Abstract: A system for determining a lane change decision based on a downstream traffic state can include a processor, a communications device, and a memory. The processor can be disposed on an ego vehicle. The memory can store an acceleration gain module and a communications module. The acceleration gain module can include instructions that cause the processor to: (1) obtain, via the communications device and from a monitoring system, information about an end of stopped or slow moving downstream traffic in a lane and (2) calculate a result of an equation for the lane change decision. The equation can include information about virtual vehicles positioned near the end or a geographical location. The communications module can include instructions that cause the processor to cause a signal, with information based on the result, to be sent to a component of the ego vehicle for an action to be performed by the component.

Abstract: The disclosure describes a method for an ego vehicle. The method includes determining that a vehicle needs to park. The method further includes identifying a parking facility within proximity to the vehicle and a corresponding congestion level of the parking facility. The method further includes selecting a first parking model from two or more parking models based on the corresponding congestion level being closer to a first congestion range than a second congestion range. The method further includes identifying a parking spot for the vehicle within the parking facility based on the first parking model.

Abstract: The disclosure includes embodiments for early detection of abnormal driving behavior. A method according to some embodiments includes sensing, by a sensor set of an ego vehicle, a remote vehicle to generate sensor data describing driving behavior of the remote vehicle. The method includes comparing the sensor data to a set of criteria for abnormal driving behavior. The method includes determining that a subset of the set of criteria are described by the sensor data. The subset satisfies a threshold for early detection of abnormal driving behavior. For example, detecting the subset triggers the determination that abnormal driving behavior is detected. The method includes determining that the remote vehicle is engaged in abnormal driving behavior based on satisfaction of threshold. In some embodiments, an abnormal driving behavior is one which satisfies a threshold for abnormality or matches a pattern of an identified abnormal driving behavior.

Abstract: System, methods, and other embodiments detecting and localizing objects within an image in a wide-view format using a synthetic representation. The method includes converting a real image in a wide-view format to a synthetic representation using a style model, wherein the synthetic representation depicts a distorted view of an object. The method also includes identifying features of the object using an extraction model that distinguishes different scales of the synthetic representation and a simulated scene to define structures associated with the distorted view. The method also includes detecting the object using a decoder model that identifies an attribute and a bounding box of the object from the features. The method also includes executing a task using the attribute and the bounding box to localize the object in the simulated scene.

Abstract: The disclosure includes embodiments for managing vehicles under anomalous driving behavior. In some embodiments, a method includes determining, by the processor, an ego behavior associated with the ego vehicle and a remote behavior associated with a remote vehicle. The method includes calculating a variance between the ego behavior and the remote behavior. The method includes determining a presence of an anomaly based on the variance satisfying a threshold, wherein the satisfying the threshold indicates the ego behavior is incompatible with the remote behavior. The method includes generating a driving management plan which is configured to mitigate the anomaly and is consistent with the ego behavior. The method includes implementing the driving management plan so that the threshold is no longer satisfied.

Abstract: A system for determining an identification of an occupant within a vehicle can include a processor, a communications interface, and a memory. The memory can store an information reception module, an identification prediction module, and an actuation module. The information reception module can cause the processor to receive, via the communications interface, from at least one sensor, and at a time when a source of a propulsion force for the vehicle is in an off state, at least one signal having information about at least one detectable characteristic associated with the occupant within the vehicle. The identification prediction module can cause the processor to produce, based on the information, a prediction of the identification of the occupant. The actuation module can cause the processor to cause, in response to a production of the prediction, a setting of a component of the vehicle to have a specific value.

Abstract: Systems and methods for managing driver habits are disclosed herein. One embodiment learns undesirable driving habits of a driver over time as the driver operates a vehicle; identifies, for each learned undesirable driving habit, one or more situational triggers associated with that undesirable driving habit; receives information from one or more of vehicle sensors and one or more external sources; predicts that the driver will engage in a particular undesirable driving habit; and carries out one or more of the following avoidance strategies to assist the driver in refraining from engaging in the particular undesirable driving habit: communicating one or more speed advisories to the driver; suggesting an alternate route to the driver; and presenting the driver with one or more of a coupon, an offer, and a discount at a place of business to encourage the driver to take a break by stopping at the place of business.

Abstract: Systems, methods, and other embodiments described herein relate to automatically learning parking preferences for a driver and generating parking recommendations. In one embodiment, a method includes generating training data based at least in part on: 1) trajectory data indicating a trajectory of a vehicle during a plurality of parking events, each in which a driver of the vehicle selected a parking candidate from among a plurality of parking candidates, and 2) attribute data indicating attributes of each of the plurality of parking candidates, removing, from the training data, data associated with at least one available parking candidate based on one or more conditions that indicate the driver did not consider the at least one available parking candidate, and training a decision model, based on remaining training data, to estimate a preferred parking candidate for the driver from among a set of available parking candidates.

