Abstract: The method for performing real-time probabilistic safety assessments for nuclear power plants is an artificially intelligent, self-updating Living probabilistic safety assessment (PSA). An initial probabilistic safety assessment model is developed for a particular nuclear power plant. A simulation of the nuclear power plant is developed, and real-time parameters are also measured from the nuclear power plant. Feature extraction is performed on the results of the probabilistic safety assessment model combined with corresponding results from the simulation and the measured real-time parameters. A set of system state labels is generated, and an updated probabilistic safety assessment model is developed using the extracted features and the set of system state labels as training data for a machine learning algorithm.

Abstract: An open-ear device performs active noise cancellation (ANC) for a user. A sensor located outside an ear of a user which does not occlude an ear canal of the ear and measures vibrational data indicative of a sound pressure level at a location outside the ear, or a level of pinna vibration of the user. A prediction pipeline generates a prediction of sound pressure within the ear canal using an individualized model, taking into account the measured vibrational data and the unique geometric shape of the user's head and pinna. This sound pressure prediction is used to generate audio instructions for rendering playback at an noise cancellation source, such as a bone conduction transducer and/or cartilage transducer, to perform ANC for the user by cancelling at least portion of the sound received at the ear canal.

Abstract: The present disclosure provides neural fields for sparse novel view synthesis of outdoor scenes. Given just a single or a few input images from a novel scene, the disclosed technology can render new 360° views of complex unbounded outdoor scenes. This can be achieved by constructing an image-conditional triplanar representation to model the 3D surrounding from various perspectives. The disclosed technology can generalize across novel scenes and viewpoints for complex 360° outdoor scenes.

Abstract: An apparatus for training a deep learning model for classifying emotions from biosignals includes: a memory configured to store a program for training the deep learning model; and a processor configured to train the deep learning model by executing the program, wherein, when the processor executes the program, the processor inputs an input matrix to an attention layer constituting the deep learning model, the input matrix being composed of a plurality of features each mapped to a plurality of channels and a plurality of feature groups as the biosignals are acquired from a plurality of channels and the biosignals acquired from each channel are divided into the plurality of feature groups, and the attention layer operates to mask the input matrix using an attention matrix in which an importance of features in each channel is reflected.

Abstract: An apparatus, including processing unit (PU) cores and computer readable storage devices storing machine instructions for determining a distance between a target object and a radar sensor circuit. The PU cores receive a beat signal generated by the radar sensor circuit and compensate for a phase difference between the received beat signal and a reconstruction of the received beat signal to obtain a phase compensated beat signal. The phase compensated beat signal is then filtered to remove spurious reflections by demodulating the phase compensated beat signal using an estimated frequency of the phase compensated beat signal. The PU cores then apply a low pass filter to the demodulated phase compensated beat signal, resulting in a modified beat signal. The PU cores then determine the distance between the target object and the radar sensor circuit using the modified beat signal.

Abstract: A high pressure, high temperature spray cooling system for heat transfer. The system includes: a supply tank having a heater and a high pressure pump; and a spray chamber spaced apart from the supply tank. The spray chamber has a workpiece receiving bed and a nozzle spaced apart from the workpiece receiving bed. The supply line has a first supply line end and a second supply line end. The first supply line end is connected to the supply tank and the second supply line end is connected to the nozzle so that the contents within the supply tank are supplied to the nozzle through the supply line at a high pressure to cool a workpiece on the workpiece receiving bed.

Abstract: The disclosure provides implicit representations for multi-object 3D shape, 6D pose and size, and appearance optimization, including obtaining shape, 6D pose and size, and appearance codes. Training is employed using shape and appearance priors from an implicit joint differential database. 2D masks are also obtained and are used in an optimization process that utilizes a combined loss minimizing function and an Octree-based coarse-to-fine differentiable optimization to jointly optimize the latest shape, appearance, pose and size, and 2D masks. An object surface is recovered from the latest shape codes to a desired resolution level. The database represents shapes as Signed Distance Fields (SDF), and appearance as Texture Fields (TF).

Abstract: A method for multi-sensor calibration includes imaging a calibration target with a first sensor using a first modality to obtain a first set of data and a second sensor using a second modality that is different from the first modality to obtain a second set of data. A border of the calibration target is identified based on the first set of data. A first centroid location of the calibration target is identified based on the border of the calibration target. A border of a pattern disposed on the calibration target is identified based on the second set of data. A second centroid location of the calibration target is identified based on the border of the pattern. Calibration data for the first sensor and the second sensor is generated based on the first centroid location and the second centroid location.

Abstract: A computing device inputs a sample generated from an electrocardiogram signal to a heartbeat classification model, generates a feature map from the sample through multiple first layers of the heartbeat classification model, generates an attention mask based on an assistant feature generated from the feature map and the sample, generates a masked feature map by masking the feature map with the attention mask; and performs classification of the sample from the masked feature map through a second layer of the heartbeat classification model.

