Abstract: Automatic generation of optimized aggregation metrics for usage-based insurance (UBI) includes performing a high-dimensional Bayesian optimization on a data archive, the data archive including a plurality of signals from vehicles and corresponding UBI effects, wherein the high-dimensional Bayesian optimization includes performing testing on weighted groups of the plurality of signals using one or more aggregation functions; and transmitting the one or more aggregation functions to a vehicle to cause the vehicle to provide aggregated signals for input to a UBI model to predict a UBI rate for the vehicle.

Abstract: Systems and methods for mitigating certain spoofing of vehicle features are disclosed herein. An example method can include determining input torque values obtained from a steering torque sensor associated with a steering wheel of a vehicle, wherein the input torque values are obtained over a period of time, determining road disturbances using a road disturbance model, applying a driver model that is indicative of human driver hands-on-wheel behaviors, determining when input torque values are indicative of a spoof or human interaction with the steering wheel using the input torque values, the road disturbance model, and the driver model, and executing a remediating measure when the input torque values are indicative of the spoof.

Abstract: A shelving system has a plurality of support posts and one or more shelf assemblies, each shelf assembly having a height adjustment mechanism. A height adjustment mechanism can include a bracket attached to a shelf portion, wherein the bracket includes a channel for receiving a support post, and a latch element pivotably attached to the bracket. The latch element pivots between a first position in which the latch element engages an aperture in the support post to support the shelf assembly on the support post, and a second position in which the latch element is out of the aperture to allow the shelf assembly to move along the support post. The latch element can include a hook with a notch positionable to receive an edge of the aperture, and a side face positionable to rest or press against the support post.

Abstract: A computer that includes a processor and a memory, the memory including instructions executable by the processor to determine a first prediction with a machine learning system based on receiving a first image from a first camera and determine a second prediction with a machine learning system based on receiving a second image from a second camera. When the first prediction does not equal the second prediction within a user determined tolerance, determine color consistency based on comparing pixel values from the first image with a threshold determined based on previously determined pixel values, determine color correction parameters by determining pixel statistics based on pixel values from the first image to include in an image signal processing system and apply the color correction parameters to a third image from the first camera by receiving the third image at the image signal processing system.

Abstract: A computer includes a processor and a memory, and the memory stores instructions executable by the processor to receive operation data generated by first components of a vehicle; determine that the vehicle is being operated autonomously by an on-board autonomous-operation module based on the operation data without using output from the on-board autonomous-operation module; and upon determining that the vehicle is being operated autonomously, actuate a second component of the vehicle.

Abstract: A computer includes a processor and a memory, and the memory stores instructions executable by the processor to sequentially activate a plurality of lamps aimed at a scene, the lamps mounted to a vehicle; receive image data of the scene generated while sequentially activating the lamps; generate a map of surface normals of the scene by executing a photometric stereo algorithm on the image data; and in response to a speed of the vehicle being below a speed threshold, navigate the vehicle based on the map of the surface normals.

Abstract: A computer-implemented method of predicting dynamics of objects in a surrounding of a vehicle is disclosed. The method starts with a step of receiving a first data sets characterizing dynamics of the objects respectively. Then, each of the first data sets is propagated through an encoder outputting a latent representation for each of the first data sets. Then, a graph based on the latent representations is generated. Then, the graph is propagated through a Graph Neural Network outputting an updated graph. Based on the updated graph a decoder outputs a predicted dynamic for selected object for a subsequent time step.

Abstract: A vehicle including a steering wheel, a first detection unit, a second detection unit and a processor is disclosed. The first detection unit may be configured to detect a first parameter associated with the steering wheel, and the second detection unit may be configured to capture an input associated with a vehicle occupant. The processor may be configured to obtain the input and the first parameter, and estimate a second parameter associated with a vehicle occupant interaction with the steering wheel based on the input. The processor may be further configured to correlate the first parameter and the second parameter, and determine that a predefined condition may be met based on the correlation. The processor may further transmit a notification responsive to a determination that the predefined condition may be met.

Abstract: A computer includes a processor and a memory, and the memory stores instructions executable by the processor to identify personally identifiable information (PII) in an initial image frame; encode a region containing the PII of the initial image frame to a latent vector with a lower dimensionality than the region; anonymize the region from the initial image frame, resulting in a modified image frame; and associate the latent vector with the modified image frame.

Abstract: A sock for improving comfort for a user is provided. The sock includes a foot area having a closed end, an open end opposite the closed end, and a notch extending from an edge of the open end of the sock toward the closed end. When a part of a user's body is inserted into the sock, the notch is configured to expand from a first state, in which edges of the notch are at a first angle with respect to each other, to a second state, in which the edges of the notch are at a second angle with respect to each other, wherein the second angle is greater than the first angle.

Abstract: Ensuring privacy consent for handling of occupant vehicle data is provided. A feature identification vector indicative of an identity of a vehicle occupant of a vehicle is identified. The feature identification vector is used to identify whether consent for use of vehicle data was provided by the vehicle occupant. The consent is requested responsive to the identity of the vehicle occupant not having consented to data collection. Responsive to the consent being given by the vehicle occupant, the consent and the feature identification vector of the vehicle occupant is stored in a storage of the vehicle. The vehicle data is uploaded in accordance with whether the consent was granted for the vehicle occupant.

