Abstract: Embodiments herein include an oversized automated vehicle having a trailer portion and tractor portion performing a J-hook turn at an intersection. The automated vehicle includes an autonomy system for gathering and processing various types of data. The autonomy system generates a candidate trajectory, including the trailer path and the tractor path performing a J-hook turn, and generates driving instructions for the automated vehicle according to a satisfactory candidate trajectory. The autonomy system applies an optimization function for iteratively generating and modifying the candidate trajectory. Using the map data for the intersection, the optimization function modifies the candidate trajectory, such that if the tractor or the trailer were to drive along the predicted tractor path and trailer path, then the tractor and the trailer would perform the J-hook turn while fitting within the drivable surface boundary and inner boundaries (e.g., driving lanes).

Abstract: Embodiments include systems and methods for determining states of traffic lights and managing behavior of an automated vehicle approaching an intersection. An autonomy system applies an object recognition engine trained to recognize a traffic light, and identify and confirm the state of the traffic light. A first neural network trained for object detection recognizes a traffic light and defines a bounding box around the recognized traffic light. A second neural network receives the region of the image bounded by the box as constituting the traffic light and “reads” the light. The automated vehicle uses other information, such as states of pedestrian traffic lights, detection of objects in and near the intersection, and glare on one or more cameras, to supplement its determination of the right of way through the intersection. The autonomy system generates a driving instruction based on the traffic light combined.

Abstract: Embodiments herein include an oversized automated vehicle having a trailer portion and tractor portion performing a J-hook turn at an intersection. The automated vehicle includes an autonomy system for gathering and processing various types of data. The autonomy system generates a candidate trajectory, including the trailer path and the tractor path performing a J-hook turn, and generates driving instructions for the automated vehicle according to a satisfactory candidate trajectory. The autonomy system applies an optimization function for iteratively generating and modifying the candidate trajectory. Using the map data for the intersection, the optimization function modifies the candidate trajectory, such that if the tractor or the trailer were to drive along the predicted tractor path and trailer path, then the tractor and the trailer would perform the J-hook turn while fitting within the drivable surface boundary and inner boundaries (e.g., driving lanes).

Abstract: Embodiments herein include an oversized automated vehicle having a trailer portion and tractor portion performing a J-hook turn at an intersection. The automated vehicle includes an autonomy system for gathering and processing various types of data. The autonomy system generates a candidate trajectory, including the trailer path and the tractor path performing a J-hook turn, and generates driving instructions for the automated vehicle according to a satisfactory candidate trajectory. The autonomy system applies an optimization function for iteratively generating and modifying the candidate trajectory. Using the map data for the intersection, the optimization function modifies the candidate trajectory, such that if the tractor or the trailer were to drive along the predicted tractor path and trailer path, then the tractor and the trailer would perform the J-hook turn while fitting within the drivable surface boundary and inner boundaries (e.g., driving lanes).

Abstract: Embodiments herein include an autonomy system of an automated vehicle performing operations for detecting and avoiding lazy or unpredictable vehicles on a roadway. The autonomy system generates bounding boxes for tracking traffic vehicles recognized in sensor data. For close vehicles, the autonomy system generates expanded bounding boxes. The autonomy system determines the close vehicle is an encroaching vehicle by detecting that the expanded bounding box of the close vehicle overlapped the middle lane line or enters into the automated vehicle's current lane of travel. The autonomy system may simulate the behavior of the close vehicle by forward-propagating predicted positions of the close vehicle and expanded bounding box over some future time. In response to detecting the encroaching vehicle, the autonomy system may determine whether to perform an avoidance action or select a particular avoidance action, such as biasing away from the encroaching vehicle.

Abstract: Embodiments include systems and methods for determining states of traffic lights and managing behavior of an automated vehicle approaching an intersection. An autonomy system applies an object recognition engine trained to recognize a traffic light, and identify and confirm the state of the traffic light. A first neural network trained for object detection recognizes a traffic light and defines a bounding box around the recognized traffic light. A second neural network receives the region of the image bounded by the box as constituting the traffic light and “reads” the light. The automated vehicle uses other information, such as states of pedestrian traffic lights, detection of objects in and near the intersection, and glare on one or more cameras, to supplement its determination of the right of way through the intersection. The autonomy system generates a driving instruction based on the traffic light combined.