System, methods, and other embodiments described herein relate to training a depth model for joint depth completion and prediction. In one arrangement, a method includes generating depth features from sparse depth data according to a sparse auxiliary network (SAN) of a depth model. The method includes generating a first depth map from a monocular image and a second depth map from the monocular image and the depth features using the depth model. The method includes generating a depth loss from the second depth map and the sparse depth data and an image loss from the first depth map and the sparse depth data. The method includes updating the depth model including the SAN using the depth loss and the image loss.

To obtain a rider-assistance system capable of providing a rider of a straddle-type vehicle with a sense of comfort and safety during a turn, and a control method for such a rider-assistance system.

The present invention provides the rider-assistance system that assists with driving by the rider of the straddle-type vehicle and includes a controller. The controller includes: an object identification section that identifies an object approaching a side of the straddle-type vehicle on the basis of output of a communication device that wirelessly receives information output from infrastructure equipment or another vehicle; a body position information acquisition section that acquires position information of at least a part of a body of the rider on the turning straddle-type vehicle; a collision possibility determination section that determines a collision possibility of the rider with the object identified by the object identification section on the basis of the position information acquired by the body position information acquisition section; and a safety operation performing section that causes the rider-assistance system to perform safety operation in the case where the collision possibility determination section determines that the collision possibility is high.

Sensor data is received from an array of sensors configured to capture one or more objects in an external environment of an autonomous vehicle. A first sensor group is selected from the array of sensors based on proximity data or environmental contexts. First sensor data from the first sensor group is prioritized for transmission based on the proximity data or environmental contexts.

An apparatus for controlling a lane change of a vehicle includes: a sensor to sense an external vehicle, an input device to receive a lane change command from a driver of the vehicle, and a control circuit to be electrically connected with the sensor and the input device. The control circuit may receive the lane change command using the input device, calculate a minimum operation speed for lane change control, and determine whether to accelerate the vehicle based on a distance between a preceding vehicle which is traveling on the same lane as the vehicle and the vehicle, when a driving speed of the vehicle is lower than the minimum operation speed when receiving the lane change command.

Methods and systems for controlling navigation of a vehicle are disclosed. The system will first identify a plurality of goal points corresponding to a drivable area that a vehicle is traversing or will traverse, where the plurality of goal points are potential targets that an uncertain road user (URU) within the drivable area can use to exit the drivable area. The system will then receive perception information relating to the URU within the drivable area, and identify a target exit point from the plurality goal points based on a score. The score is computed based on the received perception information and a loss function. The system will generate a trajectory of the URU from a current position of the URU to the target exit point, and control navigation of the vehicle to avoid collision with the URU.

A method for intersection assistance, the method may include obtaining sensed information regarding an environment of the vehicle; determining an occurrence of an intersection related situation, based on the sensed information; generating one or more intersection driving related decisions; wherein the generating comprises processing, by one or more narrow AI agents of a group of narrow AI agents, at least one out of (a) at least a first part of the sensed information, and (b) an outcome of a pre-processing of at least a second part of the sensed information; and responding to the one or more intersection driving related decisions; wherein the responding comprises at least one out of (a) executing the one or more intersection driving related decisions, and (b) suggesting executing the one or more intersection driving related decisions.

The present disclosure relates to an apparatus and method for controlling a brake system based on a driver's forward gaze.

A pedestrian tracking system includes: a buffer or a memory configured to store a trajectory sequence of a pedestrian; a step attention module and a control module. The step attention module iteratively performs a step attention process to predict states of the pedestrian. Each iteration of the step attention process includes the step attention module: learning the stored trajectory sequence to provide time-dependent hidden states, reshaping each of the time-dependent hidden states to provide two-dimensional tensors; condensing the two-dimensional tensors via convolutional networks to provide convolutional sequences; capturing global information of the convolutional sequences to output a set of trajectory patterns represented by a new sequence of tensors; learning time-related patterns in the new sequence and decoding the new sequence to provide one or more of the states of the pedestrian; and modifying the stored trajectory sequence to include the predicted one or more of the states of the pedestrian.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for predicting how likely it is that a target agent in an environment will yield to another agent when the pair of agents are predicted to have overlapping future paths. In one aspect, a method comprises obtaining a first trajectory prediction specifying a predicted future path for a target agent in an environment; obtaining a second trajectory prediction specifying a predicted future path for another agent in the environment; determining that, at an overlapping region, the predicted future path for the target agent overlaps with the predicted future path for the other agent; and in response: providing as input to a machine learning model respective features for the target agent and the other agent; and obtaining the likelihood score as output from the machine learning model.

A device may receive accelerometer data and video data for a vehicle and may identify bounding boxes and object classes for objects near the vehicle. The device may identify tracks for the objects and may filter out tracks that are not associated with vehicles or vulnerable road users to generate one or more tracks or an indication of no tracks. The device may generate a collision cone identifying a drivable area of the vehicle to identify objects more likely to be involved in a collision and may filter out tracks from the one or more tracks, based on the bounding boxes, and to generate a subset of tracks or another indication of no tracks. The device may determine scores for the subset of tracks and may identify a track of the subset of tracks with a highest score. The device may perform actions based on the identified track.