Abstract: Systems and methods for object detection. Object detection may be used to control autonomous vehicle(s). For example, the methods comprise: obtaining, by a computing device, a LiDAR dataset generated by a LiDAR system of the autonomous vehicle; and using, by the computing device, the LiDAR dataset and image(s) to detect an object that is in proximity to the autonomous vehicle. The object being is detected by: computing a distribution of object detections that each point of the LiDAR dataset is likely to be in; creating a plurality of segments of LiDAR data points using the distribution of object detections; merging the plurality of segments of LiDAR data points to generate merged segments; and detecting the object in a point cloud defined by the LiDAR dataset based on the merged segments. The object detection may be used by the computing device to facilitate at least one autonomous driving operation.

Abstract: Aspects and implementations of the present disclosure relate to performance and safety improvements for autonomous trucking systems, including techniques of obtaining an identification of an unfavorable condition on a route of an autonomous vehicle (AV), causing the AV to exit the route, and performing one or more waiting loops until the unfavorable condition is resolved, the AV is rerouted, assistance arrives, and the like.

Abstract: A parking support system of a vehicle, comprises: a detection unit configured to detect information of a periphery of the vehicle; a control unit configured to control, based on the information detected by the detection unit, a parking space entry operation of the vehicle to a parking space and a parking space exit operation of the vehicle from the parking space; and a stationary state control unit configured to maintain a stationary state of the vehicle after one of the parking space entry operation and the parking space exit operation has been completed. The stationary state control unit maintains the stationary state by first maintenance control when the parking space entry operation has been completed and maintains the stationary state by second maintenance control, which is different from the first maintenance control when the parking space exit operation has been completed.

Abstract: An imitation learning-based machine-learned (ML) model to augment or replace the prediction and/or planner components of an autonomous vehicle may be trained using a two stage and multi-discipline approach. A first stage of training may include training the ML component to output a predicted action associated with a target vehicle and modifying the ML component to reduce a difference between the predicted action and the observed action taken by the target vehicle. A second stage may use reinforcement learning to further tun the ML component. The resultant model may be used on its own, with enough training data, or to rank or weight candidate trajectories generated by a planning component of the vehicle. The ML component may provide embeddings of environment features to first transformer(s) that output to a long short-term memory that outputs to second transformer(s) to determine the predicted action.

Abstract: A vehicle control device recognizes a surrounding situation of a vehicle, controls steering and acceleration/deceleration of the vehicle without depending on an operation of a driver, determines any one of a plurality of driving modes including a first driving mode and a second driving mode as a driving mode of the vehicle, changes the driving mode of the vehicle to a driving mode in which the task is heavier when a task is not executed by the driver, determines whether there is an error in the map information and determines whether there is a recognition error, and when an error is determined in the map information, there is a recognition error while the vehicle is traveling in the second driving mode, and a preceding vehicle is recognized within a first predetermined distance, sets a travel continuation distance in the second driving mode to be longer.

Abstract: Disclosed are an apparatus and a method for controlling driving of a vehicle. The apparatus includes a sensor that obtains information about an external environment of the vehicle, and a controller that calculates a target driving speed based on at least one of weather information, road surface information, or a minimum sensing distance detectable by the sensor, and sets a preset driving speed as the target driving speed. Accordingly, the driving of the vehicle is controlled by accurately reflecting external environment information and setting the target driving speed of the vehicle, so that it is possible to minimize user anxiety about the stability of autonomous driving and reduce the risk of an accident.

Abstract: A vehicle system is provided that optimizes vehicle operation by utilizing artificial intelligence and machine learning and providing dynamic changes to the vehicle characteristics. The vehicle system includes one or more sub-systems controlled by a vehicle operating system utilizing artificial intelligence and machine learning.

Abstract: A method includes obtaining information about a lead vehicle. The lead vehicle is traveling directly ahead of the host vehicle and the information includes velocity of the lead vehicle and a gap g between the host vehicle and the lead vehicle. Based on detecting a change in the velocity of the lead vehicle in the information, a corresponding desired acceleration is computed for the host vehicle using perceived acceleration of the lead vehicle to maintain the gap g within a specified range of gap values. The perceived acceleration of the lead vehicle is based on the change in the velocity indicated by the information. Based on a check of parameters involved in the computing the corresponding desired acceleration, a modified desired acceleration is computed for the host vehicle using a lag for the lead vehicle that results in an intended acceleration of the lead vehicle differing from the perceived acceleration.

Abstract: In some embodiments, the present disclosure provides an algorithm for a data-driven intelligent adaptive model of an automobile cab based on a biometric identification technology. The algorithm includes the following steps: acquiring driving posture data, the driving posture data including driver information, automobile man-computer parameter information, and seat position information; building a feature extraction model of the driving posture data and screening feature vectors playing a significant role in predicting a seat position by the feature extraction model; building a prediction model of a comfortable seat position by a back propagation (BP) neural network and inputting the screened feature vectors to perform training on the prediction model; and inputting specific feature vectors to obtain the seat position information and accordingly adjust the seat position.

Abstract: Systems, methods, and non-transitory computer-readable media configured to obtain one or more series of successive sensor data frames during a navigation of a vehicle. Disengagement data is obtained. The disengagement data indicates whether a vehicle is in autonomous mode. A training dataset with which train a machine learning model is determined based on the one or more series of successive sensor data frames and the disengagement data. The training dataset includes a subset of the one or more series of successive sensor data frames and a subset of the disengagement data, the machine learning model being trained to identify unsafe driving conditions.

