Abstract: Certain aspects of the present disclosure provide techniques for uncertainty estimation, such as for deep learning (DL)-based object tracking systems. A method generally includes processing, with a deep learning network, an input state associated with an object to predict an output state for the object, wherein the output state comprises a plurality of state estimates for a first time period; and generating a state covariance based at least in part on estimated covariance between at least two state estimates of the plurality of state estimates, wherein the state covariance represents an estimated uncertainty associated with the output state.

Abstract: Systems and techniques are provided for image processing for calibration. For example, a computing device can obtain image frames of a three-dimensional (3D) scene, each image frame including a plurality of two-dimensional (2D) points. Each 2D point corresponds to a 3D point. The computing device can determine a subset of 3D points by applying bundle adjustment on a set of 3D points (distributed over a field of view of the camera) and on fixed parameters of a camera. The computing device can determine inlier points from the subset of 3D points with a reprojection error less than a threshold value. The computing device can determine a final 3D points and final parameters of the camera by applying the bundle adjustment on the inlier points and on a prior set of camera parameters. The computing device can apply, to the camera, final intrinsic parameters of the final parameters of the camera.

Abstract: Systems and methods for spatial compatibility determination and weight determination include a base node transmitting a sequence of beamformed sounding signals to a plurality of nodes including at least two nodes; receiving, responsive to the sequence and at the base node, receive signal strength data and signal to interference data for each node of the plurality of nodes; and, calculating, at the base node, a compatibility metric for the at least two nodes. Scheduling the at least two nodes to use a same time-frequency resource based on the compatibility metric and, and transmitting signals using the same time-frequency resource to the at least two nodes, is described. Determining initial downlink and uplink payload transmit weights from the sequence of transmit beamforming sequences is described. Calculating a per-user transmit weight for a base station is also described. Predicting the expected channel quality, pathloss, and selection of signal modulation is described.

Abstract: Systems and techniques are described herein for tracking objects. For instance, a method for tracking objects is provided. The method may include generating features based on a sensor-data frame; detecting an object based on the features; generating a bounding box based on the object; tracking the bounding box over a plurality of sensor-data frames to generate a tracklet, wherein the tracklet comprises a respective bounding box for each sensor-data frame of the plurality of sensor-data frames and an identifier; and combining the bounding box and a bounding box of the tracklet to generate an output bounding box.

Abstract: Systems and techniques are provided for image processing for calibration. For example, a computing device can obtain image frames of a three-dimensional (3D) scene, each image frame including a plurality of two-dimensional (2D) points. Each 2D point corresponds to a 3D point. The computing device can determine a subset of 3D points by applying bundle adjustment on a set of 3D points (distributed over a field of view of the camera) and on fixed parameters of a camera. The computing device can determine inlier points from the subset of 3D points with a reprojection error less than a threshold value. The computing device can determine a final 3D points and final parameters of the camera by applying the bundle adjustment on the inlier points and on a prior set of camera parameters. The computing device can apply, to the camera, final intrinsic parameters of the final parameters of the camera.

Abstract: Systems and methods for spatial compatibility determination and weight determination include a base node transmitting a sequence of beamformed sounding signals to a plurality of nodes including at least two nodes; receiving, responsive to the sequence and at the base node, receive signal strength data and signal to interference data for each node of the plurality of nodes; and, calculating, at the base node, a compatibility metric for the at least two nodes. Scheduling the at least two nodes to use a same time-frequency resource based on the compatibility metric and, and transmitting signals using the same time-frequency resource to the at least two nodes, is described. Determining initial downlink and uplink payload transmit weights from the sequence of transmit beamforming sequences is described. Calculating a per-user transmit weight for a base station is also described. Predicting the expected channel quality, pathloss, and selection of signal modulation is described.

Abstract: An apparatus comprises one or more memories, and one or more processors in communication with the one or more memories. The one or processors are configured to execute an automated driving system having a modular hybrid architecture. The modular hybrid architecture includes a plurality of task units, and the modular hybrid architecture includes one or more human-defined interfaces, and one or more AI-defined interfaces. The one or more processors are further configured to receive input data from one or more sensors process the input data using the automated driving system having the modular hybrid architecture, and control at least one operation of a vehicle according to an output of the automated driving system.

Abstract: Methods, systems, and apparatuses to update or retrain a trained machine learning model/process based on edge ground truth data generated by edge devices. For example, a computing device may obtain, from a database, a first dataset of a first trained machine learning process and a second dataset of a second trained machine learning process. The first dataset may include at least a first value of a parameter of the first trained machine learning process. Additionally, the computing device may transmit, to a first edge device of a plurality of edge devices, a message including the first dataset and the second dataset. Further, the computing device may receive, from the first edge device, a third dataset associated with the message, and update a portion of the first dataset with the third dataset. The updated portion of the first dataset may include an updated value of the parameter.

Abstract: The disclosed invention makes use of a wheel speed signal to detect a wheel anomaly such as a loose wheel or a wheel with zero pressure. The wheel speed signal is used as a basis to determine a first and a second detection signal. A further basis for determining the first and second detection signals are a first and second reference signal, respectively. The anomaly of e.g. a loose wheel is detected, according to the teaching of the invention if at least one of the detection signals exceeds a threshold. In particular, the disclosure relates to methods, systems and computer program products to achieve the mentioned objective.

