Organizations: The Hong Kong University of Science and Technology (Guangzhou), Guangzhou 511453, China · The Hong Kong University of Science and Technology, Hong Kong SAR, China
Autonomous landing of Uncrewed Aerial Vehicles (UAVs) on oscillating marine platforms is severely constrained by wave-induced multi-frequency oscillations, wind disturbances, and prediction phase lags in motion prediction. Existing methods either treat platform motion as a general random process or lack explicit modeling of wave spectral characteristics, leading to suboptimal performance under dynamic sea conditions. To address these limitations, we propose SpecFuse: a novel spectral-temporal fusion predictive control framework that integrates frequency-domain wave decomposition with time-domain recursive state estimation for high-precision 6-DoF motion forecasting of Uncrewed Surface Vehicles (USVs). The framework explicitly models dominant wave harmonics to mitigate phase lags, refining predictions in real time via IMU data without relying on complex calibration. Additionally, we design a hierarchical control architecture featuring a sampling-based HPO-RRT* algorithm for dynamic trajectory planning under non-convex constraints and a learning-augmented predictive controller that fuses data-driven disturbance compensation with optimization-based execution. Extensive validations (2,000 simulations + 8 lake experiments) show our approach achieves a 3.2 cm prediction error, 4.46 cm landing deviation, 98.7% / 87.5% success rates (simulation / real-world), and 82 ms latency on embedded hardware, outperforming state-of-the-art methods by 44%-48% in accuracy. Its robustness to wave-wind coupling disturbances supports critical maritime missions such as search and rescue and environmental monitoring. All code, experimental configurations, and datasets will be released as open-source to facilitate reproducibility.
Figures & tables
Fig. 1: Overview of the UAV-USV maritime rescue operation. The proposed SpecFuse framework (spectral-temporal fusion + learning-augmented control) enables the UAV to predict the USV’s 6-DoF motion (50Hz) and adjust trajectories dynamically, ensuring precise landing ( <5cm deviation) for medical resupply missions.
Fig. 2: Proposed control architecture. The Motion Prediction module forecasts the platform’s state, while the Learning-Augmented controller utilizes multi-source feedback to command the UAV actuators.
Fig. 3: Numerical validation in simulation of adaptive trajectory planning for UAV/USV cooperative landing under high-fidelity marine disturbances. The blue line represents the UAV’s planned trajectory, while the square platform with coordinate axes indicates the pose of the oscillating marine platform.
Method
ep ( cm ) ↓
Δp ( cm ) ↓
SR ↑
Latency ( ms ) ↓
FFT-KF [ 16 ]
5.7 ± 1.4
9.3 ± 2.5
89.1%
105 ± 4
MPC-VIO [ 14 ]
6.4 ± 1.9
11.8 ± 3.3
83.0%
118 ± 3
NMPC [ 12 ]
8.9 ± 2.8
14.1 ± 4.1
77.8%
142 ± 5
Proposed
3.2 ± 0.6
4.8 ± 1.1
98.8%
82 ± 2
TABLE I: Simulation Performance Comparison with Challenging Marine Disturbances
Module
Latency ( ms )
Peak RAM ( MB )
Spectral Prediction
34.2
82
HPO-RRT* Planning [ 27 ]
41.7
157
Learning-Based Velocity Control
6.1
29
Total System
81.9
268
TABLE II: Computational Efficiency Analysis on Embedded Hardware
Fig. 4: Experimental setup in the outdoor environment. A wave-generating vehicle follows a predefined trajectory around the USV, with the UAV’s field of view indicated.
Fig. 5: Autonomous UAV landing on an oscillating marine platform in an outdoor lake environment. The left column displays the third-person view with the UAV highlighted by red circles, while the right column shows the corresponding onboard downward-facing camera views of the landing target.
No.
Mean Wind
Max Wind
Wave Height
No.
Mean Wind
Max Wind
Wave Height
( m/s )
( m/s )
( m )
( m/s )
( m/s )
( m )
1
7.2
9.4
1.5
5
7.5
9.2
1.6
2
8.1
10.2
1.7
6
10.5
12.7
2.0
3
6.8
8.9
1.4
7
8.7
10.9
1.8
4
9.3
11.8
1.9
8
13.2
15.6
2.3
TABLE III: Calculated Environmental Parameters for Outdoor Lake Experiments
No.
