28 October 2006 An autonomous navigation method for intelligent virtual agent in virtual urban environment
Author Affiliations +
Proceedings Volume 6421, Geoinformatics 2006: Geospatial Information Technology; 642102 (2006) https://doi.org/10.1117/12.712588
Event: Geoinformatics 2006: GNSS and Integrated Geospatial Applications, 2006, Wuhan, China
Abstract
This paper presents a dual approach of autonomous navigation for intelligent virtual agent in large complex virtual urban environments. A hierarchical global road map of the environment for global path planning is precomputed at first by using of the constrained Delauny triangulation algorithm to partition the free space, which takes into consideration the uneven feature of terrain. The accuracy of the road map may be adjusted by resizing the length of constrained segment. At runtime, A* algorithm is performed to automatically computes a collision-free and constrained path between two specified locations, and then the local navigation is realized by obtaining the environmental information immediately through visual image based 3D virtual vision. Experimental results demonstrate that the approach can provide satisfying global path and real-time obstacle avoidance effect.
© (2006) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Qing Zhu, Qing Zhu, Litao Han, Litao Han, Shenghua Xu, Shenghua Xu, Yeting Zhang, Yeting Zhang, } "An autonomous navigation method for intelligent virtual agent in virtual urban environment", Proc. SPIE 6421, Geoinformatics 2006: Geospatial Information Technology, 642102 (28 October 2006); doi: 10.1117/12.712588; https://doi.org/10.1117/12.712588
PROCEEDINGS
7 PAGES


SHARE
Back to Top