Scene matching based visual SLAM navigation for small unmanned aerial vehicle

A scene matching based visual SLAM (simultaneous localization and mapping) navigation algorithm is proposed for SUAV (small unmanned aerial vehicle) which described by EKF (Extended Kalman Filtering). Firstly, a scene matching method with weighted Hausdorff distance was introduced for waypoints accurate abstraction. On this foundation, the SUAV´s nonlinear state model was analyzed to establish nonlinear relationship model between the measurement and the waypoints, and then the state of the model was predicted and estimated to deal with data association and extend the state for new waypoints. Last, through continuously predicting and estimating, the algorithm located the SUAV accurately by visual information. Simulation results show that the proposed algorithm could effectively reduce the estimation error of navigation system and improves the positioning accuracy for SUAV.