Abstract: From each of a plurality of cameras, a visual input of a location is received over a network. For each visual input from the plurality of cameras, a coupling correction is performed between a shaking of the camera with respect to the visual input by subtracting velocity vectors of the plurality of cameras from velocity vectors of pixels defining the visual input to provide a processed input. It is determined whether a shaking identified in the processed input is above a predetermined threshold based on the processed input, thereby detecting one or more anomalies. From the one or more anomalies, at least one of a location, magnitude, or depth of an earthquake are inferred based on the shaking identified in the processed input of each of the plurality of cameras.

Abstract: A computer implemented method rasterizes point cloud data. A number of processor units rasterizes the point cloud data into rasterized layers based on classes in which each rasterized layer in the rasterized layers corresponds to a class in the classes. The number of processor units creates key value pairs from the rasterized layers. The number of processor units store the key value pairs in a key value store. According to other illustrative embodiments, a computer system and a computer program product for rasterizing point cloud data are provided.

Abstract: An environment is captured with a set of sensors to generate a thermal image from an infrared sensor and a visual image from a visual light camera. The thermal image is used to predict movement of objects detected in the environment. The thermal image may be combined with the visual image as a channel of data with the visual image as an input to a prediction model that predicts object movement. Alternatively, the visual image may be used as a guide to identify relevant portions of the thermal image. Objects may be detected and segmented in the visual image and the corresponding portions of the thermal image are segmented and used to predict thermal characteristics for the object. The thermal characteristics may then be used for object movement prediction.

Abstract: To provide control for an autonomous vehicle based on a physical object near an autonomous vehicle (AV), the AV monitors its environment with an imaging sensor, such as an infrared sensor or visible light camera. The imaging sensor captures a sensor view of the environment and an encoded signal is detected in the sensor view. The encoded signal may be emitted by a mobile device with a respective emitter for the imaging sensor, such as an IR emitter or a visual light source. After detecting the encoded signal, the AV identifies the location of the mobile device within the sensor view and the environment. Then, based on the determined location in the environment, the AV performs a control action, such as navigating to a stopping place near the location.

Abstract: Object detection for ranging sensor data may detect objects which are not actually present in the environment. To identify certain of these phantom objects, objects detected in the environment are analyzed to determine whether they are enclosed by another object and if the enclosed object has a distance from the ranging sensor higher than the enclosing object. This may suggest that the enclosing object has a surface or other feature that is sensed as additional depth that manifests as a separate detectable object. These phantom objects are identified and removed from further perception processing.

Abstract: In an approach to improve detecting and identifying objects through orbital synthetic aperture radar satellites, embodiments arrange an array of elements in a predetermined configuration, and process, by a threshold and signature analysis, detected peaks in processed image data. Further, embodiments generate a list of objects detections based on the processed peaks, and identify an object based on amplitude, polarization ration, and polarization phase difference. Additionally, embodiments, classify the identified object based on the generated list of objects, and output, by a user interface, a list of probable object detections with position coordinates and identifications based on the classified identified objects, wherein the list of probable objects are above or within a predetermine threshold of confidence.

Abstract: Machine learning model optimization systems and methods are disclosed. A system receives sensor data captured by one or more sensors of a vehicle during a first time period. The vehicle uses a first trained machine learning (ML) model for one or more decisions of a first decision type during the first time period. The system generates a second trained ML model at least in part by using the sensor data to train the second trained ML model. The system identifies an optimal trained ML model from a plurality of trained ML models. The plurality of trained ML models includes the first trained ML model and the second trained ML model. The system causes the vehicle to use the optimal trained ML model for one or more further decisions of the first decision type during a second time period after the first time period.

Abstract: The present disclosure is directed to collecting and processing data from computing devices of a plurality of autonomous vehicles (AVs). The data received from each of these AV computing devices may include raw sensor data or data that has been generated using data received by one or more sensors at respective AVs. Once this data is collected and associated with discrete locations and times, the data may be evaluated and used to generate mappings of various sorts. These mappings may include mappings of underground features generated based on an evaluations of vibration data. Alternatively, or additionally, these mapping may include mappings of landscape features, atmospheric features, or the locations of aircraft from data associated with certain types of sensing apparatus, for example radar apparatus or light detecting and ranging (LiDAR) apparatus.

Abstract: The present disclosure is directed to checking and updating the calibration of a sensing apparatus at an autonomous vehicle (AV). After a sensing apparatus of an AV has been initially calibrated (e.g., when the AV is manufactured) calibration of that sensing apparatus may be checked or updated using data collected from similar vehicles of a fleet of vehicles. Each of the different vehicles of this fleet may identity its location and report that location. A first of these vehicles may receive data that identifies the location of a second vehicle. A sensing apparatus of the first vehicle may then identify locations of features at the second vehicle and a processor of the sensing apparatus may perform calculations to identify whether calibration of the sensing apparatus has changed. Parameters that adjust the calibration of the sensing apparatus may then be updated to maintain calibration of a sensing apparatus within acceptable tolerances.

