Cargando…

Enhanced ICP for the Registration of Large-Scale 3D Environment Models: An Experimental Study

One of the main applications of mobile robots is the large-scale perception of the outdoor environment. One of the main challenges of this application is fusing environmental data obtained by multiple robots, especially heterogeneous robots. This paper proposes an enhanced iterative closest point (I...

Descripción completa

Detalles Bibliográficos
Autores principales: Han, Jianda, Yin, Peng, He, Yuqing, Gu, Feng
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2016
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4801604/
https://www.ncbi.nlm.nih.gov/pubmed/26891298
http://dx.doi.org/10.3390/s16020228
Descripción
Sumario:One of the main applications of mobile robots is the large-scale perception of the outdoor environment. One of the main challenges of this application is fusing environmental data obtained by multiple robots, especially heterogeneous robots. This paper proposes an enhanced iterative closest point (ICP) method for the fast and accurate registration of 3D environmental models. First, a hierarchical searching scheme is combined with the octree-based ICP algorithm. Second, an early-warning mechanism is used to perceive the local minimum problem. Third, a heuristic escape scheme based on sampled potential transformation vectors is used to avoid local minima and achieve optimal registration. Experiments involving one unmanned aerial vehicle and one unmanned surface vehicle were conducted to verify the proposed technique. The experimental results were compared with those of normal ICP registration algorithms to demonstrate the superior performance of the proposed method.