Dear Editor,This letter presents an adaptive control framework for unmanned aerial manipulator(UAM)perching,integrating image segmentationbased tracking with multimodal nonlinear model predictive control(NMPC).A hybri...Dear Editor,This letter presents an adaptive control framework for unmanned aerial manipulator(UAM)perching,integrating image segmentationbased tracking with multimodal nonlinear model predictive control(NMPC).A hybrid control strategy with depth derivative prediction and pitch feedforward compensation effectively resolves control parameter conflicts between pole grasping and suction-based perching on planar surfaces.An autonomous switching mechanism,leveraging detection confidence,spatial fusion,and pixel-level feature enhancement,ensures accurate target pose estimation under dynamic occlusions.展开更多
基金supported in part by the National Natural Science Foundation of China(62573168,62225305,62527807)the State Administration of Science,Technology and Industry for National Defense(JCKY2024603C035)the Open Project of the National Key Laboratory of Autonomous Intelligent Unmanned Systems(ZZKF2025-4-1)。
摘要Dear Editor,This letter presents an adaptive control framework for unmanned aerial manipulator(UAM)perching,integrating image segmentationbased tracking with multimodal nonlinear model predictive control(NMPC).A hybrid control strategy with depth derivative prediction and pitch feedforward compensation effectively resolves control parameter conflicts between pole grasping and suction-based perching on planar surfaces.An autonomous switching mechanism,leveraging detection confidence,spatial fusion,and pixel-level feature enhancement,ensures accurate target pose estimation under dynamic occlusions.