Raytheon Technologies Corporation

United Technologies Corporation

Otis Elevator Company

Carrier Corporation

Goodrich Corporation

Kishore K Reddy

Michael J Giering

Edgar A Bernal

Ryan B Noraas

Vivek Venugopalan

Kin Gwn Lore

Soumalya Sarkar

Zhen Jia

Alan Matthew Finn

Michael J Birnkrant

Marcin Piech

David L Lincoln

Enrico Manes

Walter Thomas Schmidt

Catherine Thibaud

Paul R Braunwart

Amit Surana

Yanzhi Chen

Luca F Bertuccelli

Jaume Amores

Jose-Rodrigo Castillo-Garza

Charles C Coffin

Shihemn Chen

Anais Espinal

Sofiane Yous

Keith J Power

Soumik Sarkar

George S Ekladious

An example SPR detection system includes a first prism having a first surface adjacent to a first metal layer exposed to a sample gas, and a second prism having a second surface adjacent to a second metal layer exposed to a reference gas. At least one light source is configured to provide respective beams to the first and second surfaces, where each of the beams causes SPR of a respective one of the metal layers. At least one photodetector is configured to measure a reflection property of reflections of the respective beams from the metal layers during the SPR. A controller is configured to determine whether a target gas is present in the sample gas based on a known composition of the reference gas and at least one of an electrical property of the first and second metal layers during the SPR and the reflection property of the metal layers.

Surface plasmon resonance detection system

A method of monitoring a conveyance apparatus within a conveyance system including: detecting a vibratory signature at a primary location during a commissioning phase using a primary sensing apparatus located at the primary location; detecting a vibratory signature at a secondary location during the commissioning phase using a secondary sensing apparatus located at the secondary location; determining a transfer algorithm in response to the vibratory signature at the primary location during the commissioning phase and the vibratory signature at the secondary location during the commissioning phase; detecting a vibratory signature at the primary location during a normal operation phase using the primary sensing apparatus located at the primary location; and converting the vibratory signature at the primary location during the normal operation phase to a vibratory signature at the secondary location during the normal operation phase using the transfer algorithm.

Virtual sensor for elevator monitoring

According to an embodiment of the present disclosure, a method of training a machine learning model is provided. Input data is received from at least one remote device. A classifier is evaluated by determining a classification accuracy of the input data. A training data matrix of the input data is applied to a selected context autoencoder of a knowledge bank of autoencoders including at least one context autoencoder and the training data matrix is determined to be out of context for the selected autoencoder. The training data matrix is applied to each other context autoencoder of the at least one autoencoder and the training data matrix is determined to be out of context for each other context autoencoder. A new context autoencoder is constructed.

System and method for context-based training of a machine learning model

A method includes fusing multi-modal sensor data from a plurality of sensors having different modalities. At least one region of interest is detected in the multi-modal sensor data. One or more patches of interest are detected in the multi-modal sensor data based on detecting the at least one region of interest. A model that uses a deep convolutional neural network is applied to the one or more patches of interest. Post-processing of a result of applying the model is performed to produce a post-processing result for the one or more patches of interest. A perception indication of the post-processing result is output.

Multi-modal sensor data fusion for perception systems

A method for searching video includes receiving a video query, the video query including a first video query term identifying a first foreground object, a second video query term identifying a second foreground object, and a third video query term identifying a spatiotemporal relationship; searching the video in response to the video query for the first foreground object and the second foreground object within the spatiotemporal relationship; and generating a search result in response to the searching.

Video searching using multiple query terms

Data indicative of a plurality of observations of an environment are received at a control system. Machine learning using deep reinforcement learning is applied to determine an action based on the observations. The deep reinforcement learning applies a convolutional neural network or a deep auto encoder to the observations and applies a training set to locate one or more regions having a higher reward. The action is applied to the environment. A reward token indicative of alignment between the action and a desired result is received. A policy parameter of the control system is updated based on the reward token. The updated policy parameter is applied to determine a subsequent action responsive to a subsequent observation.

Control systems using deep reinforcement learning

Disclosed is an escalator system that has: an escalator step; and load sensors secured to the escalator step, wherein the load sensors are configured to: sense an escalator loading; and transfer, to an escalator controller, sensor data indicative of the escalator loading.

Escalator steps with strain sensors

A method includes detecting at least one region of interest in a frame of image data. One or more patches of interest are detected in the frame of image data based on detecting the at least one region of interest. A model including a deep convolutional neural network is applied to the one or more patches of interest. Post-processing of a result of applying the model is performed to produce a post-processing result for the one or more patches of interest. A visual indication of a classification of defects in a structure is output based on the result of the post-processing.

Deep convolutional neural networks for crack detection from image data

A method of compressing data in the context of a decision-making task includes receiving raw data, analyzing the raw data to determine content of the raw data, and adjusting one or more one data compression parameters in a compression algorithm. The adjustment of the one or more compression parameters is based on the content of the raw data and a received decision-making task to produce a modified compression algorithm. The raw data is thereafter compressed using the modified compression algorithm and output as compressed data.

Systems and methods for compressing data

A material characterization system includes an imaging unit, a material characterization controller, and an imaging unit controller. The electronic imaging unit generates a test image of a specimen composed of a material. The electronic material characterization controller determines values of a plurality of parameters and maps the parameters to corresponding ground truth labeled outputs. The mapped parameters are applied to at least one test image to predict a presence of at least one target attribute of the specimen in response to applying the learned parameters. The test image is convert to a selected output image format so as to generate a synthetic image including the predicted at least one attribute. The electronic imaging unit controller performs a material characterization analysis that characterizes the material of the specimen based on the predicted at least one attribute included in the synthetic image.

Automated material characterization system including conditional generative adversarial networks

According to one embodiment, a method of identifying a part of a conveyance system is provided. The method comprising: capturing an image of a part of a conveyance system using a camera; classifying the part of the conveyance system using supervised learning; and displaying a classification of the part of the part on a mobile computing device.

Part recognition and damage characterization using deep learning

An imaging method includes obtaining an image with a first field of view and first effective resolution and the analyzing the image with a visual attention algorithm to one identify one or more areas of interest in the first field of view. A subsequent image is then obtained for each area of interest with a second field of view and a second effective resolution, the second field of view being smaller than the first field of view and the second effective resolution being greater than the first effective resolution.

Active visual attention models for computer vision tasks

A sensor system may comprise a sensor; a processor in electronic communication with the sensor; and/or a tangible, non-transitory memory configured to communicate with the processor, the tangible, non-transitory memory having instructions stored thereon that, in response to execution by the processor, cause the processor to perform operations. The operations may comprise recording, by the sensor, a low quality data sample; and/or applying, by the processor, a mapping function having a plurality of tuned parameters to the low quality data sample, producing a high quality data output.

Sensor system for data enhancement

A sensor system may comprise a sensor; a processor in electronic communication with the sensor; and/or a tangible, non-transitory memory configured to communicate with the processor, the tangible, non-transitory memory having instructions stored thereon that, in response to execution by the processor, cause the processor to perform operations. The operations may comprise recording, by the sensor, a preliminary type data sample; and/or applying, by the processor, a mapping function having a plurality of tuned parameters to the preliminary type data sample, producing a desired type data output.

Growing community of inventors

Vernon, CT, United States of America

Kishore K Reddy