🤖 AI Summary
This work addresses the challenge of ensuring continuous target visibility in autonomous aerial tracking—particularly under significant trajectory prediction uncertainty, when full-body (not just centroid) visibility is required, and across mixed scenarios involving single/dual targets and static/dynamic environments. We propose the first method jointly modeling target trajectory prediction uncertainty and the complete body’s reachable visibility region, introducing path homotopy constraints to enhance visibility robustness. We design an obstacle-aware visibility metric and formulate a polynomial trajectory optimization problem solved via quadratic programming, augmented by a sampling-and-verification strategy for efficient reachable-region estimation. Evaluated in high-fidelity simulation and real-world flight experiments, our approach achieves a 37.2% increase in average target visible duration and a 58.6% reduction in tracking failure rate over baseline methods. It establishes the first unified, geometrically rigorous, and real-time feasible visibility assurance framework applicable across diverse operational scenarios.
📝 Abstract
Maintaining the visibility of the target is one of the major objectives of aerial tracking missions. This paper proposes a target-visible trajectory planning pipeline using quadratic programming. Our approach can handle various tracking settings, including single and dual target following and both static and dynamic environments, unlike other works that focus on a single specific setup. In contrast to other studies that fully trust the predicted trajectory of the target and consider only the visibility of the center of the target, our pipeline considers error in target path prediction and the entire body of the target to maintain the target visibility robustly. First, a prediction module uses a sample-check strategy to quickly calculate the reachable areas of moving objects, which represent the areas their bodies can reach, considering obstacles. Subsequently, the planning module formulates a single QP problem, considering path homotopy, to generate a tracking trajectory that maximizes the visibility of the target's reachable area among obstacles. The performance of the planner is validated in multiple scenarios, through high-fidelity simulations and real-world experiments.