Abstract: System, methods, and other embodiments described herein relate to training a model to stylize low-light images for improved perception. In one embodiment, a method includes encoding, by a style model, an input image to identify first content information. The method also includes decoding, by the style model, the first content information into an albedo component and a shading component. The method also includes generating, by the style model, a synthetic image using the albedo component and the shading component. The method also includes training the style model according to computed losses between the input image and the synthetic image.

Abstract: System, methods, and other embodiments described herein relate to improving depth prediction for objects within a low-light image using a style model. In one embodiment, a method includes encoding, by a style model, an input image to identify content information. The method also includes decoding, by the style model, the content information into an albedo component and a shading component. The method also includes generating, by the style model, a synthetic image using the albedo component and the shading component. The method also includes providing the synthetic image to a depth model.

Abstract: System, methods, and other embodiments described herein relate to training a prediction system for improving depth perception in low-light. In one embodiment, a method includes computing, in a first training stage, losses associated with predicting a depth map for a synthetic image of a low-light scene, wherein the losses include a pose loss, a flow loss, and a supervised loss. The method also includes adjusting, according to the losses, a style model and a depth model. The method also includes training, in a second training stage, the depth model using a synthetic representation of a low-light image. The method also includes providing the depth model.

Abstract: Systems, methods, and other embodiments described herein relate to identifying a user of a vehicle. In one embodiment, a method includes determining an activity of the user using one or more mobile devices. The method includes detecting an event in the vehicle using one or more vehicle sensors. The method includes identifying the user based on a relationship between the activity and the event.

Abstract: A system for learning optimal driving behavior for autonomous vehicles comprises a deep neural network, a first stage training module, and a second stage training module. The deep neural network comprises a feature learning network configured to receive sensor data from a vehicle as input and output spatial temporal feature embeddings and a decision action network configured to receive the spatial temporal feature embeddings as input and output an optimal driving policy for the vehicle. The first training stage module is configured to, during a first training stage, train the feature learning network using object detection loss. The second stage training module is configured to, during a second training stage, train the decision action network using reinforcement learning.

Abstract: In an example, a method may assign a first drone of a drone network a first task, the first task may instruct the first drone to transport a first package to a first destination in a geographic area. The method may receive roadway traffic data for a plurality of roadway vehicles in the geographic area; determine, based on the roadway traffic data and during transit of the first package to the first destination by the first drone, to transfer the first package to a second drone in the drone network; and transfer the first package to the second drone in the drone network.

Abstract: An example method determines a plurality of traffic sections for a geographical traffic area, each traffic section including a road segment and vehicle(s) traveling on the road segment; monitors a responsive vehicle rate for a first traffic section of the plurality of traffic sections; and assigns, based on the responsive vehicle rate of the first traffic section, the first traffic section to one of a first section server dedicated to manage the first traffic section and a remote management server capable of managing the plurality of traffic sections of the geographical traffic area, the first section server comprising computing device(s) of responsive vehicle(s) in the first traffic section.

Abstract: A mobile structure includes a controller configured to detect a mobile device proximate to a chip placed proximate to a product within the mobile structure, instruct the mobile device to open a predetermined app in response to the detection, receive a request for information about products related to the product from the mobile device, and instruct a screen of the mobile structure to display the information about the products related to the product in response to receiving the request.

Abstract: In one embodiment, a system includes a controller. The controller is configured to perform operations including obtaining data about an area related to a user, determining vacancy information about the area based on the data, determining at least one of an accessibility and a usability of the area, determining that the area is available for occupancy by one or more mobile units based on the vacancy information and at least one of the accessibility and the usability of the area, and updating a map including the area with an indication that the area is available for occupancy by one or more mobile units.

Abstract: A method comprises receiving a service request from a vehicle, obtaining environment data with one or more sensors, determining a vehicle type of the vehicle based on the service request, determining service data responsive to the service request based on the vehicle type of the vehicle and the environment data, and transmitting a service message comprising the service data to the vehicle.

Abstract: In an example, a method captures, at a first connected vehicle situated in a travel path segment, first sensor data describing an environment proximate to the first connected vehicle. The environment includes a first unconnected vehicle. The method wirelessly receives, via a communication network at the first connected vehicle from a second connected vehicle situated in the travel path segment, second sensor data describing one or more operating characteristics of the first unconnected vehicle. The method estimates, using the first sensor data and the second sensor data, a vehicle action of the first unconnected vehicle.

Abstract: A method for autonomous path planning triggered by freeway rubbernecking includes obtaining gaze directions of drivers of a plurality of connected vehicles in a region of a road. The method also includes determining whether an average speed of the plurality of connected vehicles in the region is less than a predetermined speed. The method further includes controlling mobility of one or more of the plurality of connected vehicles in response to determining that the gaze directions of one or more drivers of the plurality of connected vehicles deviate from the moving direction of the connected vehicles and determining that the average speed of the connected vehicles is less than the predetermined speed.