A Unified Strategy for Landing and Docking using Spherical flow Divergence

Select |




Print


McCarthy, Chris; Barnes, Nick


2012-01-15


Journal Article


IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI)


34


5


1024-1031


We present a novel approach to landing and docking using time-to-contact estimates taken from optical flow divergence. Previous approaches have considered the use of optical flow for docking or landing, however, there exists no strategy that works for any angle of approach (or descent), under general motion. To this end, we propose a single unified strategy for performing controlled approaches to planar surfaces of arbitrary orientation. Central to this is the examination of optical flow field divergence under spherical projection, and use of the point of maximum divergence on the view sphere. We provide a theoretical justification for estimating τ under spherical divergence, and derive both velocity and heading control schemes based on the location and value of the max-div point. The proposed control scheme allows time-to-contact (τ ) to be measured for an arbitrary angle of approach, and requires no knowledge of surface orientation, or egomotion. The proposed landing strategy is tested in simulation, over real image sequences, and in closedloop control of a docking/landing manoeuvre.


http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=6133290&url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D6133290


nicta:4025


McCarthy, Chris; Barnes, Nick. A Unified Strategy for Landing and Docking using Spherical flow Divergence. IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI). 2012-01-15; 34(5):1024-1031. http://hdl.handle.net/102.100.100/101708?index=1



Loading citation data...

Citation counts
(Requires subscription to view)