The Aerospace Corporation

Terence S Yeoh

Stephen La Lumondiere

Nehal Desai

Neil A Ives

Martin Siu Wo Leung

George C Valley

Adam Wayne Bushmaker

David Cardoza

William T Lotshaw

Steven C Moss

Artificial intelligence (AI) techniques that map disallowed states and enable access to those states under certain conditions through a search algorithm are disclosed. In other words, scenario boundaries may be crossed by jumping from one scenario that is less desirable or even has no solution to another scenario that is more desirable.

Low probability transitions and boundary crossing into disallowed states for a more optimal solution

An artificial intelligence (AI) system is disclosed that employs windowed cellular automata to create plausible alternatives. A cellular automata-based technique may be utilized to perform pattern recognition and assess the best path available (i.e., 'instant improv'). Alternative sequences (i.e., 'pattern improv') may also be used to determine alternative paths. This instant improv and pattern improv may then be used to create completely new, plausible alternative nodes. The subsequent evaluation of the sentiment further creates new, dynamic capabilities. Through the use of windowed memory learning, recall, and interpolation, new plausible structures are generated that predict dynamic systems.

Artificial intelligence system that employs windowed cellular automata to create plausible alternatives

An architecture and process are provided that encodes information into a cellular automata memory structure such that it can be recalled utilizing unique memory anchors (engrams) in a manner that both identifies and relates each piece of information relative to other data points. The automata may be individually programmable with a limited, local ruleset that activates other cellular automata based on prior patterns that were fed into the array. Deep Learning Neural Network (DLNN) systems may be probed to understand what discriminators are being used to classify the data, which is not possible with conventional DLNN techniques.

Programmable cellular automata for memory search, recall, simulation, and improvisation

An image detection assembly includes a light source that is configured to generate at least one pulsed light beam. A modulator is configured to direct the pulsed light beam onto a device via a plurality of light patterns such that a plurality of electrical signals are generated by the device. Each electrical signal corresponds to a different light pattern. A signal processing apparatus is coupled to the device and the signal processing apparatus is configured to receive the electrical signals and to digitize each electrical signal to record a plurality of signal vectors such that each signal vector corresponds to a different electrical signal. The signal processing apparatus is also configured to generate at least one image output based, at least in part, on the recorded signal vectors and the light patterns such that the image output enables a determination of at least one transient effect on the device.

Image detection assembly and method for use in determining transient effects

Various embodiments are directed to imaging systems and methods for generating an image of a sub-surface feature of an object through a surface of the object. An illumination array may comprise a plurality of illumination sources positioned around the sub-surface feature of the object. An imaging device may comprise an objective. A computer system may be in communication with the illumination array. The computer system may be programmed to calculate an optimized illumination pattern of the plurality of illumination sources for imaging the sub-surface feature; activate the optimized illumination pattern; and instruct the imaging device to capture an image of the sub-surface feature with the imaging device based on reflected illumination from the optimized illumination pattern.

Optimized illumination for imaging

An illumination source may be directed towards a surface of an object comprising subsurface features, wherein the illumination from the source is directed at a first angle relative to the normal of the surface. The object may have a portion between the subsurface features and the surface, the portion having an index of refraction that is greater than the index of refraction of a surrounding medium that surrounds the object. An imaging device may be placed with an objective lens. The first angle may be larger than an acceptance angle of the objective lens. In some embodiments, multiple illumination beams may be generated by one or more illumination sources. The beams may be rotated relative to one another about the normal of the surface. Also, in some embodiments, multiple images may be taken with the objective of the imaging device at different positions rotated off of the normal. The multiple images may be combined to generate a composite image.

Refraction assisted illumination for imaging

Various embodiments are directed to systems and methods for imaging subsurface features of a semiconductor object comprising a first region having a first doping property and a second region having a second doping property. The semiconductor object may comprise subsurface features and material between a surface of the semiconductor object and the subsurface features. The material may have an index of refraction that is greater than an index of refraction of a surrounding medium in contact with the surface of the semiconductor object. For example, a system may comprise an imaging device comprising an objective. The imaging device may be sensitive to a first wavelength. The system may also comprise an illumination source to emit illumination substantially at the first wavelength. The illumination may be directed towards the surface of the semiconductor object at a first angle relative to a normal of the surface. The first angle is greater than an acceptance angle of the objective of the imaging device. Also, the first wavelength may have a photonic energy substantially equal to a bandgap of the first region.

Various embodiments are directed to systems and methods of imaging subsurface features of objects. An illumination source may be directed towards a surface of an object comprising subsurface features at a first angle relative to the normal of the surface. The object may have a portion between the subsurface features and the surface that has an index of refraction that is greater than the index of refraction of a surrounding medium. An imaging device may be placed with an objective lens oriented substantially normal to the surface. The first angle may be larger than an acceptance angle of the objective lens.

Growing community of inventors

El Segundo, CA, United States of America

Terence S Yeoh