the final work, chapter 4, we describe a procedure for constructing bicycle routes of minimal perceived exertion for cyclists of different levels of expertise over a multi-day tour. Given a cyclists origin, destination, selected points of interest, and a level of cycling expertise, this procedure generates a multi-day bicycle tour as a collection of successive daily paths that begin and end at overnight accommodations. We demonstrate the implementation of this procedure on an example multi-day tour route in California and present the results of a survey designed to evaluate the daily paths constructed. In summation, this dissertation contributes a new metric of perceived exertion for bicycle riders, a new topographical graph, and a procedure for constructing multi-day bicycle tour routes of minimal perceived exertion for bicyclists of different levels of expertise.