An examination of Echoic Flow based autonomous guidance using the Lego Mindstorms NXT robot
Loading...
Date
2014-12
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
The Ohio State University
Abstract
It is well known that some animals, such as bats, can find their routes home autonomously and are able to avoid crashing into each other while traveling in a group. They do this using echolocation that enables them to function at dusk and in the dark. My research project is to investigate how these complicated behaviors can be replicated using robot vehicles that have echolocation sensors. Echoic flow fields are computed and simple rules to govern subsequent behavior are implemented. Echoic flow is defined to be the ratio of a sensed parameter such as range or intensity to a change in that parameter per unit time. The ratio of these two quantities gives the time over which two bodies will come into contact (collide). This “time to collision” can be used to provide feedback to the robot to either avoid collisions or to control the form of a collision. Previous theoretical research has shown that echoic flow can be used to control the behaviors of objects in relative motion. Experimental work has shown that the Lego NXT robot and its ultrasonic echolocation sensors can enable obstacle avoidance. My project develops extends this earlier experimental research to determine if echoic flow can be used to control two robots so that one leads and has to avoid obstacles as it autonomously navigates a course while the other follows the first robot maintaining a constant time to collision.
Description
Distinguished Undergraduate Research Proposal from The College of Engineering
Keywords
Echo Location, Echoic Flow, Ultrasound wave, ultrasonic sensor measuring, cognitive science, robot control, calibration