Date of Award

8-2023

Document Type

Masters Thesis

Degree Name

M.S.

Organizational Unit

Daniel Felix Ritchie School of Engineering and Computer Science, Computer Science

First Advisor

Christopher Reardon

Second Advisor

Daniel Baack

Third Advisor

Maggie Wigness

Fourth Advisor

Matthew Rutherford

Keywords

Adversary aware navigation, Autonomous robot navigation, Context aware navigation, Preference based learning, Robot learning, Terrain aware navigation

Abstract

In autonomous robot navigation, the robot is able to understand the environment around it for intelligent navigation. From its world model of this environment, it generates a global plan for navigation from a position to a goal based on different factors. This research aims to implement autonomous robot navigation by learning terrain affordances: traversability (moving quickly) and concealment (staying hidden from an adversary) using the Preference-based Inverse Reward Learning (PbIRL) methodology. The PbIRL methodology reduces the barrier of generating initial demonstration data to learn the terrain affordances by using a human expert’s preferences to learn individual weights over the terrain types in the environment. These weights are then combined into a costmap, which is passed to a planner for path generation and navigation to a goal.

This thesis extends prior research in learning terrain costs for robot navigation using an active rewards learning Python package, APReL. The novel contribution of this thesis is that the robot not only considers traversability affordances of different terrains for navigation, but also considers concealment from an adversary in the environment. This research work also augments the APReL package with wrappers suitable for the use-case. The model will be evaluated by comparing its performance with an already existing Inverse Optimal Control (IOC) model trained from human demonstrations as a baseline learning algorithm for autonomous robot navigation.

Copyright Date

8-2023

Copyright Statement / License for Reuse

All Rights Reserved
All Rights Reserved.

Publication Statement

Copyright is held by the author. User is responsible for all copyright compliance.

Rights Holder

Aniekan Ufot Inyang

Provenance

Received from ProQuest

File Format

application/pdf

Language

English (eng)

Extent

86 pgs

File Size

7.3 MB

Discipline

Robotics, Artificial intelligence

Download



Share

COinS