Abstract: Systems and methods described herein concern improving an interior design of a vehicle under development. One embodiment presents an electronic survey to ridesharing customers from a plurality of ridesharing trips in a rideshare vehicle having a reconfigurable interior that is configured in accordance with a proposed interior design for the vehicle under development, the electronic survey pertaining to one or more features of the proposed interior design, and collects and stores, in a computer memory, responses to the electronic survey for the plurality of ridesharing trips. The proposed interior design is modified based, at least in part, on the responses to the electronic survey.

Abstract: In one embodiment, a rest stop recommendation system monitors driving behavior of a driver and stores information indicating the driving behavior as historical data, determines, based at least in part on the historical data, a driver tiredness state, a continuous driving length preference, and a refueling pattern preference, determines a time to recommend a rest stop based at least in part on the driver tiredness state, the continuous driving length preference, and the refueling pattern preference, extracts, from a map database, a plurality of rest stops within a predetermined radius of a position of the vehicle, determines rest stop characteristic preferences based at least in part on the historical data, selects one or more potential rest stops from among the plurality of rest stops based at least in part on the rest stop characteristic preferences, and presents the one or more potential rest stops to the driver in a recommendation.

Abstract: A method comprises receiving an image of a road captured by a vehicle driving on the road, receiving a map of the road, the map comprising a road geometry of the road, obtaining an edge map of the road based on the image of the road, inputting the image, the map of the road, and the edge map into a trained regressor neural network, determining an estimated snow depth for each of one or more lanes of the road based on an output of the regressor neural network, and transmitting the estimated snow depth to an edge computing device.

Abstract: A vehicular hazard mitigation system includes one or more processors and a memory communicably coupled to the processors. The memory may store a vehicular hazard mitigation module including computer-readable instructions that when executed by the processors cause the processors to determine a base vehicle currently driving in an enhanced response driving mode. After determining the base vehicle, at least one alert zone of the base vehicle is determined. A determination is then made as to whether at least one other vehicle is currently in the at least one alert zone of the base vehicle. If at least one other vehicle is currently in the at least one alert zone of the base vehicle, the vehicular hazard mitigation module may autonomously control operation of the base vehicle to shift the driving mode of the base vehicle from the enhanced response driving mode to a non-enhanced response driving mode.

Abstract: A method of generating a traffic prediction includes receiving traffic data from a plurality of reporting sources, forming a plurality of initial graphs, generating a plurality of completed graphs based on the plurality of initial graphs by at least removing noise from the plurality of initial graphs, generating a plurality of feature vectors that represent spatial relationships and temporal relationships in the plurality of completed graphs, outputting a first feature vector corresponding to a first time window as a first output, caching a copy of the first feature vector with a set of feature vectors of the plurality of feature vectors corresponding to a second time window, connecting the first feature vector with the set of feature vectors, outputting the result as a second output, fusing the first output with the second output and generating a traffic prediction based on the fused outputs.

Abstract: The disclosure includes embodiments for a location data correction service for connected vehicles. A method includes receiving, by an operation center via a serverless ad-hoc vehicular network, a first wireless message that includes legacy location data that describes a geographic location of a legacy vehicle. The method includes causing a rich sensor set included in the operation center to record sensor data describing the geographic locations of objects in a roadway environment. The method includes determining correction data that describes a variance between the geographic location of the legacy vehicle as described by the sensor data and the legacy location data. The method includes transmitting a second wireless message to the legacy vehicle, wherein the second wireless message includes the correction data so that the legacy vehicle receives a benefit by correcting the legacy location data to minimize the variance.

Abstract: A health risk assessment system for estimating a health risk posed by transporting a user in a vehicle. The system includes one or more processors and a memory communicably coupled to the one or more processors and storing a health risk assessment module. The health risk assessment module includes computer-readable instructions that when executed by the one or more processors cause the one or more processors to attempt to acquire information regarding a travel history of the user. The module is also configured to, using acquired travel history information, determine a value of a first health status parameter and to compare the value of the first health status parameter to a first threshold. Responsive to the value of the first health status parameter exceeding the threshold, the module controls operation of the vehicle to implement at least one infection prevention measure.

