AirDreamer: Generalist Drone Navigation with World Models
2026-06-02 • Robotics
RoboticsArtificial Intelligence
AI summaryⓘ
The authors created a new way for drones to find their way in places they've never been before. Instead of using fixed rules that only work well in known environments, their method learns by trying things out and understanding the surroundings better. They tested it both in simulations and with real drones, and it worked better than previous methods, especially in tricky situations. Their approach also works well when moving from simulated tests to the real world without extra adjustments.
drone navigationreinforcement learningworld modelenvironment generalizationsim-to-real transfersparse rewardlocal minimarobot autonomynavigation success rate
Authors
Zian Liu, Andong Yang, Chunkai Yang, Ruidong An, Chao Gao, Guyue Zhou
Abstract
Navigating a drone in unseen and cluttered environments requires reliable generalization to unseen scene layouts and understanding of environmental structure relative to the robot's capabilities. Previous methods, which assume the same environment configuration, often rely heavily on human-designed perception pipelines and predefined rules to guide the robot toward the target. This process is environment-dependent and generalizes poorly across environments. Inspired by animal navigation behavior, we design a navigation framework that navigates with a reinforcement-learning-based policy on top of a world-model-based environment understanding to overcome these issues. In addition, a sparse reward function without hand-crafted shaping terms is designed to avoid local minima traps and encourage yaw control behaviors. In simulation and on real drones, our method exhibits emergent capabilities for navigating complex, unseen environments and escaping local optima where other methods fail. In challenging maps, it achieves a 5.3% higher navigation success rate than best baseline. Furthermore, the proposed framework achieves effective sim-to-real transfer without any tuning during deployment. The code will be publicly available.