Abstract: Artificial-intelligence-based river information system. In an embodiment, a first training dataset is used to train a travel time prediction model to predict a travel time along the waterway for a given trip. In addition, a second training dataset is used to train a river level prediction model to predict a river level along the waterway for a given time. For each of a plurality of trips, a request is received that specifies the trip and a time of the trip, and, in response to the request, the travel time prediction model is used to predict a travel time for the trip, and the river level prediction model is used to predict a river level of the waterway at one or more points along the trip. Then, a voyage plan is generated based on one or both of the predicted travel time and the predicted river level.

Abstract: A method for training an image depth estimation model. A sample environmental image, sample environmental point cloud data and sample edge information of the sample environmental image are input into a to-be-trained model; initial depth information of each of pixel points in the sample environmental image and a feature relationship between each of the pixel points and a corresponding neighboring pixel point of each of the pixel points are determined through the to-be-trained model, the initial depth information of each of the pixel points is optimized according to the feature relationship to obtain optimized depth information of each of the pixel points, and a parameter of the to-be-trained model is adjusted according to the optimized depth information to obtain the image depth estimation model.

Abstract: An adjustable transportation container system is provided including a container including a plurality of container segments defining a geometry, a container adjustment device interconnecting the plurality of container segments, one or more processors, and one or more memory modules. The one or more memory modules include a computer-readable medium storing computer-readable instructions that, when executed by the one or more processors, cause the one or more processors to identify a weight distribution of the container, identify an adjusted geometry of the container, and actuate the container adjustment device to change the geometry of the container to conform to the adjusted geometry.

Abstract: A method for learning depth-aware keypoints and associated descriptors from monocular video for monocular visual odometry is described. The method includes training a keypoint network and a depth network to learn depth-aware keypoints and the associated descriptors. The training is based on a target image and a context image from successive images of the monocular video. The method also includes lifting 2D keypoints from the target image to learn 3D keypoints based on a learned depth map from the depth network. The method further includes estimating a trajectory of an ego-vehicle based on the learned 3D keypoints.

Abstract: A method is described for validating a current position of a vehicle. The method includes receiving position data from the vehicle, receiving position validation data associated with the vehicle, and validating the current position of the vehicle provided by the position data against the position validation data to determine an accuracy value for the current position. The position data provides the current position of the vehicle, and the position validation data is generated by a source distinct from the vehicle.

Abstract: A system includes: sensors that includes a position sensor to determine a vehicle speed, an angular displacement sensor to determine a roll angle corresponding to an angular displacement of the vehicle in a roll direction and a roll rate corresponding to an angular velocity of the vehicle in the roll direction, and a steering torque sensor to determine a steering torque applied to a steering handle of the vehicle; and a stabilizing unit coupled to the sensors. The stabilizing unit: determines a synthesized torque based on the roll angle, the roll rate, and the vehicle speed; determines a tuning parameter based on a comparison of the synthesized torque and the steering torque; determines a stabilizing torque based on the tuning parameter and the synthesized torque; and provides an actuating signal corresponding to the stabilizing torque for applying the stabilizing torque to the steering handle of the vehicle.

Abstract: System, methods, and other embodiments described herein relate to accurately distinguishing a traffic light from other illuminated objects in the traffic scene and detecting states using hierarchical modeling. In one embodiment, a method includes detecting, using a machine learning (ML) model, two-dimensional (2D) coordinates of illuminated objects identified from a monocular image of a traffic scene for control adaptation by a control model. The method also includes assigning, using the ML model, computed probabilities to the illuminated objects for categories within a hierarchical ontology of environmental lights associated with the traffic scene, wherein one of the probabilities indicates existence of a traffic light instead of a brake light in the traffic scene. The method also includes executing a task by the control model for a vehicle according to the 2D coordinates and the computed probabilities.

