CaltechTHESIS
  A Caltech Library Service

Visual Prediction of Rover Slip: Learning Algorithms and Field Experiments

Citation

Angelova, Anelia Nedelcheva (2008) Visual Prediction of Rover Slip: Learning Algorithms and Field Experiments. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/F7FY-5T13. https://resolver.caltech.edu/CaltechETD:etd-10032007-121619

Abstract

Perception of the surrounding environment is an essential tool for intelligent navigation in any autonomous vehicle. In the context of Mars exploration, there is a strong motivation to enhance the perception of the rovers beyond geometry-based obstacle avoidance, so as to be able to predict potential interactions with the terrain. In this thesis we propose to remotely predict the amount of slip, which reflects the mobility of the vehicle on future terrain. The method is based on learning from experience and uses visual information from stereo imagery as input. We test the algorithm on several robot platforms and in different terrains. We also demonstrate its usefulness in an integrated system, onboard a Mars prototype rover in the JPL Mars Yard.

Another desirable capability for an autonomous robot is to be able to learn about its interactions with the environment in a fully automatic fashion. We propose an algorithm which uses the robot's sensors as supervision for vision-based learning of different terrain types. This algorithm can work with noisy and ambiguous signals provided from onboard sensors. To be able to cope with rich, high-dimensional visual representations we propose a novel, nonlinear dimensionality reduction technique which exploits automatic supervision. The method is the first to consider supervised nonlinear dimensionality reduction in a probabilistic framework using supervision which can be noisy or ambiguous.

Finally, we consider the problem of learning to recognize different terrains, which addresses the time constraints of an onboard autonomous system. We propose a method which automatically learns a variable-length feature representation depending on the complexity of the classification task. The proposed approach achieves a good trade-off between decrease in computational time and recognition performance.

Item Type:Thesis (Dissertation (Ph.D.))
Subject Keywords:learning from automatic supervision; nonlinear dimensionality reduction; robot-terrain interaction; rover slip; self-supervised learning; terrain classification
Degree Grantor:California Institute of Technology
Division:Engineering and Applied Science
Major Option:Computer Science
Thesis Availability:Public (worldwide access)
Research Advisor(s):
  • Matthies, Larry H. (advisor)
  • Perona, Pietro (co-advisor)
Thesis Committee:
  • Perona, Pietro (chair)
  • Matthies, Larry H.
  • Lapusta, Nadia
  • Murray, Richard M.
  • Abu-Mostafa, Yaser S.
Defense Date:14 September 2007
Non-Caltech Author Email:anelia.angelova (AT) gmail.com
Record Number:CaltechETD:etd-10032007-121619
Persistent URL:https://resolver.caltech.edu/CaltechETD:etd-10032007-121619
DOI:10.7907/F7FY-5T13
ORCID:
AuthorORCID
Angelova, Anelia Nedelcheva0000-0003-1822-7943
Default Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:3886
Collection:CaltechTHESIS
Deposited By: Imported from ETD-db
Deposited On:25 Oct 2007
Last Modified:08 Nov 2023 00:44

Thesis Files

[img]
Preview
PDF (Angelova_PhD_Thesis.pdf) - Final Version
See Usage Policy.

8MB

Repository Staff Only: item control page