Perception-based autonomous landing for aircraft

An apparatus, comprising:

• a memory configured to store computer-readable program code;
• and
• processing circuitry configured to access the memory, and execute the computer-readable program code to cause the apparatus to at least:
• receive a sequence of images of an airfield, captured by a camera onboard an aircraft approaching a runway;
• and for at least an image of the sequence of images,
• apply the image to a machine learning model trained to perform an object detection and segmentation in which at least one of the runway on the airfield, or a runway marking on the runway, is detected in the image, and in which a mask is produced that comprises a segment of pixels of the image that are assigned to an object class for the runway or the runway marking;
• apply the mask to a corner detector to detect one or more interest points on the mask and thereby the runway or the runway marking in the image;
• match the one or more interest points on the runway or the runway marking in the image, to one or more corresponding points on the airfield or the runway marking that have known runway-framed local coordinates;
• and
• perform a perspective-n-point estimation, using the one or more interest points and the known runway-framed local coordinates, to determine a pose estimate of the camera and thereby the aircraft relative to the runway or the runway marking, wherein the pose estimate is to be utilized in at least one of guidance or control of the aircraft.