Systems, methods, and autonomous vehicles may obtain one or more images associated with an environment surrounding an autonomous vehicle; determine, based on the one or more images, an orientation of a head worn item of protective equipment of an operator of a vehicle; determine, based on the orientation of the head worn item of protective equipment, a direction of a gaze of the operator and a time period associated with the direction of the gaze of the operator; determine, based on the direction of the gaze of the operator and the time period associated with the direction of the gaze of the operator, a predicted motion path of the vehicle; and control, based on the predicted motion path of the vehicle, at least one autonomous driving operation of the autonomous vehicle.

A system for operating an autonomous agent with incomplete environmental information can include and/or interface an autonomous operating system and an autonomous agent. A method for operating an autonomous agent with incomplete environmental information includes any or all of: receiving a set of inputs; determining a set of known objects in the ego vehicle's environment; determining a set of blind regions in the ego vehicle's environment; and inserting a set of virtual objects into the set of blind regions; selecting a set of virtual objects based on the set of blind regions; operating the autonomous agent based on the set of virtual objects; and/or any other suitable processes.

Provided is a system and method that can control a merge of an autonomous vehicle when other vehicles are present on the road. In one example, the method may include iteratively estimating a series of values associated with one or more vehicles in an adjacent lane with respect to an ego vehicle, identifying a trend associated with the one or more vehicles from the iteratively estimated series of values, determining merge intentions of the one or more vehicles with respect to the ego vehicle based on the identified trend over time, verifying the merge intentions against a simulated change in the trend, selecting a merge position of the ego vehicle with respect to the one or more vehicles within the lane based on the verified merge intentions, and executing an instruction to cause the ego vehicle to perform a merge operation based on the selected merge position.

A method of classifying a driving maneuver performed by another vehicle in an environment of an ego-vehicle. In the method: a time series of a metrologically determined position of the other vehicle relative to the ego-vehicle that extends to a time step t is provided; spatial profiles of lanes in which the other vehicle may be located are provided; for a plurality of driving maneuvers from a predetermined catalog of possible driving maneuvers, conditional probabilities for the other vehicle to perform this driving maneuver at the time t are respectively determined with a predetermined model by using the time series of the position and the profiles of the lanes; by using these conditional probabilities, a most likely position and/or a probability distribution of positions of the other vehicle at the time step t is determined.

An object warning apparatus for a vehicle including an acquiring module acquiring object information on objects surrounding the vehicle and traffic rule information including traffic rules associated with locations of the vehicle and the objects; a processing module determining (1) whether an object currently violates any one of the traffic rules; (2) whether the object will violate any one of the traffic rules when the vehicle passes the object; and (3) whether the object potentially endangers the vehicle, and to generate an appropriate warning instruction; and an outputting module outputting the warning instruction to a HMI of the vehicle to trigger the HMI so that the HMI warns a human driver of the presence of the object, or to an AD control unit of the vehicle so that the AD control unit executes appropriate automatic driving manipulations.

A system and method for providing probabilistic-based lane-change decision making and motion planning that include receiving data associated with a roadway environment of an ego vehicle. The system and method also include performing gap analysis to determine at least one gap between neighboring vehicles that are traveling within the target lane to filter out an optimal merging entrance for the ego vehicle to merge into the target lane and determining a probability value associated with an intention of a driver of a following neighboring vehicle to yield to allow the ego vehicle to merge into the target lane. The system and method further include controlling the ego vehicle to autonomously continue traveling within the current lane or autonomously merge from current lane to the target lane based on at least one of: if the optimal merging entrance is filtered out and if the probability value indicates an intention of the driver to yield.

Systems and methods for generating instructions for a vehicle to navigate an unsignaled intersection are provided. The method may include: generating an expected return over a sequence of actions of the vehicle; determining an optimal policy by selecting an action with a maximum value for the vehicle; executing dynamic frame skipping to expedite learning a repeated action of the vehicle; prioritize an experience replay by utilizing an experience replay buffer to break correlations between sequential steps of the vehicle; generate a plurality of state-action representations based on at least one of the expected return, the optimal policy, the dynamic frame skipping, or the prioritized experience replay; generate the instructions for navigating the unsignaled intersection based on the plurality of state-action representations; and transmit the instructions for navigating the unsignaled intersection to the vehicle such that the vehicle executes the instructions to navigate the unsignaled intersection.

An apparatus for assisting driving of a vehicle includes a camera mounted to the vehicle for viewing an area in front of the vehicle, a radar sensor mounted to the vehicle to sense around the vehicle, and a controller connected to the camera and/or the radar sensor to detect obstacles and perform collision avoidance control. The controller is further configured to recognize a first obstacle approaching in a lateral direction from an outside of a driving lane of the vehicle, generate and store a collision avoidance path for avoiding collision with the first obstacle, recognize a third obstacle passing behind a second obstacle in the lateral direction after the recognition of the first obstacle is interrupted by the second obstacle, and perform collision avoidance control of the vehicle based on a similarity between the first obstacle and the third obstacle and based on the stored collision avoidance path.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating vehicle intent predictions using a neural network. One of the methods includes obtaining an input characterizing one or more vehicles in an environment; generating, from the input, features of each of the vehicles; and for each of the vehicles: processing the features of the vehicle using each of a plurality of intent-specific neural networks, wherein each of the intent-specific neural networks corresponds to a respective intent from a set of intents, and wherein each intent-specific neural network is configured to process the features of the vehicle to generate an output for the corresponding intent.

Provided are autonomous vehicles (AV), computer program products, and methods for maneuvering an AV in a roadway, including receiving forecast information associated with predicted trajectories of one or more actors in a roadway, determining a relevant trajectory of an actor based on correlating a forecast for predicted trajectories of the actor with the trajectory of the AV, regenerate a distance table for the relevant trajectory previously generated for processing constraints, generate a plurality of margins for the AV to evaluate, the margins based on a plurality of margin types for providing information about risks and effects on passenger comfort associated with a future proximity of the AV to the actor, classifying an interaction between the AV and the actor based on a plurality of margins, and generating continuous scores for each candidate trajectory that is also within the margin of the actor generated for the relevant trajectory.