Abstract: Techniques for optimizing flight safety operations for an aircraft are described. In operation, aircraft operational parameters corresponding to a plurality of flight operations are retrieved. The aircraft operational parameters are then analyzed using a first machine learning model to identify a first flight operation, where the first flight operation comprises at least one aircraft operational parameter with deviation beyond a threshold. At least one potential flight safety incident corresponding to the first flight operation is then identified using the at least one aircraft operational parameter. The at least one potential flight safety incident is then analyzed using a second machine learning model to identify a corrective action for the potential flight safety incident, where the second machine learning model is trained using flight safety artifacts comprising a plurality of flight safety incidents and corrective actions to be initiated in response to the plurality of flight safety incidents.

Abstract: Systems, methods, and other embodiments described herein relate to improving identification of a path for an ego vehicle on a roadway. In one embodiment, a method includes, in response to acquiring sensor data from at least one sensor of the ego vehicle about a surrounding environment, identifying roadway elements from the sensor data as cues about the path. The roadway elements include one or more of lane markers of the roadway and surrounding vehicles. The method includes grouping the roadway elements into two or more groups according to characteristics of roadway elements indicating common curvatures. The method includes analyzing the two or more groups according to a confidence heuristic to determine a priority group from the two or more groups that corresponds with a trajectory of the ego vehicle. The method includes providing an identifier for the priority group to facilitate at least path planning for the ego vehicle.

Abstract: System, methods, and other embodiments described herein relate to improving identification of a path for an ego vehicle on a roadway. In one embodiment, a method includes, in response to acquiring sensor data from at least one sensor of the ego vehicle about a surrounding environment, identifying roadway elements from the sensor data as cues about the path. The roadway elements include one or more of lane markers of the roadway and surrounding vehicles. The method includes grouping the roadway elements into two or more groups according to characteristics of roadway elements indicating common curvatures. The method includes analyzing the two or more groups according to a confidence heuristic to determine a priority group from the two or more groups that corresponds with a trajectory of the ego vehicle. The method includes providing an identifier for the priority group to facilitate at least path planning for the ego vehicle.

Abstract: A lane identity in a roadway in which a vehicle is traveling can be determined. The roadway can include a plurality of lanes. The determining can include generating a lane identification confidence belief indicating a probability that the vehicle is in a particular lane of the plurality of lanes of the roadway. Generating the lane identification confidence belief can be based on: any detected lane crossings, the number of lanes in the roadway, the lane marker type to a left side and to a right side of the vehicle at the current position of the vehicle or ahead of the current position of the vehicle in a forward direction of travel of the vehicle, and a weighted average of an instantaneous lane identification confidence belief and a lane identification confidence belief prior to a current sample time period.

Abstract: Lane feature data is processed to compute a feed-forward lane curvature of a left lane boundary and a right lane boundary. A look-ahead lane width and a near lane width are computed based on left and right lane boundaries. A lane width increase is computed to detect a lane split or lane merge based on differences between increasing lane widths. A side of the vehicle on which the lane split or merge occurred is identified or determined. The lane boundary on the side on which the lane split or merged occurred is ignored, and a single-sided lane centering calculation is performed based on the non-ignored lane boundary.

Abstract: An automated vehicle can generate a path. Sensor data of at least a forward portion of an external environment of the vehicle can be acquired. One or more travel lane markers of a current travel path of the automated vehicle can be identified from the acquired sensor data. A detection range of the acquired sensor data can be determined. A set of travel lane parameters based on the identified one or more travel lane markers can be determined. In at least some instances, the set of travel lane parameters can include at least: offset, curvature, curvature derivative, and yaw. It can be determined whether any of the determined travel lane parameters are reliable based on the determined detection range. In one or more arrangements, a vehicle path can be determined based at least partially on the travel lane parameters that are determined as being reliable.

Abstract: An automated vehicle can generate a path. Sensor data of at least a forward portion of an external environment of the vehicle can be acquired. One or more travel lane markers of a current travel path of the automated vehicle can be identified from the acquired sensor data. A detection range of the acquired sensor data can be determined. A set of travel lane parameters based on the identified one or more travel lane markers can be determined. In at least some instances, the set of travel lane parameters can include at least: offset, curvature, curvature derivative, and yaw. It can be determined whether any of the determined travel lane parameters are reliable based on the determined detection range. In one or more arrangements, a vehicle path can be determined based at least partially on the travel lane parameters that are determined as being reliable.

