Lane-level route planning includes obtaining lane-level information of a road, where the road includes a first lane and a second lane and the lane-level information includes first lane information related to the first lane and second lane information related to the second lane; converting the lane-level information to probabilities for a state transition function; receiving a destination; and obtaining a policy as a solution to a model that uses the state transition function.

An on-board system of a vehicle scans for target entities in at least one lane to a side of the vehicle and determines position and state of motion of detected target entities. From a state of the vehicle, an intention is inferred of a driver to move the vehicle into one of the at least one lane. If the on-board system detects a risk of collision between a target entity and the vehicle, then the motion of the vehicle is impeded by the system applying brakes of the vehicle and/or reducing a driving torque of the vehicle. A speed of the vehicle is monitored and a motion of the vehicle is not impeded if the speed of the vehicle is above a threshold speed.

An active road surface maintenance system and method developed for connected vehicles with the aid of a mobility digital twin (MDT) framework. A method performed in a cloud-based digital space includes receiving data regarding a physical object from a physical space connected to a vehicle. The method also includes processing the data using machine learning to model road surface conditions, in which respective penalty values are assigned to corresponding road surfaces, a respective penalty value being higher the lower a condition of the corresponding road surface. The method also includes deriving instructions based on the modeled road surface conditions and the respective penalty values to guide actuation of the vehicle along a trajectory. The method further includes transmitting the instructions to the physical space connected to the vehicle to guide actuation of the vehicle.

A system obtains scene context data characterizing the environment. The scene context data includes data that characterizes a trajectory of an agent in a vicinity of a vehicle up to a current time point. The system identifies a plurality of initial target locations, and generates, for each of a plurality of target locations that each corresponds to one of the initial target locations, a respective predicted likelihood score that represents a likelihood that the target location will be an intended final location for a future trajectory of the agent. For each target location in a first subset of the target locations, the system generates a predicted future trajectory for the agent given that the target location is the intended final location for the future trajectory. The system further selects, as likely future trajectories of the agent, one or more of the predicted future trajectories.

A driving support apparatus according to the invention estimates the position of a moving body by controlling a position estimation unit when the tracking-target moving body leaves a first area or a second area to enter a blind spot area and detects the position of the moving body by controlling a position detection unit when the moving body leaves the blind spot area to enter the first area or the second area. In this manner, the trajectory of the tracking-target moving body is calculated so that the trajectory of the moving body detected in the first area or the second area and the trajectory of the moving body estimated in the blind spot area are continuous to each other and driving support is executed based on the calculated trajectory of the tracking-target moving body.

A travel controller recognizes an action of a driver of a vehicle from image data of the driver. The travel controller obtains information indicating that determination of whether autonomous driving of the vehicle is permissible cannot be given. When the information indicating that determination of whether the autonomous driving of the vehicle is permissible cannot be given is obtained, the travel controller operates a human interface to request the driver for an instruction to drive the vehicle. The travel controller determines whether the driver is giving an instruction to drive the vehicle from an action of the driver recognized in response to the request for an instruction to drive the vehicle. When determined in the determination process that the driver is giving an instruction to drive the vehicle, the travel controller operates a drive system of the vehicle to permit autonomous driving of the vehicle.

Methods, systems, and non-transitory computer readable media are configured to perform operations comprising determining a plurality of predetermined situations in which lighting operation of an ego vehicle should automatically transition; determining occurrence of a predetermined situation of the plurality of predetermined situations; and causing an automatic transition in lighting operation of the ego vehicle.

An advanced pedestrian warning or alert system is a system for automotive vehicles is described herein in various embodiments. This system detects if a vulnerable road user, for example a pedestrian, is at an unsafe distance from the moving vehicle. If a pedestrian is too close to the vehicle when the vehicle is in motion, then the system will issue auditory and/or visual warnings to notify the pedestrian back to safety.

An autonomous driving safety system for sharing a risk-based operation design domain includes an autonomous driving system configured to control a vehicle driving unit according to information detected by a sensor unit to perform autonomous driving, in which the autonomous driving system includes an operation design domain update unit configured to evaluate a risk of at least one of a static operation design domain and a dynamic operation design domain recognized by the sensor unit while driving to update the operation design domain.

Systems and methods for modeling and predicting scene occupancy in an environment of a robot are disclosed herein. One embodiment processes past agent-trajectory data, map data, and sensor data using one or more encoder neural networks to produce combined encoded input data; generates a weights vector for a Gaussian Mixture Model (GMM) based on the combined encoded input data; produces a volumetric spatio-temporal representation of occupancy in an environment of a robot by generating, for a plurality of modes of the GMM in accordance with the weights vector, corresponding sample probability distributions of scene occupancy based on respective means and variances of the plurality of modes, wherein the respective means and variances sample coefficients of a set of learned basis functions; and controls the operation of the robot based, at least in part, on the volumetric spatio-temporal representation of occupancy.