Pred. Err.
Land. Dev.
Land. Time
Succ.
1
3.1 cm
4.2 cm
22.3 s
✓
2
3.9 cm
5.1 cm
23.1 s
✓
3
2.8 cm
3.8 cm
21.9 s
✓
4
3.6 cm
4.7 cm
22.8 s
✓
5
3.0 cm
4.0 cm
22.5 s
✓
6
4.2 cm
5.4 cm
23.4 s
✓
TABLE IV: Real-world Landing Results in Outdoor Lake Experiments
Autonomous landing of unmanned aerial vehicle (UAV) on unmanned surface vehicle (USV) is difficult in high sea states. Existing hierarchical landing methods often infer future USV motion only implicitly from short observation histories, which limits their ability to adapt to severe and rapidly changing sea conditions. This paper presents a prediction-informed hierarchical control framework. A pretrained model is developed to explicitly predict the USV attitude during high-level landing decisions. The prediction helps identify suitable landing windows and choose the right touchdown time, while a conventional low-level flight controller maintains UAV stability and tracks motion commands. To evaluate the framework under realistic disturbances, a real-sim validation setup is constructed in which USV-motion data collected in real sea conditions are transferred to a MuJoCo USV. The experiments demonstrate the feasibility of using explicit attitude prediction to improve landing-window selection and support timing-aware UAV landing planning in rapidly varying marine environments.
Kaizhen Li, Xi Zhou, Dan Zhang +1
School of Mechatronic Engineering and Automation Shanghai University Shanghai, China
Autonomous landing of unmanned aerial vehicles (UAVs) on wave-disturbed marine platforms remains challenging due to stochastic platform motion, time-varying platform attitude, and uncertain touchdown conditions. Existing model-based methods often require accurate motion prediction and online optimization, while end-to-end learning approaches may suffer from high training complexity and limited interpretability. This paper presents WaveLander, a hierarchical control framework via reinforcement learning (RL) that decouples vertical landing decision-making from low-level flight stabilization. The RL policy maps a compact platform-relative observation to a scalar vertical velocity reference, while a conventional low-level flight controller maintains attitude stability and lateral tracking. This formulation reduces dynamic platform landing to a low-dimensional, timing-aware control problem and enables smooth landing behavior without explicit switching rules. Simulation results under randomized wave-induced platform motions show that WaveLander achieves robust landing performance and generalizes to unseen disturbance conditions, demonstrating the potential of hierarchical learning-based control for marine UAV recovery.
Chun-Kit Li, Iok Long Sit, Ming Fung Siu +4
Department of Mathematics, The Hong Kong University of Science and Technology, Hong Kong SAR. · Cheng Kar-Shun Robotics Institute, The Hong Kong University of Science and Technology, Hong Kong SAR. · Division of Integrative Systems and Design, The Hong Kong University of Science and Technology, Hong Kong SAR. +2
Autonomous aerial inspection of marine infrastructure is frequently compromised by stochastic sea states, introducing risks of high-kinetic impacts, post-landing toppling, and sensory occlusion. This paper proposes a decoupled, multi-vehicle landing framework synchronizing an Unmanned Surface Vehicle (USV) equipped with a 3-RPU stabilized platform with a robust Unmanned Aerial Vehicle (UAV). The architecture utilizes two independent Deep Reinforcement Learning (DRL) agents: a Soft Actor-Critic (SAC) agent providing high-frequency wave-motion compensation for the landing deck, and a multimodal RL agent for the UAVs final approach. Evaluated in high-fidelity maritime simulations, the system achieved a 100% landing success rate across 15 trials in wave states varying from calm to rough. Results show a mean stabilization efficacy of 87.8%, maintaining the landing surface within 1 degree of the horizontal plane for 96% of the mission duration in rough conditions, effectively contributing to safer landings.
Francisco S. Neves, Pedro N. Pereira, Raul D. S. G. Campilho +1
Faculty of Engineering, University of Porto (FEUP) · Instituto Superior de Engenharia do Porto (ISEP)