AeroAct: Action-Centered World-Action Models for Language-Conditioned Quadrotor Flight
Robotics
Summary
The authors introduce AeroAct, a new model that helps drones follow language instructions by predicting sequences of smooth flight actions based on what the drone sees and feels. Unlike previous methods that use simple commands or waypoints, AeroAct learns from future video frames during training to better anticipate the effects of its actions. They created a special system to gather data combining visuals, movement, and language, including a low-cost device to mimic drone flights. Tests in simulation and real flights show their method improves how well drones track targets and search for objects over time.
Authors
Xinhong Zhang, Qiyuan Zhu, Yubo Huang, Haolin Chen, Runqing Wang, Yuhao Mo, Zhongxin Chen, Yu Hu, Xinjiang Wang, Jian Sun, Gang Wang
Abstract
Language-conditioned quadrotor flight requires a policy to ground semantic goals, anticipate the visual consequences of ego-motion, and output control references that remain smooth and dynamically executable under rapidly changing first-person views. Existing aerial vision-language navigation and vision-language-action methods commonly use discrete actions, high-level waypoints, or instantaneous velocity commands, which provide limited supervision about how flight actions change future observations. We present AeroAct, an action-centered world-action model (WAM) for quadrotor navigation. To the best of our knowledge, AeroAct is the first WAM instantiated and demonstrated for real-world aerial flight. The model adapts a pretrained video diffusion Transformer to predict local trajectory-action chunks from egocentric visual history, proprioception, and language. Future first-person frames are used during training as dense consequence supervision, while deployment directly decodes actions without generating future video. To obtain aligned visual, state, language, and dynamically feasible action data, we build a DiffAero-based pipeline with complementary Isaac Lab and 3D Gaussian splatting renderers. We further introduce a low-cost handheld collection device that couples camera observations with motion estimates to recreate flight-like egocentric trajectories, and a self-guidance procedure that improves temporal consistency across overlapping trajectory chunks. Closed-loop simulation and real-world experiments show that temporal visual context improves target tracking and object-search performance, and that WAM-based policies can be executed on a physical quadrotor.