Abstract: A navigation system for a marine vessel incudes one or more proximity sensors, each at a sensor location on the marine vessel and configured to measure proximity of objects in an area surrounding the marine vessel and generate proximity measurements, and a control system configured to receive the proximity measurements measured by the one or more proximity sensors. From the received proximity measurements, four linearly-closest proximity measurements to the marine vessel are identified, including one closest proximity measurement to the marine vessel in each of a positive X direction, a negative X direction, a positive Y direction, and a negative Y direction. A most important object (MIO) dataset is generated identifying the four linearly-closest proximity measurements and propulsion of the marine vessel is controlled based at least in part on the MIO dataset.

Abstract: A device and a method for controlling collision avoidance of a trailer vehicle in platooning include a jackknife induction determination module or operation that determines whether to execute jackknife induction control in a leading vehicle when an emergency braking situation occurs during platooning of the trailer vehicle. The device and method also include a jackknife induction control module or operation that provides an additional braking distance of a following vehicle at the rear of a tractor by inducing pivoting of a trailer while controlling the tractor and trailer of the leading vehicle, respectively, to allow a jackknife phenomenon to occur. Stability of the platooning is improved by reducing a possibility of the occurrence of a collision accident by the following vehicle even during emergency braking.

Abstract: The disclosure generally pertains to systems and methods to protect a side-view mirror of a vehicle. An example method executed by a processor includes obtaining a first dimensional parameter associated with a body portion of a vehicle (for example, a width or a length of the vehicle) and a second dimensional parameter associated with a side-view mirror attached to the vehicle (for example, a protrusion distance of the side-view mirror). A turning path characteristic associated with a movement of the vehicle may then be determined, followed by determining a probability of a collision between the side-view mirror and an object located outside the vehicle. Determining the probability may be based on the first dimensional parameter, the second dimensional parameter, and the turning path characteristic. If the probability of the collision exceeds a threshold value, an alert may be issued, or a preventive maneuver executed in order to prevent the collision.

Abstract: One embodiment of a system for analyzing the in-lane driving behavior of an external road agent generates a sequence of sparse 3D point clouds based on a sequence of depth maps corresponding to a sequence of images of a scene. The system performs flow clustering based on the sequence of depth maps and a sequence of flow maps to identify points across the sequence of sparse 3D point clouds that belong to a detected road agent. The system generates a dense 3D point cloud by combining at least some of those identified points. The system detects one or more lane markings and projects them into the dense 3D point cloud to generate an annotated 3D point cloud. The system analyzes the in-lane driving behavior of the detected road agent based, at least in part, on the annotated 3D point cloud.

Abstract: A method for autonomously tethering an autonomous electric powered trailer to a tethering partner includes identifying a target tethering partner for the autonomous electric powered (AEP) trailer, detecting a likely n-dimensional position of a tethering nexus of the target tethering partner based on an assessment of sensor data; computing, a set of automated tethering instructions based on the likely n-dimensional position of the tethering nexus; and autonomously tethering, via the one or more computers, a coupler of the AEP trailer to the tethering nexus based on executing the set of automated tethering instructions.

Abstract: A system and corresponding method for autonomous driving of a vehicle are provided. The system comprises at least one neural network (NN) that generates at least one output for controlling the autonomous driving. The system further comprises a main data path that routes bulk sensor data to the at least one NN and a low-latency data path with reduced latency relative to the main data path. The low-latency data path routes limited sensor data to the at least one NN which, in turn, employs the limited sensor data to improve performance of the at least one NN's processing of the bulk sensor data for generating the at least one output. Improving performance of the at least one NN's processing of the bulk sensor data enables the system to, for example, identify a safety hazard sooner, enabling the autonomous driving to divert the vehicle and avoid contact with the safety hazard.

Abstract: A system for determining a distance between a camera and a hitch ball of a vehicle. The system includes a camera configured to be mounted on a vehicle and an electronic processor connected to the camera. The electronic processor is configured to determine a first position and a second position associated with a point on a trailer, determine an angle of rotation between the first position and the second position, and using the first position, the second position, and the angle of rotation, determine a null space of a matrix. In one example, the matrix is A = [ 1 0 0 - cu xo cl x 0 1 0 - cu y ? 0 cl y 0 0 0 cu z ? 0 - cl z 0 0 1 - cu z ? 0 cu z ? 1 ? cu x ? 1 cl x - cos ? ? sin ? ? 1 0 0 ] and the null space of the matrix includes a value of a distance (dl) between the camera and the hitch ball.