Abstract: A lane identity in a roadway in which a vehicle is traveling can be determined. The roadway can include a plurality of lanes. The determining can include generating a lane identification confidence belief indicating a probability that the vehicle is in a particular lane of the plurality of lanes of the roadway. Generating the lane identification confidence belief can be based on: any detected lane crossings, the number of lanes in the roadway, the lane marker type to a left side and to a right side of the vehicle at the current position of the vehicle or ahead of the current position of the vehicle in a forward direction of travel of the vehicle, and a weighted average of an instantaneous lane identification confidence belief and a lane identification confidence belief prior to a current sample time period.

Abstract: A method and apparatus for determining a lane identity of a vehicle travelling in a roadway where the roadway contains a plurality of lanes is determined based on information from sensors associated with the vehicle and map data for detecting the lane marker type to a left side and right side of the vehicle, any lane crossings during travel of the vehicle along the roadway, a history of lane marker type up to a predetermined travel distance along the roadway, and the lane marker type at the instantaneous position of the vehicle in the roadway. The lane identity method and apparatus outputs a confidence belief identifying the probability that the vehicle is traveling in one particular lane of the roadway.

Abstract: A method and apparatus to predict the most probable path of a vehicle along a roadway which may contain a junction splitting the roadway into a rootpath and one or more sub-paths, each containing one or more lanes. The method and apparatus also process vehicle sensor information pertaining to driver corrections, road boundary detection, lane marker detection, map information availability, and road surface quality at the current position of the vehicle as well as from previous passes of the vehicle past the same location and other vehicles passing the same location to provide a warning to the driver to retake control of the vehicle if information used by the automatic vehicle control will become unavailable a predetermined travel distance ahead of the current position of the vehicle.

Abstract: Lane feature data is processed to compute a feed-forward lane curvature of a left lane boundary and a right lane boundary. A look-ahead lane width and a near lane width are computed based on left and right lane boundaries. A lane width increase is computed to detect a lane split or lane merge based on differences between increasing lane widths. A side of the vehicle on which the lane split or merge occurred is identified or determined. The lane boundary on the side on which the lane split or merged occurred is ignored, and a single-sided lane centering calculation is performed based on the non-ignored lane boundary.

Abstract: A method and apparatus for determining a lane identity of a vehicle travelling in a roadway where the roadway contains a plurality of lanes is determined based on information from sensors associated with the vehicle and map data for detecting the lane marker type to a left side and right side of the vehicle, any lane crossings during travel of the vehicle along the roadway, a history of lane marker type up to a predetermined travel distance along the roadway, and the lane marker type at the instantaneous position of the vehicle in the roadway. The lane identity method and apparatus outputs a confidence belief identifying the probability that the vehicle is traveling in one particular lane of the roadway.

Abstract: A method and apparatus to predict the most probable path of a vehicle along a roadway which may contain a junction splitting the roadway into a rootpath and one or more sub-paths, each containing one or more lanes. The method and apparatus also process vehicle sensor information pertaining to driver corrections, road boundary detection, lane marker detection, map information availability, and road surface quality at the current position of the vehicle as well as from previous passes of the vehicle past the same location and other vehicles passing the same location to provide a warning to the driver to retake control of the vehicle if information used by the automatic vehicle control will become unavailable a predetermined travel distance ahead of the current position of the vehicle.

Abstract: A platoon model allows improved prediction of preceding vehicle future state. In this context, the preceding vehicle is a vehicle immediately ahead of the host vehicle, and the dynamic state of the preceding vehicle was predicted based on data received from one or more vehicles in the platoon. The intelligent driver model (IDM) was extended to model car-following dynamics within a platoon. A parameter estimation approach may be used to estimate the model parameters, for example to adapt to different driver types. An integrated approach including both state prediction and parameter estimation was highly effective.

Abstract: A platoon model allows improved prediction of preceding vehicle future state. In this context, the preceding vehicle is a vehicle immediately ahead of the host vehicle, and the dynamic state of the preceding vehicle was predicted based on data received from one or more vehicles in the platoon. The intelligent driver model (IDM) was extended to model car-following dynamics within a platoon. A parameter estimation approach may be used to estimate the model parameters, for example to adapt to different driver types. An integrated approach including both state prediction and parameter estimation was highly effective.