Page 1 |
Save page Remove page | Previous | 1 of 160 | Next |
|
small (250x250 max)
medium (500x500 max)
large ( > 500x500)
Full Resolution
All (PDF)
|
This page
All
Subset |
MOBILE ROBOT OBSTACLE AVOIDANCE USING
A COMPUTATIONAL MODEL OF THE LOCUST BRAIN
by
Manu Viswanathan
A Thesis Presented to the
FACULTY OF THE USC VITERBI SCHOOL OF ENGINEERING
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(COMPUTER SCIENCE)
May 2011
Copyright 2011 Manu Viswanathan
Object Description
| Title | Mobile robot obstacle avoidance using a computational model of the locust brain |
| Author | Viswanathan, Manu |
| Author email | mviswana@usc.edu; mvnathan@att.net |
| Degree | Master of Science |
| Document type | Thesis |
| Degree program | Computer Science (Robotics & Automation) |
| School | Viterbi School of Engineering |
| Date defended/completed | 2011-01-10 |
| Date submitted | 2011 |
| Restricted until | Unrestricted |
| Date published | 2011-01-26 |
| Advisor (committee chair) | Itti, Laurent |
| Advisor (committee member) |
Sukhatme, Gaurav Schaal, Stefan Celikel, Tansu |
| Abstract | The Lobula Giant Movement Detector (LGMD), a visual interneuron in the locust's brain, responds preferentially to objects approaching along collisional trajectories. The goal of the Robolocust project is to build a robot that interfaces with this neuron and uses the LGMD spikes to make steering decisions.; However, before interfacing with actual locusts, we would like to simulate the LGMD and develop suitable obstacle avoidance algorithms that take LGMD spikes as their sensory input. To this end, we have implemented a computational model of the LGMD that uses a laser range finder mounted on an iRobot Create to generate artificial spikes that are then fed into obstacle avoidance algorithms.; The main contribution of this thesis is a Bayesian state estimator that models the time-to-impact as a hidden variable and uses the LGMD spikes to gauge how far away the robot is from approaching obstacles. Additionally, we present three different LGMD-based obstacle avoidance algorithms that we have developed. |
| Keyword | mobile robotics; insect vision; collision detection; time-to-impact; obstacle avoidance; locust LGMD; probabilistic sensor model; Bayesian state estimation |
| Language | English |
| Part of collection | University of Southern California dissertations and theses |
| Publisher (of the original version) | University of Southern California |
| Place of publication (of the original version) | Los Angeles, California |
| Publisher (of the digital version) | University of Southern California. Libraries |
| Provenance | Electronically uploaded by the author |
| Type | texts |
| Legacy record ID | usctheses-m3629 |
| Rights | Viswanathan, Manu |
| Repository name | Libraries, University of Southern California |
| Repository address | Los Angeles, California |
| Repository email | http://www.usc.edu/isd/libraries/services/ask_a_librarian/email/ |
| Filename | etd-Viswanathan-4280 |
| Archival file | uscthesesreloadpub_Volume26/etd-Viswanathan-4280.pdf |
Description
| Title | Page 1 |
| Full text | MOBILE ROBOT OBSTACLE AVOIDANCE USING A COMPUTATIONAL MODEL OF THE LOCUST BRAIN by Manu Viswanathan A Thesis Presented to the FACULTY OF THE USC VITERBI SCHOOL OF ENGINEERING UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree MASTER OF SCIENCE (COMPUTER SCIENCE) May 2011 Copyright 2011 Manu Viswanathan |
Comments
Post a Comment for Page 1
