Aircraft landing pose estimation method based on vision-inertia tight coupling

Abstract

The invention provides an aircraft landing pose estimation method based on vision-inertia tight coupling. The method comprises the steps of: an actual feature point detection process: acquiring video,enhancing an image, detecting a target and outputting a feature point, a synthesized feature point generation process: reading a pose parameter of an inertia measurement unit, reading geographic information of an airport landmark point and calculating a synthesized feature point, and a relative pose solution process: reading the actual feature point and the synthesized feature point, reading thegeographic information of the airport landmark point, and calculating a relative pose between an aircraft and a landing platform.

Description

technical field [0001] The invention relates to the field of integrated navigation, in particular to an aircraft landing pose estimation method based on visual-inertial tight coupling. Background technique [0002] Traditional aircraft approach and landing navigation often use Kalman filter (KF), PnP method or nonlinear optimization (ULO). When the visual measurement model constructed is a nonlinear time-varying system, even if extended Kalman filter (EKF) (PF) is also difficult to significantly improve the filtering accuracy, and the calculation time is longer; the PnP algorithm is greatly affected by the accuracy of target detection in the image, and depends on the number of image features. When the number of image features is small (such as <5) pose The solution accuracy is poor; the nonlinear optimization method finds the optimal solution through multiple iterations to achieve the minimum value of the objective function, and its calculation time is uncertain, which ca...

AI Technical Summary

Problems solved by technology

[0002] Traditional aircraft approach and landing navigation often use Kalman filter (KF), PnP method or nonlinear optimization (ULO). When the visual measurement model constructed is a nonlinear time-varying system, even if extended Kalman filter (EKF) (PF) is also difficult to significantly improve the filtering accuracy, and the calculation time is longer; the PnP algorithm is greatly affected by the accuracy of target detection in the image, and depends on the number of image features. When the number of image features is small (such as <5) pose The solution accuracy is poor; the nonlinear optimization method finds the optimal solution through multiple iterations to achieve the minimum value of the objective function, and its calculation time is uncertain, which cannot meet the strong real-time requirements of airborne applications

Images