Back to search

Cornell University

Qualitative Relational Mapping And Navigation For Long-Term Robotic Operation

Abstract

dc:description.abstract

The research presented in this work focuses on several aspects of the remote operation of ground vehicles, notably Navigation and Mapping for autonomous robots and the effects of time delay in tele-operated vehicles. Navigation and mapping of large, unstructured spaces is achieved by accumulating constraints on the geometrical relationships between landmarks. These relationships are tracked using two qualitative representations of space, one based on qualitative angles between landmark triples, and a second which also considers qualitative edge lengths. For the first representation, measurements and graph inference are performed by way of manually computed lookup tables based on geometrical constraints on qualitative states. For the second representation, measurements are generated online using a branch-and-bound algorithm to solve a set of nonlinear feasibility problems, while lookup tables for inference are generated using a similar, offline approach. Estimates of the Relative Neighborhood Graph are extracted from the qualitative map and used to perform long-distance navigation. The effects of human control of remote vehicles are considered, focusing on the question of how operators are able to compensate for time delays when teleoperating vehicles in continuous motion. Statistical models fit to experimental data using the Least Angle Regression and Sparse Multinomial Regression algorithms show that human operators anticipate future control needs by predicting rover motion forward through time to determine predicted off-track errors. The relative contributions of environmental features to model predictive power is used to determine how feature 'importance' varies as a function of time delay.

Degree

thesis:*
Name thesis:degree_name
Ph. D., Mechanical Engineering
Level thesis:degree_level
Doctor of Philosophy
Discipline thesis:degree_discipline
Mechanical Engineering
Grantor
Cornell University
Year dc:date.issued
2014

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • McClelland, Mark
Chair dc:contributor.chair
  • Campbell, Mark
Committee members dc:contributor.committeemember
  • Peck, Mason
  • Psiaki, Mark Lockwood
  • Kress Gazit, Hadas

Subjects

dc:subject × 3

Rights

Language dc:language.iso
en_US

Identifiers

dc:identifier.*
Dc Identifier Bibid
8641206

Chain of custody

source
Harvested from
Cornell University
Base URL
ecommons.cornell.edu/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

McClelland, Mark. Qualitative Relational Mapping And Navigation For Long-Term Robotic Operation. Doctor of Philosophy thesis, Cornell University, 2014. https://hdl.handle.net/1813/37108