Abstract: A computer implemented method rasterizes point cloud data. A number of processor units rasterizes the point cloud data into rasterized layers based on classes in which each rasterized layer in the rasterized layers corresponds to a class in the classes. The number of processor units creates key value pairs from the rasterized layers. The number of processor units store the key value pairs in a key value store. According to other illustrative embodiments, a computer system and a computer program product for rasterizing point cloud data are provided.

Abstract: In some examples, a method of vector-raster data fusion includes receiving vector data for a geographical location, and statistically analyzing the vector data to obtain vector statistics. In some examples the method further includes rasterizing the vector statistics, and storing at least one of the vector data and the rasterized vector statistics together in a key-value store together with previously stored raster data for the geographical location. In some examples, the vector data further includes metadata, and the method further includes storing the metadata in at least one of the key-value store or a separate vector database.

Abstract: A computer produces predictions throughout a raster field in response to point data, by obtaining a partially empty matrix of point data, filling a matrix of extrapolated raster data by dilating the point data in a first convolutional neural network, and generating a matrix of aggregate raster data by combining the extrapolated raster data with organic raster data in a second convolutional neural network.

Abstract: Techniques for image processing and transformation are provided. A plurality of images and a plurality of maps are received, and a system of neural networks is trained based on the plurality of images and the plurality of maps. A first image is received, and a first map is generated by processing the first image using the system of neural networks.

Abstract: A computer-implemented method for determining field boundaries and crop forecasts in each field is provided. The method includes deriving vegetation indices for each geo-spatial pixel of each image of multi-spectral imagery at a plurality of points in time, constructing minimum bounding boxes for each image according to the vegetation indices, and generating, based on a neural network analysis of each image and the minimum bounding boxes, a geo-spatial plot of crops including a predicted plot of future crop usage for an area including each field in the multi-spectral imagery.

Abstract: Methods and systems for generating a composite image in remote sensing applications are described. In an example, a device can receive an image having a plurality of points specified in an image space. The device can extract a portion of the image and transform points among the extracted portion from the image space to a parameter space defined by a distance parameter and an orientation parameter. The device can identify a set of intersection points in the parameter space that indicate at least one occurrence of a geometry feature in the extracted portion of the image. The device can augment the portion of the image with a plurality of new pixels based on the identified set of intersection points. The device can generate a composite image using the augmented image, where the composite image can include a plurality of augmented images corresponding to other portions of the image.

Abstract: A charging device includes a passive auxiliary circuit and a rectifier which is connected downstream of the auxiliary circuit. The passive auxiliary circuit includes input nodes and output nodes. Between the input node and the output nodes, two impedances are connected. Here, an imaginary component of the first impedance has a positive non-zero value and an imaginary component of the second impedance a negative non-zero value or vice versa.

Abstract: A computer-implemented method, a computing system, and a computer program product, for automatically labeling an amount of unlabeled data for training one or more classifiers of a machine learning system. A method includes iteratively processing unlabeled data items. Receiving an unlabeled data item into each autoencoder in an autoencoder architecture. Each autoencoder processing with a lowest loss of information the unlabeled data item that is likely associated with a label associated with the autoencoder, while processing with a higher loss of information the unlabeled data item that is likely not associated with the label. Predicting, based on loss of information, a probability distribution for the unlabeled data item. Automatically associating the label to the unlabeled data item, based on the label being associated with a highest probability in a peaking probability distribution associated with the unlabeled data item. The autoencoder architecture can include a cloud computing network architecture.

Abstract: Methods and systems for managing vegetation include training a machine learning model based on an image of a training data region before a weather event, an image of the training data region after the weather event, and information regarding the weather event. A risk score is generated for a second region using the trained machine learning model based on an image of the second region and predicted weather information for the second region. The risk score is determined to indicate high-risk vegetation in the second region. A corrective action is performed to reduce the risk of vegetation in the second region.

Abstract: From each of a plurality of cameras, a visual input of a location is received over a network. For each visual input from the plurality of cameras, a coupling correction is performed between a shaking of the camera with respect to the visual input by subtracting velocity vectors of the plurality of cameras from velocity vectors of pixels defining the visual input to provide a processed input. It is determined whether a shaking identified in the processed input is above a predetermined threshold based on the processed input, thereby detecting one or more anomalies. From the one or more anomalies, at least one of a location, magnitude, or depth of an earthquake are inferred based on the shaking identified in the processed input of each of the plurality of cameras.

Abstract: A method of generating an aggregate forecast includes obtaining historical forecasts for a number of time steps and at least one location, obtaining historical conditions for the time steps and the at least one location, training a machine learning algorithm to produce an aggregate historical forecast in response to the historical conditions and the historical forecasts, and producing an aggregate current forecast by running the trained machine learning algorithm on current forecasts. The historical forecasts and the current forecasts vary in at least one of spatial resolution or temporal resolution, and include a first forecast that is valid for a first time step and a second forecast that is valid for a second time step.