Abstract: A system for navigating a host vehicle relative to a road segment may receive a captured image acquired by a camera onboard the host vehicle, wherein the captured image includes representations of at least a first object and a second object present in an environment of the host vehicle; provide the captured image to a trained model configured to receive the captured image as input, apply large language model functionality to establish a semantic constraint between the first object and the second object, and generate an output that identifies the semantic constraint between the first object and the second object; determine, based on the semantic constraint between the first object and the second object, at least one navigational action for the host vehicle; and cause the host vehicle to implement the at least one navigational action.

Abstract: A system may receive, from a map associated with a road segment, at least one indicator of a lane of travel associated with the road segment, location information associated with the lane of travel, an identifier of a road element associated with the road segment, and location information associated with the road element; generate a graph representation of a junction associated with the road segment including a plurality of nodes and one or more semantic links; provide the graph representation to a trained model configured to infer a relationship between the road element and the lane of travel; based on the output of the trained model, store in the map road element relationship; and distribute the map for use in navigating along the road segment.

Abstract: A Printed Circuit Board (PCB) may include one or more single-ended PCB traces configured to route single-ended Radio-Frequency (RF) signals between an integrated circuit and one or more waveguides, wherein first ends of the one or more single-ended PCB traces are to be coupled to the integrated circuit; and one or more PCB-to-waveguide transitions configured to couple second ends of the one or more single-ended PCB traces to the one or more waveguides. For example, a PCB-to-waveguide transition may include a PCB probe connected to a second end of a single-ended PCB trace of the one or more single-ended PCB traces, the PCB probe configured to couple RF energy of the single-ended RF signals between the single-ended PCB trace and a waveguide of the one or more waveguides; and a via configured to electrically connect the PCB probe to a ground layer of the PCB.

Abstract: Systems and methods may generate a map for use in navigating a host vehicle relative to a road segment by receiving drive information from a vehicle that previously traversed the road segment, providing a representation of at least one road topography indicator to a trained model, wherein the trained model is configured to provide a driving protocol indicator as an output; generating, based on the driving protocol indicator, a representation of at least one driving convention associated with the driving protocol indicator and storing the representation of the at least one driving convention in the map; and distributing the map to at least one host vehicle navigation system for use in navigating the host vehicle along the road segment relative to the stored representation of the at least one driving convention.

Abstract: A system for navigating a host vehicle relative to a road segment may store a speed limit indicator in a map among navigation information associated with the road segment. The system may determine a first navigational action for the host vehicle based on the received speed limit indicator; cause at least one system associated with the host vehicle to implement the first navigational action; receive an image acquired by at least one camera onboard the host vehicle; determine a speed limit affecting condition represented in the received image; determine a condition-based speed limit based on the speed limit affecting condition; determine a second navigational action for the host vehicle based on the condition-based speed limit; and cause the at least one system associated with the host vehicle to implement the second navigational action.

Abstract: System and methods are disclosed for navigating a host vehicle. In one implementation, at least one processing device may be programmed to receive a plurality of images representative of an environment of the host vehicle, identify, in at least one of the plurality of images a representation of, a target vehicle, determine a separation distance between the host vehicle and the target vehicle, if the separation distance is less than a comfort distance threshold, cause a reduction in a throttle level associated with a throttle system of the host vehicle without activating a braking system associated with the host vehicle, and if the separation distance is determined to be greater than but trending toward a safe distance threshold, which is less than the comfort distance threshold, cause activation of the braking system associated with the host vehicle.

Abstract: Systems and methods are provided for navigating a host vehicle.

Abstract: A host vehicle-based feature harvester is disclosed. In one implementation, the feature harvester includes memory and a processor configured to receive a plurality of images captured by a camera onboard the host vehicle, the plurality of images being representative of an environment of the host vehicle; analyze at least one image from the plurality of images to identify a representation of a lane mark; select at least one sample area of the representation of the lane mark, wherein the at least one sample area is associated with an image location of at least a portion of the representation of lane mark; determine a location identifier of the at least one sample area; determine a surface quality indicator associated with the at least one sample area; and cause transmission of the location identifier and the surface quality indicator to an entity remotely-located relative to the host vehicle.

Abstract: Techniques are disclosed for reducing false positives for generating warnings to avoid potential collisions between a vehicle and vulnerable road users (VRUs). This is accomplished via an onboard vehicle safety system that uses crowdsourced map data to determine whether a vehicle is capable of performing a maneuver that results in a lateral shift of the vehicle (which may include a lane-shifting or turning maneuver) within a predetermined threshold time period. The ability for the vehicle to make the turning maneuver, among other driving scenarios, may be used to by the safety system to intelligently determine whether a warning or other action is needed to avoid a potential collision with a VRU. In this way, the occurrence and number of false warnings/interventions are minimized or at least reduced, leading to more attentive drivers and thereby improving VRU safety.

Abstract: Techniques are disclosed to enable the generation of ground truth datasets that identify a continuous time movement trajectory of dynamic objects such as non-stationary vehicles. The techniques as described herein may utilize iterative closest point (ICP) processes to compute a rigid-body transform that describes a 3D correspondence between different pairs of temporal point cloud data sample sets at various discrete time periods that are identified with an object along a movement trajectory. From these rigid-body transforms at each of the discrete time periods, a pose graph optimization may be performed to compute the continuous time movement trajectory. The ground truth datasets may be used to enable training of machine learning models, which may be used by a vehicle computing system when navigating a driving environment.

