Adaptive formation control and transformation in virtual crowds via deep reinforcement learning

Adaptive formation control and transformation in virtual crowds via deep reinforcement learning

연구 분야: Software Development

학회: The Visual Computer

In the field of computer graphics, crowd simulation is crucial for constructing realistic virtual environments. This paper introduces a novel approach to adaptive formation control and transformation in virtual crowds using deep reinforcement learning. Our method enables virtual agents to autonomously generate, maintain, and adaptively transform formations relying solely on local perception and interaction. The proposed framework comprises three key components: dynamic formation configuration based on alignment, a formation control algorithm embedded with formation constraints, and an adaptive formation transformation mechanism. Simulation results demonstrate that our method achieves formation errors around 5% in formation tasks, surpassing existing approaches in both convergence speed and formation control precision. It further exhibits rapid, stable, flexible, and highly scalable formation control capabilities, offering a novel paradigm for creating realistic and environmentally adaptive crowd animations. Code and trained policies for this paper are at https://github.com/SYDDX/AFCTVC-DRL.