Abstract: Systems and methods for distributed cooperative localization in a connected vehicular platform are disclosed herein. At a first sensor-rich vehicle, one embodiment receives, from a second sensor-rich vehicle via direct vehicle-to-vehicle (V2V) communication, an estimated location of the first sensor-rich vehicle and an associated confidence level; estimates a location of the second sensor-rich vehicle based on first sensor data and assigns a confidence level to the estimated location; transmits to the second sensor-rich vehicle via direct V2V communication, the estimated location of the second sensor-rich vehicle and the assigned confidence level; accepts, based on communication between the first and second sensor-rich vehicles and predetermined acceptance criteria, an assignment to perform localization for a legacy vehicle; estimates a location of the legacy vehicle based on second sensor data; and transmits, to the legacy vehicle, the estimated location of the legacy vehicle.

Abstract: The disclosure includes embodiments a first connected vehicle to transmit a set of wireless communications based on a communication plan generated by a cloud server by observing a set of digital twin simulations. A method includes wirelessly transmitting, by the first connected vehicle that is included in a set of connected vehicles, first context data that describes a context of the first connected vehicle and a set of wireless communications to be sent by the first connected vehicle. The method includes receiving plan data that describes a communication plan for transmitting the set of wireless communications, wherein the set of digital twin simulations are executed by the cloud server based at least in part on the first context data and a set of other context data provided by the set of connected vehicles. The method includes wirelessly transmitting the set of wireless communications in compliance with the communication plan.

Abstract: An object detection system can identify a specific object in a two-dimensional image. A processor can receive the two-dimensional image with bounding boxes around objects. The processor can cause a point to appear in the two-dimensional image. The point can represent a position, in three-dimensional space, of the specific object. The processor can determine an existence of a condition. The condition can be that the point is enclosed by a plurality of bounding boxes. The processor can receive, in response to the existence of the condition, a depth image. The processor can determine, in response to the existence of the condition and based on information included in the depth image and a location of the point in the two-dimensional image, a specific bounding box that encloses the point. The processor can cause, based on a location of the specific bounding box, an indication of the specific object to be presented.

Abstract: An object detection system can identify a specific object in a two-dimensional image. A processor can receive the two-dimensional image with bounding boxes around objects. The processor can cause a point to appear in the two-dimensional image. The point can represent a position, in three-dimensional space, of the specific object. The processor can determine an existence of a condition. The condition can be that the point is enclosed by a plurality of bounding boxes. The processor can receive, in response to the existence of the condition, a depth image. The processor can determine, in response to the existence of the condition and based on information included in the depth image and a location of the point in the two-dimensional image, a specific bounding box that encloses the point. The processor can cause, based on a location of the specific bounding box, an indication of the specific object to be presented.

Abstract: A method comprises making initial predictions of whether a first vehicle will perform a lane change at a plurality of future time steps based on sensor data captured by an egovehicle; and in response to making an initial prediction that the first vehicle will perform a lane change at a first one of the future time steps, making final predictions that the first vehicle will perform a lane change at each of a plurality of time steps subsequent to the first one of the future time steps.

Abstract: An embodiment of the present disclosure takes the form of a method carried out by a perception-network device. The perception-network device provides a first perspective view of a scene to a first branch of a neural network, and generates a feature map via the first branch based on the first perspective view. The perception-network device augments the generated feature map with features of a complementary feature map generated by a second branch of the neural network provided with a second perspective view of the scene. The perception-network device generates a perception inference via the neural network based on the augmented feature map.

