Title of Presentation: Coupled Vision and Inertial Navigation for Pin-Point Landing
Primary (Corresponding) Author: Andrew Johnson
Organization of Primary Author: NASA Jet Propulsion Laboratory
Co-Authors: A. Mourikis,
Abstract: Our research task in the Mars Technology Program focuses on terrain relative state estimation during planetary descent to enable pin-point landing. In this task we have developed estimation algorithms that fuse inertial and visual measurements as well as computer vision algorithms that automatically extract landmark locations and terrain feature tracks through a sequence of images. These algorithms have been tested extensively with simulated data and data collected during multiple field test campaigns on a variety of platforms.
Recently we have developed an algorithm that automatically produces 2D-to-3D correspondences between descent images and a surface map (mapped landmarks) and 2D-to-2D correspondences through a sequence of descent images (persistent features). These correspondences are input into a Kalman filter to estimate position, velocity and attitude in the map-based reference frame. The filter tightly couples IMU and camera measurements in a resource-adaptive and hence real-time capable fashion. Results from a sounding rocket test, covering the dynamic profile of typical planetary landing scenarios, show estimation errors of magnitude 0.16 m/s in velocity and 7.6 m in position at touchdown. These results vastly improve current state of the art and meet the requirements of future planetary exploration missions.