Abstract: Systems, methods, and apparatuses of a welding system are disclosed and include a first stage of a scanning device for scanning weld parts to generate a three-dimensional (3D) profile of a weld target wherein the 3D profile captures matching imperfections of a meeting together of the set of weld parts when performing the weld operation; and the second stage of a monitoring device to monitor the weld operation and to generate a data of high-resolution measurements of the weld operation; wherein the first stage further includes the monitoring device determining a weld schedule based on the 3D profile, and to adjust the weld schedule while the weld operation progresses to adapt to predicted distortion based on the 3D profile and to sensed distortion; wherein the second stage further includes a plurality of sensors to sense a set of components associated with the weld operation to generate high-resolution data of measurements.

Abstract: A method includes: selectively actuating switches of an inverter module and first, second, third, and fourth switches of a motor; and charging a battery with power from a charger using at least one of first, second, and third stator windings of the motor, the battery having a first voltage and being electrically connected in parallel with the inverter module, and the charger having a second voltage that is less than the first voltage.

Abstract: An example automated driving system computer can include a board, one or more processors coupled to the board, a first layer of a first thermal interface material applied on the one or more processors, a heat spreader having a first side and a second side, the first side in contact with the first layer of the first thermal interface material, a second layer of a second thermal interface material applied on the second side of the heat spreader, and a cold plate in contact with the second layer of the second thermal interface material.

Abstract: Methods and system for training a neural network for depth estimation in a vehicle. The methods and systems receive respective training image data from at least two cameras. Fields of view of adjacent cameras of the at least two cameras partially overlap. The respective training image data is processed through a neural network providing depth data and semantic segmentation data as outputs. The neural network is trained based on a loss function. The loss function combines a plurality of loss terms including at least a semantic segmentation loss term and a panoramic loss term. The panoramic loss term includes a similarity measure regarding overlapping image patches of the respective image data that each correspond to a region of overlapping fields of view of the adjacent cameras. The semantic segmentation loss term quantifies a difference between ground truth semantic segmentation data and the semantic segmentation data output from the neural network.

Abstract: A multimedia system is provided for a scalable infotainment system of a motor vehicle. The scalable infotainment system includes a camera, which is releasably attached to the motor vehicle and supports camera software components for generating a media signal. The multimedia system includes a media player and a computer attached to the motor vehicle. The computer has a processor and a non-transitory computer readable storage medium, which stores player software components including multiple streaming protocols used by the media player to play a media stream. The processor is programmed to transmit to the camera a player setup signal including data associated with the player software components. The processor is further programmed to receive from the camera a media signal associated with a media stream generated by the camera by using a common one of the streaming protocols supported by the camera and the media player.

Abstract: In an exemplary embodiment, a vehicle system is provided that includes a sensor array and a processor. The sensor array is configured to generate vehicle sensor data pertaining to a target vehicle in proximity to the vehicle. The processor is coupled to the sensor array, and is configured to at least facilitate: beginning an overtake maneuver for the vehicle, via instructions provided to the drive system, when conditions for an overtake of the target vehicle by the vehicle are initially determined by the processor to be valid; subsequent to the beginning of the overtake maneuver, determining via the processor whether the conditions for the overtake of the target vehicle are still valid, based on updated sensor data reflecting variant traffic conditions; and further controlling the overtake maneuver based on the determining as to whether the conditions for the overtake of the target vehicle are still valid.

Abstract: A reconfigurable connector system for distributing power between an input power source and a cable harness connector having M cable harness terminals includes an energy center module (ECM) and a connecting adapter. The ECM has a module housing, a PCB having more than M pin holes, a plurality of e-fuses connected with the pin holes, a controller connected with the e-fuses, and an input connector connected to the controller and/or the e-fuses. The connecting adapter has a body with first and second sides and M bridging adapters within the body. Each bridging adapter includes a first connection point accessible from the first side and configured to be directly connectable with a cable harness terminal, one or more second connection points accessible from the second side and configured to be connectable with a pin hole, and a busbar connecting the first connection point with the one or more second connection points.

