Event-Based Adaptive Koopman Framework for Optic Flow-Guided Landing on Moving Platforms

2025-01-28

AI Summary

Safe soft landing of resource-constrained UAVs on dynamically moving platforms (e.g., ships, vehicles) remains challenging due to abrupt platform motion, ground effect, and sensor noise. Method: This paper proposes an event-driven adaptive control framework leveraging optical flow perception. It integrates Koopman operator theory with online adaptive system identification to construct a real-time-updatable dynamical model, coupled with an event-triggered model predictive controller (MPC) designed to avoid Zeno behavior and ensure global convergence. Contributions/Results: (1) First online adaptive mechanism within the Koopman framework explicitly compensates for unknown platform motion and ground effect; (2) A synergistic optical-flow–event–MPC architecture balances computational efficiency and robustness. Simulation results demonstrate significantly improved landing accuracy and stability over both non-adaptive event-triggered and conventional time-triggered adaptive methods under strong ground effect and sensor noise.

Technology Category

• Intelligent Robots: State Estimation
• Humans and AI: Planning and Decision Support for Human-Machine Teams
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

This paper presents an optic flow-guided approach for achieving soft landings by resource-constrained unmanned aerial vehicles (UAVs) on dynamic platforms. An offline data-driven linear model based on Koopman operator theory is developed to describe the underlying (nonlinear) dynamics of optic flow output obtained from a single monocular camera that maps to vehicle acceleration as the control input. Moreover, a novel adaptation scheme within the Koopman framework is introduced online to handle uncertainties such as unknown platform motion and ground effect, which exert a significant influence during the terminal stage of the descent process. Further, to minimize computational overhead, an event-based adaptation trigger is incorporated into an event-driven Model Predictive Control (MPC) strategy to regulate optic flow and track a desired reference. A detailed convergence analysis ensures global convergence of the tracking error to a uniform ultimate bound while ensuring Zeno-free behavior. Simulation results demonstrate the algorithm's robustness and effectiveness in landing on dynamic platforms under ground effect and sensor noise, which compares favorably to non-adaptive event-triggered and time-triggered adaptive schemes.

Problem

• Drone Landing
• Moving Platform
• Stability and Precision

Innovation

• Koopman Method
• Real-time Prediction
• Adaptive Control Strategy