Engineering Papers⌕ Search

Engineering topics

Compton, Michael

Publications and source records attributed to Compton, Michael.

7 records
All sourcesNASA NTRSDOE OSTI
NASA NTRS1994

Goal Directed Model Inversion: A Study of Dynamic Behavior

Goal Directed Model Inversion (GDMI) is an algorithm designed to generalize supervised learning to the case where target outputs are not available to the learning system. The output of the learning system becomes the input to some external device or transformation, and only the output of this device or transformation can be compared to a desired target. The fundamental driving mechanism of GDMI is to learn from success. Given that a wrong outcome is achieved, one notes that the action that produced that outcome 0 "would have been right if the outcome had been the desired one." The algorithm then proceeds as follows: (1) store the action that produced the wrong outcome as a "target" (2) redefine the wrong outcome as a desired goal (3) submit the new desired goal to the system (4) compare the new action with the target action and modify the system by using a suitable algorithm for credit assignment (Back propagation in our example) (5) resubmit the original goal. Prior publications by our group in this area focused on demonstrating empirical results based on the inverse kinematic problem for a simulated robotic arm. In this paper we apply the inversion process to much simpler analytic functions in order to elucidate the dynamic behavior of the system and to determine the sensitivity of the learning process to various parameters. This understanding will be necessary for the acceptance of GDMI as a practical tool.

Colombano, Silvano P.↗
NASA NTRS1994

Goal Directed Model Inversion: Learning Within Domain Constraints

Goal Directed Model Inversion (GDMI) is an algorithm designed to generalize supervised learning to the case where target outputs are not available to the learning system. The output of the learning system becomes the input to some external device or transformation, and only the output of this device or transformation can be compared to a desired target. The fundamental driving mechanism of GDMI is to learn from success. Given that a wrong outcome is achieved, one notes that the action that produced that outcome "would have been right if the outcome had been the desired one." The algorithm makes use of these intermediate "successes" to achieve the final goal. A unique and potentially very important feature of this algorithm is the ability to modify the output of the learning module to force upon it a desired syntactic structure. This differs from ordinary supervised learning in the following way: in supervised learning the exact desired output pattern must be provided. In GDMI instead, it is possible to require simply that the output obey certain rules, i.e., that it "make sense" in some way determined by the knowledge domain. The exact pattern that will achieve the desired outcome is then found by the system. The ability to impose rules while allowing the system to search for its own answers in the context of neural networks is potentially a major breakthrough in two ways: 1) it may allow the construction of networks that can incorporate immediately some important knowledge, i.e. would not need to learn everything from scratch as normally required at present, and 2) learning and searching would be limited to the areas where it is necessary, thus facilitating and speeding up the process. These points are illustrated with examples from robotic path planning and parametric design.

Colombano, Silvano P.↗
NASA NTRS1994

Goal Directed Model Inversion: Learning Within Domain Constraints

Goal Directed Model Inversion (GDMI) is an algorithm designed to generalize supervised learning to the case where target outputs are not available to the learning system. The output of the learning system becomes the input to some external device or transformation, and only the output of this device or transformation can be compared to a desired target. The fundamental driving mechanism of GDMI is to learn from success. Given that a wrong outcome is achieved, one notes that the action that produced that outcome "would have been right if the outcome had been the desired one." The algorithm makes use of these intermediate "successes" to achieve the final goal. A unique and potentially very important feature of this algorithm is the ability to modify the output of the learning module to force upon it a desired syntactic structure. This differs from ordinary supervised learning in the following way: in supervised learning the exact desired output pattern must be provided. In GDMI instead, it is possible to require simply that the output obey certain rules, i.e., that it "make sense" in some way determined by the knowledge domain. The exact pattern that will achieve the desired outcome is then found by the system. The ability to impose rules while allowing the system to search for its own answers in the context of neural networks is potentially a major breakthrough in two ways: (1) it may allow the construction of networks that can incorporate immediately some important knowledge, i.e., would not need to learn everything from scratch as normally required at present; and (2) learning and searching would be limited to the areas where it is necessary, thus facilitating and speeding up the process. These points are illustrated with examples from robotic path planning and parametric design.

Colombano, Silvano P.↗
NASA NTRS1993

Portable computing - A fielded interactive scientific application in a small off-the-shelf package

Experience with the design and implementation of a portable computing system for STS crew-conducted science is discussed. Principal-Investigator-in-a-Box (PI) will help the SLS-2 astronauts perform vestibular (human orientation system) experiments in flight. PI is an interactive system that provides data acquisition and analysis, experiment step rescheduling, and various other forms of reasoning to astronaut users. The hardware architecture of PI consists of a computer and an analog interface box. 'Off-the-shelf' equipment is employed in the system wherever possible in an effort to use widely available tools and then to add custom functionality and application codes to them. Other projects which can help prospective teams to learn more about portable computing in space are also discussed.

Groleau, Nicolas↗
NASA NTRS1990

The 1990 progress report and future plans

This document describes the progress and plans of the Artificial Intelligence Research Branch (RIA) at ARC in 1990. Activities span a range from basic scientific research to engineering development and to fielded NASA applications, particularly those applications that are enabled by basic research carried out at RIA. Work is conducted in-house and through collaborative partners in academia and industry. Our major focus is on a limited number of research themes with a dual commitment to technical excellence and proven applicability to NASA short, medium, and long-term problems. RIA acts as the Agency's lead organization for research aspects of artificial intelligence, working closely with a second research laboratory at JPL and AI applications groups at all NASA centers.

Friedland, Peter↗
NASA NTRS1987

Exploratory evaluation of a moving-model technique for measurement of dynamic ground effects

An attempt is presently made to measure the dynamic or time-dependent ground effects that may be encountered by aircraft during approach and landing by means of a ground-based testing technique that employs a model moving horizontally over an upwardly inclined ground plane to simulate the rate of descent. Results have been thus obtained for a 60-deg delta wing and for an F-18 full configuration, with and without thrust reversal, at forward speeds of up to 100 ft/sec. An analysis of the results gathered indicates that ground effects were reduced when the rate of descent was included in the test conditions.

Kemmerly, Guy T.↗