Abstract: For example, a polarized antenna may include a dielectric layer including an antenna via; a first metal layer on a first side of the dielectric layer, the first metal layer including an antenna feeding trace connected to a first end of the antenna via; and a second metal layer on a second side of the dielectric layer. For example, the second metal layer may include a plurality of antenna elements including a first antenna element, a second antenna element, and a third antenna element. For example, the third antenna element may be connected to a second end of the antenna via. For example, the second metal layer may include a plurality of antenna traces serially connecting the plurality of antenna elements.

Abstract: The disclosed technique is an any-in, first-out (“AIFO”) memory device that improves upon conventional first-in, first-out memory devices. The technique mimics a first-in, first-out queue to provide sequential storage and retrieval in scenarios where transactional ordering cannot be guaranteed. By associating an identifier and a write-pointer in a look-up table with read requests prior to transmission, responses may be received out of order and/or interleaved without delaying processing until the next response in the sequence is received. An internal validity flag may be included in the AIFO memory array to represent the validity of each line in the memory array. Once a read pointer reaches a row that is marked as invalid, the read command pauses and discontinues further reading until a write to the row that marks the row as valid. To further optimize writing, a polarity of the validity may be toggled each read cycle.

Abstract: System and techniques for vehicle environment modeling with a camera are described herein. A time-ordered sequence of images representative of a road surface may be obtained. An image from this sequence is a current image. A data set may then be provided to an artificial neural network (ANN) to produce a three-dimensional structure of a scene. Here, the data set includes a portion of the sequence of images that includes the current image, motion of the sensor from which the images were obtained, and an epipole. The road surface is then modeled using the three-dimensional structure of the scene.

Abstract: A navigation system may receive a captured image acquired by a camera onboard a host vehicle; determine a localized position of the host vehicle relative to a map representative of the road segment; generate based on the localized position of the host vehicle and three-dimensional location information associated with the plurality of topography features, a two-dimensional image representation of a plurality of road topography features; analyze the captured image and the two-dimensional image representation and determine three-dimensional information for at least one object represented in the captured image, but not represented in the map representative of the road segment, based on the three-dimensional location information associated with the plurality of topography features included in the map; determine, based on the three-dimensional information for the at least one object, at least one navigational action for the host vehicle; and cause the host vehicle to implement the at least one navigational action.

Abstract: A system may receive drive information from each of a plurality of harvesting vehicles that traversed a road segment, wherein the drive information includes at least one indicator of an actual trajectory traveled by the harvesting vehicle and altitude information associated with a road surface included among a plurality of road surfaces associated with the road segment, wherein each of the plurality of road surfaces is located at a different altitude. The system may aggregate the drive information to determine at least one host vehicle drivable path, store a representation of the drivable path, generate a road surface representation for each road surface, store the road surface representations, assign a detected road feature to a road surface representation, and distribute the map to at least one host vehicle navigation system for use in navigating the host vehicle along the road segment relative to the host vehicle drivable path.

Abstract: Systems and methods for navigating a host vehicle are disclosed. In one implementation, a system includes a processor configured to receive an image captured by a camera onboard the host vehicle; provide the image to a trained model configured to generate an output identifying two or more target trajectories associated with each of the two or more features represented in the image; determine, based on the output generated by the trained model, location information for the two or more target trajectories associated with each of the two or more features; determine at least one navigational action for the host vehicle based on the location information determined for at least one of the target trajectories; and cause the host vehicle to implement the at least one navigational action.

Abstract: Systems and methods are provided for generating a crowd-sourced map for use in vehicle navigation. In one implementation, a system may include at least one processor configured to receive drive information collected from vehicles that traversed a junction; aggregate the received drive information to determine positions of traffic lights and spline representations for drivable paths; input the determined positions and the spline representations to a trained model configured to generate a traffic light relevancy mapping indicating a traffic light relevancy for traffic light to drivable path pairs of the junction; input an observed vehicle behavior to the at least one trained model to generate an updated traffic light relevancy mapping; store in the crowd-sourced map the indicators of traffic light relevancy for the traffic light to drivable path pairs; and transmit the crowd-sourced map to a vehicle for use in navigating the road segment.

Abstract: A multi-core processor configured to improve processing performance in certain computing contexts is provided. The multi-core processor includes multiple processing cores that implement barrel threading to execute multiple instruction threads in parallel while ensuring that the effects of an idle instruction or thread upon the performance of the processor is minimized. The multiple cores can also share a common data cache, thereby minimizing the need for expensive and complex mechanisms to mitigate inter-cache coherency issues. The barrel-threading can minimize the latency impacts associated with a shared data cache. In some examples, the multi-core processor can also include a serial processor configured to execute single threaded programming code that may not yield satisfactory performance in a processing environment that employs barrel threading.

Abstract: Systems and methods are disclosed for navigating a host vehicle. In one implementation, at least one processor may be programmed to receive an image captured by a camera; identify an oncoming target vehicle; determine that a planned trajectory for the host vehicle includes a turn that crosses a projected trajectory of the oncoming target vehicle in a potential turn-across-path event; determine a driving jurisdiction associated with the road; determine, based on the driving jurisdiction, whether the target vehicle is approaching a road feature that negates the potential turn-across-path event; and either implement or forego implementing a remedial action based on whether the target vehicle is approaching a road feature that negates the potential turn-across-path event.

Abstract: Provided herein are system, apparatus, device, method and/or computer program product embodiments, and/or combinations and sub-combinations thereof, for predicting a driver path of a vehicle. In the method, an image of a driver of the vehicle is received. Based on the image, a gaze direction of the driver is determined. Based at least in part on the gaze direction, a path of the vehicle is predicted. The vehicle is controlled based on the predicted path.