Abstract: System, methods, and other embodiments described herein relate to single-shot multi-object three-dimensional (3D) shape reconstruction and categorical six-dimensional (6D) pose and size estimation. In one embodiment, a method includes inferring a heatmap based upon a feature pyramid, where the feature pyramid is generated based upon a red green blue depth (RGB-D) image that includes objects. The method further includes sampling a 3D parameter map at locations corresponding to peaks in the heatmap, where the 3D parameter map is inferred based upon the feature pyramid, and where the locations include latent shape codes, 6D poses, and one-dimensional (1D) scales. The method further includes generating point clouds based upon the latent shape codes, the 6D poses, and the 1D scales.

Abstract: A device includes one or more processors configured to receive radar data, and generate a plurality of occupancy grid maps based on the radar data. Each of the occupancy grid maps corresponds to a respective one of a plurality of candidate angles. The one or more processors is also configured to select one of the candidate angles as a sensor mount angle based on the occupancy grid maps, and trigger an action based on the sensor mount angle and the radar data.

Abstract: A device includes one or more processors configured to receive radar data, and generate a plurality of occupancy grid maps based on the radar data. Each of the occupancy grid maps corresponds to a respective one of a plurality of candidate angles. The one or more processors is also configured to select one of the candidate angles as a sensor mount angle based on the occupancy grid maps, and trigger an action based on the sensor mount angle and the radar data.

Abstract: An apparatus, including processing unit (PU) cores and computer readable storage devices storing machine instructions for determining a distance between a target object and a radar sensor circuit. The PU cores receive a beat signal generated by the radar sensor circuit and compensate for a phase difference between the received beat signal and a reconstruction of the received beat signal to obtain a phase compensated beat signal. The phase compensated beat signal is then filtered to remove spurious reflections by demodulating the phase compensated beat signal using an estimated frequency of the phase compensated beat signal. The PU cores then apply a low pass filter to the demodulated phase compensated beat signal, resulting in a modified beat signal. The PU cores then determine the distance between the target object and the radar sensor circuit using the modified beat signal.

Abstract: A method, apparatus and system for efficient navigation in a navigation space includes determining semantic features and respective 3D positional information of the semantic features for scenes of captured image content and depth-related content in the navigation space, combining information of the determined semantic features of the scene with respective 3D positional information using neural networks to determine an intermediate representation of the scene which provides information regarding positions of the semantic features in the scene and spatial relationships among the sematic features, and using the information regarding the positions of the semantic features and the spatial relationships among the sematic features in a machine learning process to provide at least one of a navigation path in the navigation space, a model of the navigation space, and an explanation of a navigation action by the single, mobile agent in the navigation space.

Abstract: In the proposed low complexity technique a hierarchical approach is created. An initial FFT based detection and range estimation gives a coarse range estimate of a group of objects within the Rayleigh limit or with varying sizes resulting from widely varying reflection strengths. For each group of detected peaks, demodulate the input to near DC, filter out other peaks (or other object groups) and decimate the signal to reduce the data size. Then perform super-resolution methods on this limited data size. The resulting distance estimations provide distance relative to the coarse estimation from the FFT processing.

Abstract: An apparatus, including processing unit (PU) cores and computer readable storage devices storing machine instructions for determining a distance between a target object and a radar sensor circuit. The PU cores receive a beat signal generated by the radar sensor circuit and compensate for a phase difference between the received beat signal and a reconstruction of the received beat signal to obtain a phase compensated beat signal. The phase compensated beat signal is then filtered to remove spurious reflections by demodulating the phase compensated beat signal using an estimated frequency of the phase compensated beat signal. The PU cores then apply a low pass filter to the demodulated phase compensated beat signal, resulting in a modified beat signal. The PU cores then determine the distance between the target object and the radar sensor circuit using the modified beat signal.

Abstract: A vehicle includes a generator, an outlet, and a controller. The generator is electrically connected to the outlet. The controller is programmed to, responsive to outlet temperature being less than a threshold, deliver an electric current from the generator to an external device that is connected to the outlet. The controller is further programmed to, responsive to the outlet temperature exceeding the threshold, inhibit delivering the current from the electric machine to the external device.

Abstract: In the proposed low complexity technique a hierarchical approach is created. An initial FFT based detection and range estimation gives a coarse range estimate of a group of objects within the Rayleigh limit or with varying sizes resulting from widely varying reflection strengths. For each group of detected peaks, demodulate the input to near DC, filter out other peaks (or other object groups) and decimate the signal to reduce the data size. Then perform super-resolution methods on this limited data size. The resulting distance estimations provide distance relative to the coarse estimation from the FFT processing.

Abstract: A device includes one or more processors configured to receive radar data, and generate a plurality of occupancy grid maps based on the radar data. Each of the occupancy grid maps corresponds to a respective one of a plurality of candidate angles. The one or more processors is also configured to select one of the candidate angles as a sensor mount angle based on the occupancy grid maps, and trigger an action based on the sensor mount angle and the radar data.

Abstract: A method for multi-sensor calibration includes imaging a calibration target with a first sensor using a first modality to obtain a first set of data and a second sensor using a second modality that is different from the first modality to obtain a second set of data. A border of the calibration target is identified based on the first set of data. A first centroid location of the calibration target is identified based on the border of the calibration target. A border of a pattern disposed on the calibration target is identified based on the second set of data. A second centroid location of the calibration target is identified based on the border of the pattern. Calibration data for the first sensor and the second sensor is generated based on the first centroid location and the second centroid location.