Abstract: An electronic control unit (ECU), or node, is configured to use a single key for generating requests from a security peripheral for a MAC. The security peripheral includes the stored shared key. The security peripheral may further include a policy enabling it to detect if a request from the V-ECU is valid, in which case it generates a MAC. The security peripheral is also used to store information in a MAC Generate Allow List (MGAL). In some embodiments, the receiving nodes in a network receive data based on a security peripheral's response to a transmit nodes requests for a MAC. The receiving nodes use this knowledge to avoid being spoofed.

Abstract: Protective sleeves, kits-of-parts for manufacturing sleeves, and methods for manufacturing sleeves are provided. An exemplary protective sleeve includes an inlet end connector; intermediate linking connectors; and an outlet end connector. The intermediate linking connectors are interconnected in a series from the inlet end connector to the outlet end connector; and the inlet end connector, the intermediate linking connectors, and the outlet end connector comprise fire-resistant material.

Abstract: Methods and systems are provided for determining and mitigating vehicle pull force from passing other vehicles.

Abstract: A method and system for distributed publisher-subscriber event communication. The distributed publisher-subscriber event communication may include determining a first event published with a first service operating on a first device, determining a second device subscribing to the first event, unicasting the first event from the first device to the second device, determining a plurality of second applications operating on the second device subscribing to the first event, and multicasting the first event received at the second device to each of the second applications.

Abstract: An electronic control unit (ECU), or node, is configured to use a single key for all virtual ECUs (V-ECUs) having a message to transmit. The ECU also may include a security peripheral that includes the stored shared key. The security peripheral may further include a policy enabling it to detect if a request from the V-ECU is valid, in which case it generates a MAC. The security peripheral is also used to store information in a MAC Generate Allow List (MGAL), which may define the policy about when the V-ECU can transmit. The ECU may include a crypto-engine, which stores shared keys from other nodes used to verify MACs transmitted by other nodes.

Abstract: In exemplary embodiments, methods and systems are provided that include or utilize one or more automotive equipped cameras and a processor. The one or more automotive equipped cameras are configured to be onboard a vehicle and to obtain camera images during operation of the vehicle, the camera images including a view of one or more wheels of the vehicle. The processor is coupled to the one or more cameras, and is configured to at least facilitate: processing the camera images; and determining one or more states of the one or more wheels, including an alignment thereof, a health thereof, or both, based on the processing of the camera images.

Abstract: A method of reforming a negative electrode layer of a secondary lithium battery may include execution of a reforming cycle that reforms a major facing surface of the negative electrode layer by eliminating at least a portion of a lithium dendrite or other lithium-containing surface irregularity that has formed on the major facing surface of the negative electrode layer.

Abstract: A clearance risk detection system is provided in a vehicle. The clearance risk detection system is configured to: identify a terrain object in a vehicle travel path; classify the terrain object based on whether the host vehicle can clear the terrain object or cannot clear the terrain object; provide a first indicator that indicates that the host vehicle can clear the terrain object when appropriate; provide a second indicator that indicates that the host vehicle can clear the terrain object with a suggested raising of vehicle ride height when appropriate; automatically raise the vehicle ride height when appropriate; and provide a third indicator that indicates that the host vehicle cannot clear the terrain object even with raising vehicle ride height when appropriate.

Abstract: The subject disclosure relates to techniques for improving performance of a machine learning algorithm that at least receives data descriptive of a 3-D object and provides an output, where the 3-D object occurs infrequently in a training dataset. A process of the disclosed technology can include determining that the machine learning algorithm performed below a threshold performance score when receiving data descriptive of the 3-D object in a real-world scene, wherein the 3-D object is classified as a first type of object, creating at least one 3-D representation of the first type of object for use in a simulation, modifying a plurality of simulated scenes to include the at least one 3-D representation of the first type of object, and training the machine learning algorithm with the modified simulated scenes, whereby the machine learning algorithm has greater exposure to the first type of object.

Abstract: The subject disclosure relates to features for improving wheelchair accessibility in autonomous vehicles (AVs) and in particular, for enabling automatic ingress/egress of a wheelchair ramp to facilitate the loading and unloading of a passenger wheelchair. In some aspects, a process of the disclosed technology includes steps for identifying one or more visual reference features on at least one surface of an autonomous vehicle (AV), automatically deploying a ramp to facilitate ingress of a wheelchair, and tracking ingress of the wheelchair based on the one or more visual reference features. Systems and machine-readable media are also provided.