Search papers, labs, and topics across Lattice.
2
0
3
9
UAV swarms can now adapt to changing conditions and replan trajectories in real-time by inferring expert-like behaviors from a learned probabilistic world model, avoiding computationally expensive re-optimization.
Active Inference offers a scalable and cognitively grounded alternative to reinforcement learning for UAV swarm control, showing faster convergence and improved stability in dynamic environments.