Abstract: Controlling an unmanned aerial vehicle to traverse a portion of an operational environment of the unmanned aerial vehicle may include obtaining an object detection type, obtaining object detection input data, obtaining relative object orientation data based on the object detection type and the object detection input data, and performing an object avoidance operation based on the relative object orientation data. The object detection type may be monocular object detection, which may include obtaining the relative object orientation data by obtaining motion data indicating a change of spatial location for the unmanned aerial vehicle between obtaining the first image and obtaining the second image based on searching along epipolar lines to obtain optical flow data.

Abstract: An autonomous vehicle may determine that it has an amount of power remaining projected to be needed to reach a recharging point by autonomously traveling with predefined systems disabled. The vehicle may disable the predefined systems and travel towards the recharging point. Subsequent to the disabling, the autonomous vehicle may determine that it no longer has the amount of power remaining projected to be needed to reach the recharging point. The vehicle may then communicate with a server and request assistance. The vehicle may then travel to an instructed rendezvous point with a second autonomous vehicle, the rendezvous received from the server. The two vehicles may then communicate to allow the first vehicle to leverage a capability of the second autonomous vehicle, responsive to the first and second autonomous vehicles being within communication range, allowing the first vehicle to reach the recharging point via the leveraging.

Abstract: A method of controlling a steering system on a marine vessel includes, in response to receiving input to engage a quick steer mode, employing a reduced steering ratio to translate a change in position of a steering wheel to a desired change in steering angle of a marine drive. A vessel speed of the marine vessel is compared to a threshold vessel speed. Upon determining that the vessel speed exceeds the threshold vessel speed, an output limit in a direction of travel of the marine vessel is determined and the marine drive is controlled based on the output limit. A steering actuator associated with the marine drive is controlled based on the reduced steering ratio and the change in position of the steering wheel.

Abstract: Systems and methods are provided for navigating a host vehicle. At least one processing device may be programmed to receive an image of an environment of the host vehicle; detect, based on analysis of the image, a pedestrian crosswalk in the image; detect a presence of a traffic light and determine whether the traffic light is relevant to the host vehicle and the pedestrian crosswalk; determine a state of the traffic light; determine, when a pedestrian appears in the image, a proximity of the pedestrian relative to the pedestrian crosswalk; determine a planned navigational action for navigating the host vehicle relative to the pedestrian crosswalk based on a driving policy, the state of the traffic light and the proximity of the pedestrian relative to the pedestrian crosswalk; and cause one or more actuator systems of the host vehicle to implement the planned navigational action.

Abstract: Disclosed herein are methods for determining an estimated weight of a vehicle. The methods comprise operating at least one processor to: receive vehicle data associated with the vehicle, the vehicle data comprising a plurality of vehicle parameters collected during operation of the vehicle; identify one or more vehicle maneuvers based on the vehicle data, each vehicle maneuver being associated with a portion of the vehicle data; and use at least one machine learning model to determine the estimated weight of the vehicle based on the portion of the vehicle data associated with each of the one or more vehicle maneuvers, the at least one machine learning model trained using training data associated with a plurality of previous vehicle maneuvers. Also disclosed are systems for implementing methods of the present disclosure.

Abstract: A vehicular control system is presented where the contact patches of the vehicle have reduced impact on an environment. Vehicular control systems include force sensors and terrain sensors that provide real-time or near real-time information about the operating environment of a vehicle. The force sensor data may be used to derive one or more forces which can be used to infer the nature of the vehicles contact patches. As the vehicular control system detects changes in a terrain attribute from the terrain sensors, the vehicular control system determines what the contact patches should be to address the terrain change and determines the necessary forces to give rise to the target contact patches. The vehicular control systems may then adjust the operational parameter values of the vehicle to generate the target contact patches to thereby reduce the impact on the environment.