Abstract: Systems and methods described herein relate to providing guidance to vehicle drivers regarding predicted lane-change behavior of other drivers. One embodiment transforms historical vehicle trajectory data into a corresponding alternative representation; applies a clustering algorithm to group a plurality of drivers into groups of similar drivers; applies Bayesian inference to train a Bayesian neural network (BNN) for the drivers in each group; adapts the BNN for each group to generate a personalized BNN for each driver in that group; identifies a particular driver on a roadway; receives information regarding the particular driver's vehicle and one or more other nearby vehicles; estimates a probability that the particular driver will change lanes using the personalized BNN for that driver; and communicates guidance regarding predicted lane-change behavior of the particular driver to at least one nearby vehicle based, at least in part, on the estimated probability that the particular driver will change lanes.

Abstract: Systems and methods for compensating for driver speed-tracking error are disclosed herein. One embodiment computes a recommended speed for a vehicle; classifies a driver of the vehicle as a particular type of driver among a plurality of driver types based on measured speed-tracking error, wherein the speed-tracking error is a difference between the recommended speed and an actual speed of the vehicle; predicts the speed-tracking error at a future time increment based on the speed-tracking error at one or more past time increments using a nonlinear autoregressive (NAR) neural network associated with the particular type of driver; computes a compensated recommended speed for the vehicle based on the recommended speed and the predicted speed-tracking error at the future time increment; and communicates the compensated recommended speed to the driver.

Abstract: A server can assist connected vehicles merging from a ramp onto a main lane. A connected vehicle can be configured to communicate with the server. Connected vehicles on the main lane and connected vehicles on the ramp can send their locations and speeds to the server. The server can determine when at least one connected vehicle is within a geo-fence surrounding a merging point, where the main lane and the ramp meet. The server can determine an order in which the connected vehicles on the main lane and on the ramp can merge based on the connected vehicles' locations and speeds. The server can determine an advisory speed for at least one of the connected vehicles based on the order. The server can send the advisory speed to the at least one connected vehicle.

Abstract: An edge device can send a vehicle-specific alert to a connected vehicle, which is configured to communicate with the edge device. The edge device can receive observations of other vehicles on the road from connected vehicles and/or roadside units. The edge device can also receive driving history and/or other information about the other vehicles from a server. The edge device can classify the behavior of vehicles detected in the observations based on the observations, driving history, and/or other information. Based on the classified behavior, the edge device can determine whether and how the connected vehicle is impacted and, if impacted, send the vehicle-specific alert to the connected vehicle.

Abstract: Systems and methods for compensating for driver speed-tracking error are disclosed herein. One embodiment computes a recommended speed for a vehicle; classifies a driver of the vehicle as a particular type of driver among a plurality of driver types based on measured speed-tracking error, wherein the speed-tracking error is a difference between the recommended speed and an actual speed of the vehicle; predicts the speed-tracking error at a future time increment based on the speed-tracking error at one or more past time increments using a nonlinear autoregressive (NAR) neural network associated with the particular type of driver; computes a compensated recommended speed for the vehicle based on the recommended speed and the predicted speed-tracking error at the future time increment; and communicates the compensated recommended speed to the driver.

Abstract: A vehicle for interpreting a traffic scene is provided. The vehicle includes one or more sensors configured to capture an image of an external view of the vehicle, and a controller. The controller is configured to obtain the captured image of the external view of the vehicle from the one or more sensors, segment a plurality of instances from the captured image, determine relational information among the plurality of instances, and generate a hyper graph including a plurality of nodes representing the plurality of instances and a plurality of edges representing the relational information among the plurality of instances.

Abstract: The disclosure includes embodiments a first connected vehicle to transmit a set of wireless communications based on a communication plan generated by a cloud server by observing a set of digital twin simulations. A method includes wirelessly transmitting, by the first connected vehicle that is included in a set of connected vehicles, first context data that describes a context of the first connected vehicle and a set of wireless communications to be sent by the first connected vehicle. The method includes receiving plan data that describes a communication plan for transmitting the set of wireless communications, wherein the set of digital twin simulations are executed by the cloud server based at least in part on the first context data and a set of other context data provided by the set of connected vehicles. The method includes wirelessly transmitting the set of wireless communications in compliance with the communication plan.