Flocks, herds and schools: A distributed behavioral model
SIGGRAPH '87 Proceedings of the 14th annual conference on Computer graphics and interactive techniques
Artificial intelligence: a modern approach
Artificial intelligence: a modern approach
| Hi-index | 0.00 |
Many methods have been proposed to simulate group movement in complex environments with static obstacles. A critical part of this kind of simulation is the path planning method, which is used to prevent the moving objects from colliding with obstacles, and allow the group to alter its course when they are trapped in local minima. However, there is little research on the path planning for dynamic environments in which there are many moving obstacles. In this paper, we investigate how to simulate group movement in dynamic environments with moving obstacles. To this end, a roadmap is constructed during preprocessing and dynamically updated to guide a group. A self-adaptive roadmap method and a hierarchical motion-planning mechanism are proposed to enhance the simulation performance when the roadmap needs to be updated. The novelty of our self-adaptive roadmap method is that the roadmap is dynamically updated based on the obtained path, and we only require the roadmap to be partially correct. The hierarchical motion-planning mechanism also reduces the computational cost of the reconnection process which is the bottleneck of the roadmap-based path planning. Experimental results show that proposed mechanisms perform well in dynamic environments.