Learn for Variation: Efficient AAV Trajectory Learning through a Differentiable Wireless World Model

Xiucheng Wang; Zhenye Chen; Nan Cheng; Zhisheng Yin; Xuemin Shen

eess.SYMar 19, 2026

Learn for Variation: Efficient AAV Trajectory Learning through a Differentiable Wireless World Model

Authors: Xiucheng Wang, Zhenye Chen, Nan Cheng, Zhisheng Yin, Xuemin Shen

Organizations: State Key Laboratory of ISN and School of Telecommunications Engineering, Xidian University, Xi’an 710071, China · School of Aerospace Science and Technology, Xidian University, Xi’an 710071, China · Department of Electrical and Computer Engineering, University of Waterloo, Waterloo, N2L 3G1, Canada

Abstract

Autonomous aerial vehicles (AAVs) enable data collection for sixth-generation Internet-of-Things networks, but their trajectories couple nonlinear wireless rates with long-horizon service progress. This paper views the evolution of AAV kinematics, channel state, and user backlog as a structured differentiable world model and develops Learn for Variation (L4V) to exploit that model efficiently. L4V replaces a discontinuous completion-time objective with a cumulative-backlog surrogate, unrolls the mission dynamics, and propagates pathwise sensitivities to a neural policy through the discrete adjoint recursion. The resulting derivative is exact conditional on a fixed exogenous-noise realization; stochastic expected-objective optimization still requires sampling. We show that the structured adjoint grows at most polynomially with the horizon and establish a stationary-point rate for fixed-step full-gradient descent under standard smoothness assumptions. The framework also learns shared OFDMA allocation under reparameterized shadowing and Rician fading, while distributional pretraining amortizes model-based optimization into forward-only deployment on unseen layouts. Paired stress tests cover channel-generator mismatch, noisy partial observations, a fixed-resource two-AAV extension, and a circular no-fly region. Code and configurations are available at https://github.com/UNIC-Lab/L4V-AAV. Against genetic-algorithm, DQN, A2C, DDPG, and differentiable model-predictive-control implementations, L4V reduces mission time by up to $65\%$ , executes a default mission in $53$ ms, and completes all $60$ frozen-policy tests after pretraining on $1{,}600$ layouts.

Figures & tables

Fig. 1 : The L4V framework. Each blue transition block is one step of the structured wireless world model. The forward pass (black) unrolls the AAV trajectory and accumulates $\mathcal{L}$ ; the backward pass (red) implements the discrete adjoint and propagates exact conditional pathwise gradients $\nabla_{\theta}\mathcal{L}$ to the policy parameters.

Panel of Figure 2 — Fig. 2 : Performance versus service-area size $L$ (a, b) and noise power $\sigma^{2}$ (c, d). Medians over $10$ trials with inter-quartile bands; completion time on a log axis.

Panel of Figure 3 — Fig. 3 : Performance versus user density $K$ (a, b) and channel gain $\eta$ (c, d). Medians over $10$ trials with inter-quartile bands; completion time on a log axis.

Method	Scenario optimization	Execution/solve
L4V	293 s	53 ms
GA	–	2369 s
DQN	887 s	118 ms
A2C	2475 s	131 ms
DDPG	3506 s	111 ms
MPC	–	64 s

TABLE I : Scenario-specific optimization and execution costs at the default configuration. Neural execution is the warmed median rollout cost scaled to the observed median mission length; GA and MPC report median per-instance solve time. Held-out transfer is evaluated separately in Fig. 6(e) .

Panel of Figure 5 — Fig. 4 : Allocation ablation and energy-weight sensitivity. Panels (a, b) compare learned allocation with equal split over five paired instances; thin segments preserve each pairing and thick horizontal bars mark medians. The broken axes omit empty ranges while retaining and labeling the capped run. Panel (c) shows every energy-sweep trial with median/IQR summaries and a linear diagnostic. Panel (d) reports paired percentage changes from $\gamma=0$ with inter-quartile bands. Lower is better throughout, and capped or regressing trials are retained.

