Project Title:
Multilevel Motion Processing for Autonomous Helicopters
03.04-7300
Multilevel Motion Processing for Autonomous Helicopters
Advanced Decision Systems
1500 Plymouth Street
Mountain View
CA
94043-1230
Daryl T.
Lawton
(415-960-7300)
ARC
Abstract:
A fundamental problem with automating nap-of-the-earth helicopter flight is the extraction
and representation of information about the rapidly changing relationship between
the helicopter and its environment. In this project, analysis of motion involves
several explicit levels of processing and representation for determining image motion
and building environmental inferences. The different levels, which constrain and
direct each other significantly, range from very local and simple descriptions of
temporal changes in image intensity to qualitative and potentially robust descriptions
of the relative depths of coherent environmental surfaces and occlusion boundaries.
This range of representation and constraints can effectively address problems associated
with the typically brittle environmental inferences associated with motion processing.
The results will supply NASA with an extensive set of algorithms that can be tested
and combined. This is particularly important since the exact use and nature of motion
processing for helicopter systems will vary based upon several evolving factors.
Potential Commercial Application:
Potential Commercial Applications: Analysis of dynamic visual information could
be applied to industrial inspection, surveillance, change detection, and robot guidance
and navigation.