Sola & Monin & Devy & Lemaire, "Undelayed initialization in bearing only SLAM"

Undelayed initialization in bearing only SLAM

UndelayedBOSLAM.pdf

Sola, J.   Monin, A.   Devy, M.   Lemaire, T.  
CNRS, Toulouse, France;

This paper appears in: Intelligent Robots and Systems, 2005. (IROS 2005). 2005 IEEE/RSJ International Conference on
Publication Date: 2-6 Aug. 2005
On page(s): 2499- 2504
ISBN: 0-7803-8912-3
INSPEC Accession Number: 8750433
Digital Object Identifier: 10.1109/IROS.2005.1545392
Current Version Published: 2005-12-05


ref.

SolaAbstract.pdf

http://homepages.laas.fr/jsola/JoanSola/eng/bearingonly.html




 기존 SLAM에서 쓰이는 레이저 레인지 스캐너 등 range and bearing 센서 대신 공간에 대한 풍부한 정보를 주는 카메라를 쓰면, 1차원 (인식된 물체까지의 거리 정보, depth)을 잃게 되어 bearing-only SLAM이 된다.

EKF requires Gaussian representations for all the involved random variables that form the map (the robot pose and all landmark's positions). Moreover, their variances need to be small to be able to approximate all the non linear functions with their linearized forms.

두 입력 이미지 프레임 사이에 baseline을 구할 수 있을 만큼 충분한 시점 차가 존재해야 랜드마크의 위치를 결정할 수 있으므로, 이를 확보하기 위한 시간이 필요하게 된다.

http://en.wikipedia.org/wiki/Structure_from_motion

1. Extract features from images
2. Find an initial solution for the structure of the scene and the motion of the cameras
3. Extend the solution and optimise it
4. Calibrate the cameras
5. Find a dense representation of the scene
6. Infer geometric, textural and reflective properties of the scene.

sequential probability ratio test
http://en.wikipedia.org/wiki/Sequential_probability_ratio_test
http://www.agrsci.dk/plb/bembi/africa/sampling/samp_spr.html
http://eom.springer.de/S/s130240.htm

EKF (extended Kalman filter) - inconsistency and divergence
GSF (Gaussian sum filter) - computation load
FIS (Federated Information Sharing)