University of Connecticut

At&t Intellectual Property I, L.p.

Toyota Motor Engineering & Manufacturing North America, Inc.

Krishna R Pattipati

Subhabrata Bapi Sen

Yanyuan Qin

Bing Wang

Shailesh N Joshi

Shuai Hao

Ali M Bazzi

Donald Mcmenemy

Paul M D'Orlando

Weiqiang Chen

Gregory R Leaper

Rhonda Walthall

Shalabh Gupta

Nayeff A Najjar

James Z Hare

John Kaminski

Ravi Rajamani

Muhamed K Farooq

Lingyi Zhang

James Meacham

Brian French

Yaakov Bar-Shalom

Bharath Pattipati

Travis Williams

Gopi Vinod Avvari

Tai-Sik Hwang

Ruofan Jin

Balakumar Balasingam

A method, system, and non-transitory computer readable medium describing an autoencoder that creates a reduced feature space from healthy power electronics devices for training. Devices under test are then encoded and compared to the encoded features of the healthy devices to determine health of the other devices. Contextual information is used to build multiple models that compare power electronics devices from similarly operated vehicles with one another.

Autoencoder utilizing vehicle contextual information

A system and method that determines whether a heat exchanger within a complex networked system is fouling is provided. The system and method includes training classifiers indicative of a plurality of fouling conditions associated with the heat exchanger and testing the classifiers with optimal sensor data from optimal sensors to determine whether the fouling is being experienced by the heat exchanger.

Computer program product for optimal sensor selection and fusion for heat exchanger fouling diagnosis in aerospace systems

Systems and methods of testing the health of vehicular power devices are disclosed herein. A method may include producing operating points as a function of cycling current (I) and voltage drain to source (V) when a subject device is conducting current. The method may further include determining a mean of moving distribution to adapt a center of the moving distribution contrasted with a plurality of known healthy devices. The method may also include indicating an imminent fault in the subject device based upon a discontinuity among operating points above a threshold.

Systems and methods for false-positive reduction in power electronic device evaluation

A processing system including at least one processor may obtain a selection of a desired perceptual quality for a video and selects a first video chunk from among a first plurality of video chunks, wherein the first plurality of video chunks is associated with a first time block of the video, wherein the selecting is based upon a manifest file of the video, wherein each video chunk of the first plurality of video chunks is associated with a respective perceptual quality and a respective encoding bitrate, wherein the first video chunk is part of a subset of the first plurality of video chunks for which the respective perceptual quality matches the desired perceptual quality within a predefined threshold tolerance, and wherein the respective encoding bitrate of the first video chunk is lower than a highest encoding bitrate associated with the subset of the first plurality of video chunks. The processing system may obtain the first video chunk from a server and store the first video chunk in a buffer.

Chunk-based filtering to optimize video streaming quality and data usage

To predict a failure condition in a power module of a vehicle, it is determined whether a discontinuity in statistical data characterizing physical measurements of the power module meets a threshold criterion. Responsive to the discontinuity meeting the threshold criterion, a data offset in the physical measurements is computed at the discontinuity. A shift correction is applied to the physical measurements in accordance with the computed data offset responsive to a determination that the discontinuity is attributable to a restart of the power module. Other statistical data characterizing the shift-corrected physical measurements are computed and the statistical data and the other statistical data are provided to a machine learning processor that predicts the failure condition in the power module.

Mitigating data offsets for machine learning

Systems and methods for detecting a fault or model mismatch are disclosed. A system includes a processor, a memory, and one or more sensors. The sensors may detect data associated with an electronic device. The memory may store processor executable instructions to: compute Tand Q statistics, over a time period, and apply a model mismatch and fault detection logic based on the Tand Q statistics. The model mismatch and fault detection logic may: count consecutive instances where a Tstatistic exceeds a Tthreshold via a Tcounter, update a probability of fault based on the Tcounter, count consecutive instances where a Q statistic exceeds a Q threshold via a Q counter, update a probability of model mismatch based on the Q counter, and detect one of a fault or a model mismatch based on a probability of fault threshold and a probability of model mismatch threshold.

Systems and methods for detecting a fault or a model mismatch

Devices, computer-readable media, and methods for selecting a video chunk from among a plurality of video chunks associated with a first time block of a video in accordance with a designated perceptual visual quality are disclosed. For instance, a processing system including at least one processor may obtain a selection of a designated perceptual visual quality for a video, select a video chunk from among a plurality of video chunks associated with a first time block of the video, based upon a manifest file of the video, where the video chunk is selected in accordance with the designated perceptual visual quality, and where each of the plurality of video chunks is associated with a respective perceptual visual quality. The processing system may further obtain the video chunk from a server and record the video chunk in a buffer.

Perceptual visual quality video chunk selection

Method of optimal sensor selection and fusion for heat exchanger fouling diagnosis in aerospace systems

Devices, computer-readable media, and methods for differential adaptive bitrate streaming based on scene complexity are disclosed. In one example, a method includes inferring, by a processor, a complexity of a scene contained in a first chunk of an adaptive bitrate streaming video, wherein the first chunk is a next chunk of the adaptive bitrate streaming video to be downloaded to an adaptive bitrate streaming video player, and wherein the complexity indicates a relative amount of spatial and temporal information contained in the scene, and selecting, by the processor, a first track from which to download the first chunk, wherein the first track is one of a plurality of tracks containing the first chunk, and wherein each track of the plurality of tracks encoded the first chunk at a different bitrate, and wherein a bitrate of the first track is proportional to the complexity of the scene as inferred.

Differential adaptive bitrate streaming based on scene complexity

Aspects of the subject disclosure may include, for example, communicating with a source of streaming media, wherein the streaming media is variable bitrate encoded, wherein the streaming media comprises a first plurality of chunks, wherein each of the first plurality of chunks corresponds to a first portion of the streaming media, wherein the streaming media comprises a second plurality of chunks, wherein each of the second plurality of chunks corresponds to a second portion of the streaming media, wherein the streaming media comprises a third plurality of chunks, and wherein each of the third plurality of chunks corresponds to a third portion of the streaming media; obtaining, from the source, a first one of the first plurality of chunks; and determining which of the second plurality of chunks and third plurality of chunks to obtain. Other embodiments are disclosed.

Apparatus, storage medium and method for adaptive bitrate streaming adaptation of variable bitrate encodings

A method includes calculating a first estimated state of charge (SOC) of a battery at a first time, receiving a voltage value representing a measured voltage across the battery at a second time, calculating a filter gain at the second time, and calculating a second estimated SOC of the battery at the second time based on the first estimated SOC, the voltage value, and the filter gain. Another method includes storing, in a memory, a library of equivalent circuit models representing a battery, determining an operational mode of a battery based on a load associated with the battery, selecting one of the equivalent circuit models based on the determined operational mode, and calculating a state of charge of charge (SOC) of the battery using the selected equivalent circuit model.

Battery state of charge tracking, equivalent circuit selection and benchmarking

Aspects of the subject disclosure may include, for example, embodiments include obtaining a first estimated bandwidth capacity of a communication network and selecting a target buffer level to temporarily store media content. Further embodiments include requesting a first portion of the media content according to the first estimated bandwidth capacity and the target buffer level and receiving the first portion of the media content. Additional embodiments include determining a current buffer level and receiving a second estimated bandwidth capacity of the communication network. Also, embodiments include requesting a second portion of the media content according to the current buffer level and the second estimated bandwidth capacity. Other embodiments are disclosed.

Growing community of inventors

Mansfield, CT, United States of America

Krishna R Pattipati