🤖 AI Summary
This work addresses the performance gap observed when deploying multi-waypoint drone controllers from simulation to real-world platforms under environmental disturbances. To bridge this sim-to-real discrepancy, the authors propose a hierarchical terminal control architecture that decouples smooth approach trajectory generation, continuous disturbance compensation, and supervised near-target regulation, while separating the core controller structure from platform-specific tuning. The system is validated through a three-stage pipeline—from PyBullet and PX4/Gazebo simulations to physical deployment on a Tello drone—achieving a mean late-phase position error of 0.024 meters under stochastic wind disturbances. Integrating a cascaded flight control stack, Vicon-based high-precision motion capture, and dual evaluation criteria (Strict/Grace), the framework substantially enhances cross-scenario robustness and transfer reliability.
📝 Abstract
Disturbance-robust UAV position control is easy to demonstrate in benign simulations but much harder to make fast in approach, well behaved near the target, and credible beyond a single benchmark. This letter presents a layered terminal-control architecture for multi-waypoint UAV position regulation together with a staged evaluation across PyBullet, PX4/Gazebo, and hardware. Phase I uses a PyBullet benchmark with stochastic wind for rapid structural selection, identifying a controller core that separates smooth approach generation, persistent-bias compensation, and supervised near-target terminal regulation. Phase II carries only that main architecture into a more demanding PX4/Gazebo closed loop, where the outer-loop controller acts through a cascaded flight stack with delay-sensitive settling and stronger transit-to-hover coupling. This step exposes which benchmark gains survive autopilot-mediated dynamics and which refinements collapse once the loop becomes more deployment-like. In Phase I, the bare controller attains 0.024 m mean late-stage wind error. In Phase II, the final controller is selected using a transfer-oriented rule emphasizing absence of benchmark priors, cross-scenario balance, and deployable supervisory logic. Strict is used as the primary reporting reference; the supplementary retrospective Grace analysis shows that part of the residual failure set is sensitive to completion semantics rather than gross waypoint-miss behaviour. The evaluation is completed on one Vicon-tracked Tello stack through a two-level hardware study. Taken together, the results suggest that benchmark success becomes more informative when the main controller design is separated from benchmark-specific refinement and remains defensible under harder closed-loop evaluation.