Abstract: A computer includes a processor and a memory storing instructions executable by the processor to predict a quantity of misalignment of an optical sensor based on a projected motion of a vehicle, predict an error of the predicted quantity of misalignment, and actuate the vehicle based on the predicted quantity of misalignment and the predicted error. The vehicle includes the optical sensor.

Abstract: Captured PII is identified in sensor data, the sensor data being captured over time by a sensor of a vehicle of an environment surrounding the vehicle. The sensor data includes captured PII and non-PII. A notice zone surrounding the vehicle is identified, the notice zone defining an area in which a notice device of the vehicle provides notice to pedestrians that the sensor data is being captured. Any instances of the captured PII in the sensor data are obscured that are captured from the pedestrians that, over the time of the capture, did not enter the notice zone and receive the notice. The sensor data, as processed, is sent to a cloud server for storage.

Abstract: A computer includes a processor and a memory, and the memory stores instructions executable by the processor to determine a value over time derived from a region of sensor data from an optical sensor of a host vehicle; determine that the value over time matches a stored temporal pattern; upon determining that the value over time matches the stored temporal pattern, classify an object at the region as a surface reflecting a signal from an illuminator of an origin vehicle; and actuate a component of the host vehicle based on a classification of the object at the region.

Abstract: Analytical methods and systems applied to sequential event data are disclosed. An exemplary system and method analyzes datasets containing events in a plurality of journeys. The methods and systems described analyze and quantify the relative importance of events and sequences leading to outcomes where the data is complex and interconnected. In some embodiments, a graphical user interface illustrates the quantification of these datasets. In some embodiments, the graphical user interface maps the journey paths to show the relative importance of each journey path. In some embodiments, the maps of journey paths are interactive, allowing selection of paths of interest for detailed analysis. In some embodiments, the methods and systems calculate paths similar to a journey path of interest. An exemplary method and system also provides detailed recommendations for changing events within a sequence to either increase or decrease the likelihood of achieving a selected outcome.

Abstract: A computer is programmed to receive a plurality of first values for a respective plurality of variables transmitted over a communications network of a vehicle; determine a similarity measure between the first values and a plurality of second values of the variables; determine whether to transmit a collection of data transmitted over the communications network based on the similarity measure; and, upon so determining, transmit the collection of data to a server remote from the vehicle. The second values are a most similar set of values from an activation surface for a feature of the vehicle. The activation surface is a surface in a space defined by the variables dividing the space into regions in which activation of the feature is more or less likely than not, respectively, to have occurred when the variables have values in that region.

Abstract: The disclosure is generally directed to systems and methods associated with communicating privacy rights pertaining to data captured from a vehicle. An example method executed by a processor of a data capture apparatus in a vehicle can include capturing an image. In an example scenario, the image may include an individual who is located outside the vehicle. Furthermore, in some cases, the image can be a part of a video clip containing multiple images. The method further includes generating an identifier for identifying the image(s). The identifier can be a machine-readable symbol such as, for example, a QR-code symbol or a barcode. A notification that includes the identifier may be generated and conveyed to the individual for use by the individual to exercise his/her rights to privacy with respect to the images in which he/she is present.

Abstract: A method for configuring a neural network which is designed to map measured data to one or more output variables. The method includes: transformation(s) of the measured data is/are specified which when applied to the measured data, is/are meant to induce the output variables supplied by the neural network to exhibit an invariant or equivariant behavior; at least one equation is set up which links a condition that the desired invariance or equivariance be given with the architecture of the neural network; by solving the at least one equation a feature is obtained that characterizes the desired architecture and/or a distribution of weights of the neural network in at least one location of this architecture; a neural network is configured in such a way that its architecture and/or its distribution of weights in at least one location of this architecture has/have all of the features ascertained in this way.

Abstract: A system for detecting a road surface includes a processor programmed to identify, in a vehicle sensor field of view, environment features including a sky, a road surface, a road shoulder, based on map data and data received from the one or more vehicle sensors, upon determining a low confidence in identifying an area of the field of view to be a road surface or sky, to receive a polarimetric image, and to determine, based on the received polarimetric image, whether the identified area is a mirage phenomenon, a wet road surface, or the sky. The low confidence is determined upon determining that a road surface has a vanishing edge, a road lane marker is missing, or an anomalous object is present.

Abstract: Provided is a method for detecting the trajectories of one or more targets in the field of view of one or more sensors, the method comprising: receiving one or more sensor frames corresponding to the one or more sensors; defining a space of allowable target states for the one or more sensor frames; specifying a set of potential target trajectories, each comprising one allowable target state for each of the one or more sensor frames; specifying target signal parameters for each of the allowable target states, such that the target signal parameters predict the expected target signal contribution corresponding to the one or more sensor frames; specifying a data fidelity objective to quantify how well the target signal parameters match the one or more sensor frames; specifying a sequence of one or more sparsity objectives to penalize a number of detected targets; determine the trajectories of one or more targets as follows: obtain values for all the target signal parameters in all the sensor frames, the obtained