Abstract: A driving environment is perceived based on sensor data obtained from a plurality of sensors mounted on an ADV. A first set of features is extracted from the sensor data representing the driving environment, where the first set of features include one or more obstacles moving relative to the ADV. A precaution notification is generated by applying a precautionary slowdown predictive model to the first set of features and a second set of features determined based on internal states of the ADV. In response to the precaution notification, a speed planning is performed to lower a speed limit of the ADV to a predetermined percentage of the speed limit. The ADV is controlled to drive autonomously according to the lowered speed limit to perform a precautionary slowdown.

Abstract: A caravanning vehicle system includes a lead vehicle having an engine and at least one trailing vehicle having an electric motor. The at least one trailing vehicle is non-mechanically connected to the lead vehicle. The lead vehicle includes a power transfer system that is configured to transfer electric power, wirelessly, to the at least one trailing vehicle.

Abstract: A system, including a processor and a memory, the memory including instructions to be executed by the processor to receive one or more images from a vehicle, wherein a first deep neural network included in a computer in the vehicle has failed to determine an orientation of a first object in the one or more images. The instructions can include further instructions to generate a plurality of modified images with a few-shot image translator, wherein the modified images each include a modified object based on the first object. The instructions can include further instructions to re-train the deep neural network to determine the orientation of the first object based on the plurality of modified images and download the re-trained deep neural network to the vehicle.

Abstract: A working vehicle includes a vehicle body capable of traveling either by either manual steering with a steering wheel or by automatic steering with the steering wheel, a reference registration controller configured or programmed to register a traveling reference line based on a position of the vehicle body caused to travel by the manual steering, an automatic steering controller configured or programmed to perform control by the automatic steering on a set scheduled traveling line based on the traveling reference line, a display to display information about the automatic steering when forward traveling by the automatic steering is to be performed, and a notifier to issue a notification of the information about the automatic steering aside from the information displayed on the display when rearward traveling by the automatic steering is to be performed.

Abstract: Provided is an electric brake including: a brake mechanism configured to transmit, based on a braking request, a thrust force generated through drive of an electric motor to a piston configured to move brake pads to be pressed against a disc rotor; and an ECU for rear electric brake configured to control the drive of the electric motor, and to move, in a non-braking state, the piston to a predetermined clearance position at which a clearance between each of the brake pads and the disc rotor is a predetermined amount. The ECU for rear electric brake is configured to drive the electric motor so that a period from generation of the braking request to a start of the pressing of the disc rotor by the brake pads is a predetermined period regardless of a position of the piston at a time when the braking request is generated.

Abstract: Method for detecting a failure in a pneumatic system of a vehicle, the pneumatic system comprising a supply pneumatic subsystem comprising supply lines, a control pneumatic subsystem comprising control lines, a delivery pneumatic subsystem comprising delivery lines, the vehicle comprising an electronic control unit configured to implement a trained supply recurrent neural network, a trained control recurrent neural network, a trained delivery recurrent neural network, and a trained main recurrent neural network; the method comprising collect and provide the input parameters to the trained recurrent neural networks to get a output from recurrent neural networks; provide the input parameters and the output parameters from recurrent neural networks to get a main output indicative of a pneumatic system health parameter indicative of a presence or an absence of a failure in one of the supply pneumatic subsystem, or the control pneumatic subsystem, or the delivery pneumatic subsystem of the pneumatic system.

Abstract: Techniques for determining whether to yield a vehicle to an oncoming object based on a stationary object are discussed herein. A vehicle can determine a drivable area in an environment through which the vehicle is travelling. A trajectory associated with the vehicle can be determined. A probability that the oncoming object will intersect a drive path associated with a vehicle trajectory can be determined. In some case, the probability can be based on various distance metric(s) of the environment. If the oncoming object is determined to intersect the drive path, a stopping location for the vehicle can be determined to allow the oncoming object to traverse around the stationary object. The vehicle can be controlled based on sensor data captured by the vehicle and the probability.

Abstract: A method for controlling an opening device and to an opening device for a motor vehicle door element, having a control unit, an electric drive, and an actuating element, wherein the actuating element can be actuated by means of the drive, and the door element can be opened by means of the actuating element. The opening device can be controlled on the basis of a position, in particular the inclination of the motor vehicle.

Abstract: Techniques are disclosed for systems and methods to provide assisted and/or autonomous navigation for mobile structures. A docking assist system includes a logic device, one or more sensors, one or more actuators/controllers, and modules to interface with users, sensors, actuators, and/or other modules of a mobile structure. The logic device is adapted to receive docking assist parameters from a user interface and perimeter sensor data from a perimeter ranging system mounted to the mobile structure. The logic device determines docking assist control signals based on the docking assist parameters and perimeter sensor data. The logic device may then provide the docking assist control signals to a navigation control system. Control signals and/or other docking assist analysis and/or parameters may be displayed to a user and/or used to adjust a steering actuator, a propulsion system thrust, and/or other operational systems of the mobile structure.

Abstract: Techniques for determining whether to yield a vehicle to an oncoming object based on a stationary object are discussed herein. A vehicle can determine a drivable area in an environment through which the vehicle is travelling. A trajectory associated with the vehicle can be determined. A probability that the oncoming object will intersect a drive path associated with a vehicle trajectory can be determined. In some case, the probability can be based on various distance metric(s) (e.g., dynamic distance metrics) of the environment. If the oncoming object is determined to intersect the drive path, a stopping location for the vehicle can be determined to allow the oncoming object to traverse around the stationary object. The vehicle can be controlled based on sensor data captured by the vehicle and the probability.