Abstract: A system comprises a processor and a memory storing instructions. The processor receives an image for processing using a reinforcement learning based agent comprising a neural network trained using a reward function. The image includes N lane lines of a roadway, where N is a positive integer. The instructions configure the processor to traverse the image using the agent at least N times from a first end of the image to a second end of the image by: incrementally moving the agent from a first side of the image to a second side of the image after each traversal; and maximizing rewards for the agent using the reward function during each traversal of the image using the agent. The instructions configure the processor to identify the N lane lines of the roadway as a single cluster of lane lines after traversing the image using the agent at least N times.

Abstract: A method of surveying roads includes generating a dynamic flight plan for a drone using a vehicle traveling on a road as a target. The dynamic flight plan includes instructions for movement of the drone. The method includes controlling the drone as a function of position of the vehicle based on the dynamic flight plan. The method includes maintaining, based on the controlling, line of sight with the drone while the drone with an onboard camera follows the vehicle and captures images of the road being traveled by the vehicle using the onboard camera.

Abstract: A system and method are provided for controlling a vehicle. In one embodiment, a system includes: a sensor system configured to generate sensor data sensed from an environment of the vehicle; and a control module configured to, by a processor, predict parked vehicles within the scene of the environment, identify an outer edge associated with the parked vehicles, generate a virtual lane line based on the outer edge associated with the parked vehicles, and generate signal data based on the virtual lane line to at least one of display the virtual lane line within the scene and control the vehicle.

Abstract: A system comprises a processor and a memory storing instructions. The processor receives an image for processing using a reinforcement learning based agent comprising a neural network trained using a reward function. The image includes N lane lines of a roadway, where N is a positive integer. The instructions configure the processor to traverse the image using the agent at least N times from a first end of the image to a second end of the image by: incrementally moving the agent from a first side of the image to a second side of the image after each traversal; and maximizing rewards for the agent using the reward function during each traversal of the image using the agent. The instructions configure the processor to identify the N lane lines of the roadway as a single cluster of lane lines after traversing the image using the agent at least N times.

Abstract: A method of surveying roads includes generating a dynamic flight plan for a drone using a vehicle traveling on a road as a target. The dynamic flight plan includes instructions for movement of the drone. The method includes controlling the drone as a function of position of the vehicle based on the dynamic flight plan. The method includes maintaining, based on the controlling, line of sight with the drone while the drone with an onboard camera follows the vehicle and captures images of the road being traveled by the vehicle using the onboard camera.

Abstract: Methods and apparatus are provided for 3D object bounding for 2D image data for use in an assisted driving equipped vehicle. In various embodiments, an apparatus includes a camera operative to capture a two dimensional image of a field of view, a lidar operative to generate a point cloud of the field of view, a processor operative to generate a three dimensional representation of the field of view in response to the point cloud, to detect an object within the three dimensional representation, to generate a three dimensional bounding box in response to the object, to project the three dimensional bounding box onto the two dimensional image to generate a labeled two dimensional image, and a vehicle controller to controlling a vehicle in response to the labeled two dimensional image.

Abstract: Systems and method are provided for training an autonomous vehicle. In various embodiments, a method includes: storing, in a data storage device, real world data including a sequence of images of a road environment, the sequence of images generated based on a vehicle traversing the road environment; processing, in an offline simulation environment, the sequence of images with a deep reinforcement learning agent associated with a control feature of the autonomous vehicle to obtain an optimized set of control policies; and training the autonomous vehicle based on the optimized set of control polices.

Abstract: The present application generally relates to a method and apparatus for generating an action policy for controlling an autonomous vehicle. In particular, the method and apparatus are operative for determining a following distance between a host vehicle and a lead vehicle and a lead vehicle speed, generating a lane change request in response to the following distance, a host vehicle speed and the lead vehicle speed, determining an available lane in response to an image and the lane change request, generating a lane change command in response to the available lane, generating a control signal in response to the lane change request and a map data and controlling the vehicle to execute a lane change action in response to the control signal.