Flying by Inference: Active Inference World Models for Adaptive UAV Swarms

2026-04-30

AI Summary

This work addresses the computational efficiency and adaptability challenges in repeated trajectory planning for multi-UAV swarms operating in dynamic environments by proposing a hierarchical decision-making framework grounded in active inference. For the first time, active inference is integrated into UAV swarm control, leveraging an offline-constructed task-path-action dictionary to enable online fusion of symbolic task logic with continuous motion control—thereby achieving adaptive task allocation, path insertion, and obstacle-avoidance replanning without iterative optimization. The system combines genetic algorithm–generated expert data, probabilistic world model learning, KL-divergence–based anomaly detection, and hybrid filtering for state estimation. Simulations and real-world flight experiments demonstrate that, compared to an enhanced Q-learning baseline, the proposed method yields smoother, more stable behaviors and effectively corrects symbolic predictions corrupted by noisy observations.

Abstract

This paper presents an expert-guided active-inference-inspired framework for adaptive UAV swarm trajectory planning. The proposed method converts multi-UAV trajectory design from a repeated combinatorial optimization problem into a hierarchical probabilistic inference problem. In the offline phase, a genetic-algorithm planner with repulsive-force collision avoidance (GA--RF) generates expert demonstrations, which are abstracted into Mission, Route, and Motion dictionaries. These dictionaries are used to learn a probabilistic world model that captures how expert mission allocations induce route orders and how route orders induce motion-level behaviors. During online operation, the UAV swarm evaluates candidate actions by forming posterior beliefs over symbolic states and minimizing KL-divergence-based abnormality indicators with respect to expert-derived reference distributions. This enables mission allocation, route insertion, motion adaptation, and collision-aware replanning without rerunning the offline optimizer. Bayesian state estimators, including EKF and PF modules, are integrated at the motion level to improve trajectory correction under uncertainty. Simulation results show that the proposed framework preserves expert-like planning structure while producing smoother and more stable behavior than modified Q-learning. Additional validation using real-flight UAV trajectory data demonstrates that the learned world model can correct symbolic predictions under noisy and non-smooth observations, supporting its applicability to adaptive UAV swarm autonomy.

Problem

• UAV swarm
• trajectory planning
• adaptive autonomy
• collision avoidance
• uncertainty

Innovation

• active inference
• probabilistic world model
• hierarchical planning
• UAV swarm autonomy
• expert-guided learning