Abstract: Physical layer processing methods for network acquisition by remote nodes in wireless communication systems are described herein. New methods for wireless network discovery and synchronization by remote nodes are described herein that utilize spatial (e.g., antenna array) processing algorithms which may achieve enhanced functioning in challenging radio frequency environments, such as those containing interference and multipath distortion effects. These methods may include advantageous use of spatial whiteners and associated pluralities of adaptive beamformers to detect network reference and synchronization signals and estimate their parameters.

Abstract: Methods, systems, and apparatuses to update or retrain a trained machine learning model/process based on edge ground truth data generated by edge devices. For example, a computing device may obtain, from a database, a first dataset of a first trained machine learning process and a second dataset of a second trained machine learning process. The first dataset may include at least a first value of a parameter of the first trained machine learning process. Additionally, the computing device may transmit, to a first edge device of a plurality of edge devices, a message including the first dataset and the second dataset. Further, the computing device may receive, from the first edge device, a third dataset associated with the message, and update a portion of the first dataset with the third dataset. The updated portion of the first dataset may include an updated value of the parameter.

Abstract: The present disclosure describes systems and methods for the granular, interference aware selection of modulation and coding schemes (MCS) per-sub-band, per-stream, and in some examples, per-user. In some examples, channel condition metrics regarding wireless communication conditions, including interference conditions, from a communication node of a wireless access system may be received. Based at least on the received channel condition metrics indicative of interference, a multidimensional interference margin generator may determine a per-sub-band per-stream margin. A modulation and coding scheme may be selected based on the per-sub-band per-stream margin. In some examples, the selected modulation and coding scheme may be transmitted to various modulators/demodulators, encoders/decoders, and/or other communication nodes within the wireless access system. In some examples, a scheduler may select an allocation based at least on the per-sub-band per-stream margin.

Abstract: The disclosed invention makes use of slip related values to calculate friction related values and tire stiffness related values and feeds back an estimated tire stiffness relates value or a calculated friction related as a basis for further calculations. In particular, the disclosure relates to methods, apparatuses and computer program products to achieve the mentioned objective.

Abstract: The present disclosure describes systems and methods for selecting modulation and coding schemes (MCS) per-sub-band, and in some examples, per-stream. In some examples, channel condition metrics regarding wireless communication conditions from a communication node of a wireless access system may be received. Based at least on the received channel condition metrics, a modulation and coding scheme may be selected. In some examples, the selected modulation and coding scheme may be transmitted to various modulators/demodulators, encoders/decoders, and/or other communication nodes within the wireless access system.

Abstract: Physical layer processing methods for network acquisition by remote nodes in wireless communication systems are described herein. New methods for wireless network discovery and synchronization by remote nodes are described herein that utilize spatial (e.g., antenna array) processing algorithms which may achieve enhanced functioning in challenging radio frequency environments, such as those containing interference and multipath distortion effects. These methods may include advantageous use of spatial whiteners and associated pluralities of adaptive beamformers to detect network reference and synchronization signals and estimate their parameters.

Abstract: A system for monitoring the condition of a road surface travelled by a plurality of vehicles, each including at least one sensor, is provided. The system includes a central processing arrangement arranged to: map at least a part of the road surface with a number of cells; receive road surface data for the cells, which road surface data is based on measurements made by the sensors as the plurality of vehicles travel on the road surface; and calculate a probability for at least one road surface parameter for each cell travelled by the plurality of vehicles based at least on road surface data received from the plurality of vehicles.

Abstract: Physical layer processing methods for network acquisition by remote nodes in wireless communication systems are described herein. New methods for wireless network discovery and synchronization by remote nodes are described herein that utilize spatial (e.g., antenna array) processing algorithms which may achieve enhanced functioning in challenging radio frequency environments, such as those containing interference and multipath distortion effects. These methods may include advantageous use of spatial whiteners and associated pluralities of adaptive beamformers to detect network reference and synchronization signals and estimate their parameters.

Abstract: A system for monitoring the condition of a road surface travelled by a plurality of vehicles, each including at least one sensor, is provided. The system includes a central processing arrangement arranged to: map at least a part of the road surface with a number of cells; receive road surface data for the cells, which road surface data is based on measurements made by the sensors as the plurality of vehicles travel on the road surface; and calculate a probability for at least one road surface parameter for each cell travelled by the plurality of vehicles based at least on road surface data received from the plurality of vehicles.

Abstract: In accordance with one or more embodiments herein, a system for monitoring the condition of a road surface traveled by a plurality of vehicles, each comprising at least one sensor, is provided. The system comprises a central processing arrangement arranged to: map at least a part of the road surface with a number of cells; receive road surface data for the cells, which road surface data is based on measurements made by the sensors as the plurality of vehicles travel on the road surface; and calculate a probability for at least one road surface parameter for each cell traveled by the plurality of vehicles based at least on road surface data received from the plurality of vehicles.

Abstract: The disclosed invention makes use of slip related values to calculate friction related values and tire stiffness related values and feeds back an estimated tire stiffness relates value or a calculated friction related as a basis for further calculations. In particular, the disclosure relates to methods, apparatuses and computer program products to achieve the mentioned objective.