Abstract: A deep neural network(s) (DNN) may be used to detect objects from sensor data of a three dimensional (3D) environment. For example, a multi-view perception DNN may include multiple constituent DNNs or stages chained together that sequentially process different views of the 3D environment. An example DNN may include a first stage that performs class segmentation in a first view (e.g., perspective view) and a second stage that performs class segmentation and/or regresses instance geometry in a second view (e.g., top-down). The DNN outputs may be processed to generate 2D and/or 3D bounding boxes and class labels for detected objects in the 3D environment. As such, the techniques described herein may be used to detect and classify animate objects and/or parts of an environment, and these detections and classifications may be provided to an autonomous vehicle drive stack to enable safe planning and control of the autonomous vehicle.

Abstract: A deep neural network(s) (DNN) may be used to detect objects from sensor data of a three dimensional (3D) environment. For example, a multi-view perception DNN may include multiple constituent DNNs or stages chained together that sequentially process different views of the 3D environment. An example DNN may include a first stage that performs class segmentation in a first view (e.g., perspective view) and a second stage that performs class segmentation and/or regresses instance geometry in a second view (e.g., top-down). The DNN outputs may be processed to generate 2D and/or 3D bounding boxes and class labels for detected objects in the 3D environment. As such, the techniques described herein may be used to detect and classify animate objects and/or parts of an environment, and these detections and classifications may be provided to an autonomous vehicle drive stack to enable safe planning and control of the autonomous vehicle.

Abstract: Methods, apparatus, systems and articles of manufacture are disclosed for augmented reality vehicle condition inspection. An example apparatus disclosed herein includes a location analyzer to determine whether a camera is at an inspection location and directed towards a first vehicle in an inspection profile, the inspection location corresponding to a location of the camera relative to the first vehicle, an interface generator to generate an indication on a display that the camera is at the inspection location, the indication associated with an inspection image being captured, and an image analyzer to compare the inspection image captured at the inspection location with a reference image taken of a reference vehicle of a same type as the first vehicle, and determine a vehicle part condition or a vehicle condition based on the comparison of the inspection image and the reference image.

Abstract: Methods, apparatus, systems and articles of manufacture are disclosed for augmented reality measuring of equipment. An example apparatus disclosed herein includes an image comparator to compare camera data with reference information of a reference vehicle part to identify a vehicle part included in the camera data, and an inspection image analyzer to, in response to the image comparator identifying the vehicle part, measure the vehicle part by causing an interface generator to generate an overlay representation of the reference vehicle part on the camera data displayed on a user interface, and determining, based on one or more user inputs to adjust the overlay representation, a measurement corresponding to the vehicle part.

Abstract: A computer-implemented method, system and computer program product for representing source code in vector space. The source code is parsed into an abstract syntax tree, which is then traversed to produce a sequence of tokens. Token embeddings may then be constructed for a subset of the sequence of tokens, which are inputted into an encoder artificial neural network (“encoder”) for encoding the token embeddings. A decoder artificial neural network (“decoder”) is initialized with a final internal cell state of the encoder. The decoder is run the same number of steps as the encoding performed by the encoder. After running the decoder and completing the training of the decoder to learn the inputted token embeddings, the final internal cell state of the encoder is used as the code representation vector which may be used to detect errors in the source code.

Abstract: A method, system and computer program product for obtaining vector representations of code snippets capturing semantic similarity. A first and second training set of code snippets are collected, where the first training set of code snippets implements the same function representing semantic similarity and the second training set of code snippets implements a different function representing semantic dissimilarity. A vector representation of a first and second code snippet from either the first or second training set of code snippets is generated using a machine learning model. A loss value is generated utilizing a loss function that is proportional or inverse to the distance between the first and second vectors in response to receiving the first and second code snippets from the first or second training set of code snippets, respectively. The machine learning model is trained to capture the semantic similarity in the code snippets by minimizing the loss value.

Abstract: A magnetic flow metering device and method is disclosed for the measurement of corrosive flow streams. The device utilizes a unibody construction wherein the flow conduit is constructed entirely from an insulative, non-conducting material without resorting to a metallic outer housing. The portions of the electrodes in contact with the flow stream are made of a suitable conductive polymer material, resistant to the corrosive media. The electrodes also feature shields that are molded into the electrode assembly to reduce background electrical noise. The invention also utilizes an electrical configuration that actively drives the electrode shield circuit (electrodes as well as cabling) to provide a more accurate measurement of the electromotive force.

Abstract: A process pressure fluid transmitter is coupled to a differential pressure producer and measures a differential pressure across the producer. A statistical parameter is calculated upon successive differential pressure indications, and the calculated parameter is used to indicate wet gas flow. The indication can be generated locally at the process pressure transmitter, communicated via a wired process communication loop, communicated wirelessly, or any combination thereof.

Abstract: An industrial field device provides information indicative of a process variable. The field device includes a controller, communication circuitry, a process variable sensor and measurement circuitry. The communication circuitry is coupled to the controller. The process variable sensor has an electrical characteristic that changes based on a variable of a process fluid. The measurement circuitry is coupled to the process variable sensor and coupled to the controller. The controller is configured to generate communication via the communication circuitry relative to the process variable, and to provide an indication of sand flow or other solids in the fluid.

Abstract: A magnetic flow metering device and method is disclosed for the measurement of corrosive flow streams. The device utilizes a unibody construction wherein the flow conduit is constructed entirely from an insulative, non-conducting material without resorting to a metallic outer housing. The portions of the electrodes in contact with the flow stream are made of a suitable conductive polymer material, resistant to the corrosive media. The electrodes also feature shields that are molded into the electrode assembly to reduce background electrical noise. The invention also utilizes an electrical configuration that actively drives the electrode shield circuit (electrodes as well as cabling) to provide a more accurate measurement of the electromotive force.

Abstract: A device for diagnosing operation of an industrial process control or monitoring system includes an input configured to receive an input related to a process signal. A first statistical parameter module provides a first statistical parameter output related to a statistical parameter of the process signal. A filter provides a filter output related to a filtered value of the process signal. A second statistical parameter module provides a second statistical parameter output related to a statistical parameter of the filter output. A diagnostic module diagnoses operation of the industrial process based upon the first and second statistical parameters.

Abstract: A flow meter for measuring flow of a process fluid includes a sensor configured to provide a sensor output signal. The sensor output signal is related to flow of the process fluid. Circuitry in the flow meter is configured to determine a statistical parameter related to sensor output signal. Diagnostic circuitry provides a diagnostic output based upon the determined statistical parameter and the sensor output signal.

Abstract: A magnetic flow metering device and method is disclosed for the measurement of corrosive flow streams. The device utilizes a unibody construction wherein the flow conduit is constructed entirely from an insulative, non-conducting material without resorting to a metallic outer housing. The portions of the electrodes in contact with the flow stream are made of a suitable conductive polymer material, resistant to the corrosive media. The electrodes also feature shields that are molded into the electrode assembly to reduce background electrical noise. The invention also utilizes an electrical configuration that actively drives the electrode shield circuit (electrodes as well as cabling) to provide a more accurate measurement of the electromotive force.