Abstract: A system for localization of a safe landing zone comprises at least one image-capture device onboard an aerial vehicle, and an onboard processor coupled to the image-capture device. The processor is operative to execute instructions to perform a method that comprises: receive, from the image-capture device, two or more overlapping images of a landscape underneath the aerial vehicle; generate, based on the overlapping images, a landing zone heatmap of the landscape; identify, based on the landing zone heatmap, one or more regions of the landscape having potential landing zones and obstacles; and determine a location of a safe landing zone using a distance transform of the one or more regions of the landscape. The location of the safe landing zone is in an area within one of the potential landing zones that is farthest from the obstacles. The location of the safe landing zone is then stored in a database.

Abstract: Disclosed are methods and systems for generating three-dimensional reconstructions of environments. A system, for example, may include a housing having an image sensor directed in a first direction and a distance sensor directed in a second direction and a control unit including a processor and a memory storing instructions. The processor may be configured to execute the instructions to: generate a first 3D model of an environment; generate a plurality of revolved 3D models by revolving the first 3D model relative to the image sensor to a plurality of positions within a predetermined angular range; match a set of distance values to one of the revolved 3D models; determine an angular position of the second direction relative to the first direction; and generate a 3D reconstruction of the environment.

Abstract: A method for inertial sensor error modeling and compensation comprises obtaining multiple bias drift datasets for an elapsed time period for one or more gyroscopes; generating a 3D bias drift data plot using the multiple bias drift datasets; generating a partial bias drift data image based on the 3D bias drift data plot; and inputting the partial bias drift data image into a machine learning algorithm to predict how bias drift evolves over time for the gyroscopes. The machine learning algorithm uses the partial bias drift data image, the elapsed time period, and temperature history to compute a predicted bias over temperature with respect to time, to thereby predict bias drift data for a future time period for the gyroscopes. The machine learning algorithm outputs a completed bias drift image that represents drift data from the elapsed time period and the predicted bias drift data for the future time period.

Abstract: A method for inertial sensor error modeling and compensation comprises obtaining multiple bias drift datasets for an elapsed time period for one or more gyroscopes; generating a 3D bias drift data plot using the multiple bias drift datasets; generating a partial bias drift data image based on the 3D bias drift data plot; and inputting the partial bias drift data image into a machine learning algorithm to predict how bias drift evolves over time for the gyroscopes. The machine learning algorithm uses the partial bias drift data image, the elapsed time period, and temperature history to compute a predicted bias over temperature with respect to time, to thereby predict bias drift data for a future time period for the gyroscopes. The machine learning algorithm outputs a completed bias drift image that represents drift data from the elapsed time period and the predicted bias drift data for the future time period.

Abstract: Disclosed are methods and systems for generating three-dimensional reconstructions of environments. A system, for example, may include a housing having an image sensor directed in a first direction and a distance sensor directed in a second direction and a control unit including a processor and a memory storing instructions. The processor may be configured to execute the instructions to: generate a first 3D model of an environment; generate a plurality of revolved 3D models by revolving the first 3D model relative to the image sensor to a plurality of positions within a predetermined angular range; match a set of distance values to one of the revolved 3D models; determine an angular position of the second direction relative to the first direction; and generate a 3D reconstruction of the environment.

Abstract: A system for localization of a safe landing zone comprises at least one image-capture device onboard an aerial vehicle, and an onboard processor coupled to the image-capture device. The processor is operative to execute instructions to perform a method that comprises: receive, from the image-capture device, two or more overlapping images of a landscape underneath the aerial vehicle; generate, based on the overlapping images, a landing zone heatmap of the landscape; identify, based on the landing zone heatmap, one or more regions of the landscape having potential landing zones and obstacles; and determine a location of a safe landing zone using a distance transform of the one or more regions of the landscape. The location of the safe landing zone is in an area within one of the potential landing zones that is farthest from the obstacles. The location of the safe landing zone is then stored in a database.

Abstract: The present disclosure provides computer vision systems and methods for an aircraft. The computer vision systems and methods identify a runway based on video frame data from a camera mounted to the aircraft, extract runway pixel positions associated with the runway, determine the aircraft position in a real world coordinate frame based on the pixel positions, the aircraft position including an aircraft lateral position and an aircraft elevation, receive predefined aircraft position data and go-around rules for a compliant approach to the runway, calculate an aircraft position deviation based on the aircraft position and the predefined aircraft position data, determine whether the aircraft position deviation is in conformance with the go-around rules, and output an indication of conformity with the go-around rules.

Abstract: The present disclosure provides computer vision systems and methods for an aircraft. The computer vision systems and methods identify a runway based on video frame data from a camera mounted to the aircraft, extract runway pixel positions associated with the runway, determine the aircraft position in a real world coordinate frame based on the pixel positions, the aircraft position including an aircraft lateral position and an aircraft elevation, receive predefined aircraft position data and go-around rules for a compliant approach to the runway, calculate an aircraft position deviation based on the aircraft position and the predefined aircraft position data, determine whether the aircraft position deviation is in conformance with the go-around rules, and output an indication of conformity with the go-around rules.