Multi-Agent-Based Simulation of Archaeological Mobility in Uneven Landscapes

Researchers developed a multi-agent simulation framework using reinforcement learning to model archaeological mobility patterns in complex terrain. The system combines global path planning with local adaptation to simulate human and animal movement in historical landscapes, demonstrated through pursuit scenarios and transport analysis.

• →Multi-agent framework integrates reinforcement learning for adaptive navigation in archaeological simulations.
• →System processes real-world elevation data into high-fidelity 3D environments with terrain constraints.
• →Framework models diverse agent types including human groups and animal transport systems with empirical parameters.
• →Hybrid navigation strategy provides computationally efficient solution for large-scale dynamic simulations.
• →Results demonstrate significant impact of terrain morphology and agent heterogeneity on movement outcomes.