Iterative closest point via particle filtering. This code is basically used for 2D/3D point set registration.
The registration process is modelled as a state estimation problem which can be solved efficiently via filtering techiniques such as UKF, CKF and particle filter. To understand the theory, the readers are referred to [1][2][3]. Also, it is good to cite these articles if you use this code.
Special thanks goes to Prof. Jihua ZHU, who provided me with the backbone code of particle filter; And also thanks goes to Dr. Liang LI, who discussed with me about technique details including motion alignment error and density proposal.
Make sure that you also add "LiDAR/CommonFunctions" to your matlab PATH, which provides functions relating to coordinate transformation.
[1] R. Sandhu, S. Dambreville and A. Tannenbaum, "Point Set Registration via Particle Filtering and Stochastic Dynamics," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 32, pp. 1459-1473, 2010.
[2] 祝继华, 杜少毅, 李钟毓,等. 基于粒子滤波的部分对应点集刚体配准算法[J]. **科学:信息科学, 2014, 44(7):886-899.
[3] L. Li, M. Yang, C. Wang, and B. Wang, "Hybrid Filtering Framework based Robust Localization for Industrial Vehicles," IEEE Transactions on Industrial Informatics, vol. PP, p. 1-1, 2018.
If you have any problems, please feel free to contact me.
Di Wang
Laboratory of Visual Cognitive Computing and Intelligent Vehicle
Institute of Artificial Intelligence and Robotics
Xi'an Jiaotong University
Xi'an, Shaanxi, 710049, P.R. China