Fig. 5 : Executed AAV paths on one shared

K=4

instance; marker size scales with task demand and each path is labeled with its mission length.

Panel of Figure 7 — Fig. 6 : Reviewer-driven stress tests. Panels (a, b) keep the trained policy fixed and pair the random draws across deployment conditions. The nominal channel is $(\kappa,K_{\rm r},\nu)=(0.15,6,2)$ ; mild and severe shifts are $(0.25,3,2.2)$ and $(0.40,0,2.5)$ . Observation tests add a fixed location bias of standard deviation $0.30$ , independently mask each current backlog update with probability $0.50$ while holding its last estimate, or combine both. Panel (c) retains one global radio budget as the fleet grows from one to two AAVs. Panel (d) applies hard collision checks after training with or without obstacle geometry and a soft clearance barrier. Panel (e) freezes one distributionally pretrained policy on disjoint layouts. Panel (f) reports completion and circular turn variation. Thin marks retain trial-level outcomes; thick marks and bands show medians and IQRs; capped runs and success rates are retained. Logarithmic completion-time ordinates in (b) and (f) preserve extreme trials without compressing the median region.

UAV Trajectory Optimization

UAV trajectory optimization plans feasible, safe flight paths that balance mission goals, energy use, sensing, and communication.

See all 48 papers

Learning-Accelerated Optimization-based Trajectory Planning for Cooperative Aerial-Ground Handover Missions
May 19, 2026Jingshan Chen, Bochen Yu, Henrik Ebel +1Multi-Robot Motion Planning Trajectory Optimization
Institute of Engineering and Computational Mechanics, University of Stuttgart, 70569 Stuttgart, Germany · Mechanical Engineering, LUT University, 53850 Lappeenranta, Finland
LiLib: Lifelong Air-to-Ground Path-Loss Prediction on UAVs via a Drift-Triggered Model Library
Oct 5, 2026Minh TranConcept Drift Wireless Communications
Hanoi University of Science and Technology (HUST), Hanoi, Vietnam
Uncertainty-Aware World Model for Aerial Image-Goal Navigation
Aug 6, 2026Deyi Zhu, Haoyu Fan, Yinan Zhu +4Uncertainty Quantification World Models for Robotics
Tsinghua Shenzhen International Graduate School, Tsinghua University

May 19, 2026J/K move · Enter open · S save

Learning-Accelerated Optimization-based Trajectory Planning for Cooperative Aerial-Ground Handover Missions

Jingshan Chen, Bochen Yu, Henrik Ebel, Peter Eberhard

Institute of Engineering and Computational Mechanics, University of Stuttgart, 70569 Stuttgart, Germany · Mechanical Engineering, LUT University, 53850 Lappeenranta, Finland

Multi-Robot Motion Planning Trajectory Optimization Trajectory Generation Multi-Robot Coordination UAV Trajectory Optimization Learning-Augmented Algorithms

Open paper AskPDF arXiv

This paper presents a learning-augmented trajectory planning framework for cooperative unmanned aerial vehicle (UAV) and unmanned ground vehicle (UGV) handover missions. While centralized trajectory optimization ensures dynamic feasibility and task optimality, its high computational cost limits real-time applicability. We propose a neural surrogate planner utilizing decoupled encoder-decoder long short-term memory (LSTM) networks to generate coordinated handover trajectory predictions from the task specifications. These predictions serve as informed warm starts for the downstream centralized optimizer, thereby accelerating convergence to dynamically feasible solutions. Benchmark evaluations demonstrate that the learning-augmented planning framework achieves more than a threefold speedup and 100% optimization success rate compared to cold start optimization. The results indicate that combining data-driven inference with model-based refinement enables fast and reliable trajectory generation for heterogeneous multi-robot systems.

Abstract

UAV Trajectory Optimization

Explore similar work

Learning-Accelerated Optimization-based Trajectory Planning for Cooperative Aerial-Ground Handover Missions

LiLib: Lifelong Air-to-Ground Path-Loss Prediction on UAVs via a Drift-Triggered Model Library

Uncertainty-Aware World Model for Aerial Image-Goal Navigation