Reverse Engineering the Dynamics of Human Pursuit Strategies
Creators:Arumukhom Revi, Dheepak
MetadataShow full item record
Publisher:The Ohio State University
Series/Report no.:The Ohio State University. Department of Mechanical and Aerospace Engineering Honors Theses; 2016
Many animals in the wild often perform pursuit and evasion in order to survive – pursuing prey or evading predators. In modern humans, such pursuit and evasion are most often seen in various sports and children’s games. Here, we examine what pursuit strategies people use to catch an evader, as in a game of tag. We performed human subject experiments in which we tested pursuit-evasion scenarios with two subjects per trial, one the pursuer and the other the evader, and eleven subject pairs. We used multiple protocols, with differing evader motion – evader moving in a straight line, a circle, or free-form, and the evader being able to change direction or not (when moving in a straight line or circle). Pursuer and evader motion data was obtained using stereogrammetry techniques with three cameras. We compared the data with a few mathematical model of pursuit, including pure pursuit, lead, and lag pursuit, proportional navigation. We also fit the pursuer velocity data to linear models of evader variables. While the median strategy used by the pursuers was close to pure pursuit, we found that pursuers used a lead strategy when the evader was more constrained and therefore more predictable – for instance, when the evader moved in a straight line without changing direction. We also found that on average 40% of the pursuers motion variance (velocity magnitude and direction) can be explained by using simple models, relating to evader motion (velocity magnitude and direction) and the distance between the pursuer and evader.
Academic Major: Mechanical Engineering
College of Engineering
Items in Knowledge Bank are protected by copyright, with all rights reserved, unless otherwise indicated.