Energy-Efficient UAV Communications in the Presence of Wind: 3D Modeling and Trajectory Design

Xiaojun Yuan, Xinhong Dai, Bin Duo, Xiaojun Yuan 0002, Marco Di RenzoIEEE Trans. Wireless Communications 2024 · DOI

TL;DR

This paper tackles the challenge of energy-efficient communication for drones (UAVs) operating in windy conditions by developing a novel 3D propulsion energy consumption model and an adaptive trajectory design. The proposed approach significantly improves energy efficiency by allowing UAVs to dynamically adjust their flight paths and speeds to counteract wind disturbances, outperforming traditional windless designs.

What Problem Does This Paper Address?

Unmanned Aerial Vehicles (UAVs) are increasingly used for various communication tasks, but their energy efficiency is a critical bottleneck, especially for rotary-wing UAVs whose propulsion power is highly sensitive to wind. Existing research often simplifies energy consumption models or ignores wind effects, leading to suboptimal designs in real-world, dynamic environments. This paper addresses the need for an accurate energy model that accounts for 3D wind disturbances and proposes an adaptive trajectory optimization strategy to maximize energy efficiency (EE) under these conditions.

Methods

The authors introduce a Generalized Propulsion Energy Consumption Model (GPECM) for rotary-wing UAVs that incorporates 3D wind modeling and force analysis. This model captures the real-time wind's impact on propulsion power, considering UAV velocity and acceleration. To address the challenge of real-time wind information scarcity, they propose a two-phase Offline-based Online Adaptive (OBOA) design:

  • Offline Phase: Before flight, this phase uses statistical wind information (speed and direction distributions) to optimize a general UAV trajectory and user scheduling using stochastic programming (SP). This provides a statistically favorable, but suboptimal, reference trajectory.
  • Online Phase: During flight, this phase uses real-time wind measurements to adapt the UAV's instantaneous velocity slot-by-slot, referencing the offline trajectory. This allows for dynamic adjustments to minimize propulsion energy consumption and maximize EE in response to changing wind conditions.

The problem is formulated as a stochastic optimization problem, and to handle its complexity, it's decomposed into subproblems that are solved iteratively using techniques like successive convex approximation (SCA).

Datasets

Not discussed in the paper. The authors rely on simulated wind data and parameters derived from existing literature for their numerical experiments.

Key Insights and Ideas

  • Wind is a major energy drain: The paper clearly demonstrates that wind significantly impacts UAV energy consumption, and ignoring it leads to substantial inefficiencies.
  • 3D modeling is crucial: A 3D GPECM is necessary to accurately capture the complex interplay of wind, UAV motion, and energy expenditure.
  • Adaptive strategies are key: Real-time wind conditions are dynamic and unpredictable. An offline, statistically optimized trajectory is insufficient; online adaptation is essential for maximizing energy efficiency.
  • OBOA design balances planning and adaptation: The two-phase OBOA approach effectively leverages prior knowledge (offline phase) while allowing for real-time adjustments (online phase), offering a practical solution.
  • Lower flight altitudes can be more energy-efficient in headwinds: The study shows that UAVs may fly lower to counteract headwinds, saving energy and improving EE.

Key Results

  • The proposed offline design significantly improves EE compared to a windless scheme, demonstrating the importance of considering wind statistics.
  • The OBOA design further enhances EE by adapting to real-time wind changes, achieving higher performance than both the offline and windless schemes, especially in scenarios with dramatic wind fluctuations.
  • Simulations show that the OBOA design leads to a smaller angular deviation between the UAV's velocity and the real-time wind, indicating better adaptation.
  • The OBOA design saves propulsion energy (up to 20%) compared to the offline design, while the expected achievable rate is only slightly lower (no more than 10% difference).

Limitations and Future Work

  • Perfect wind measurement assumption: The online phase assumes instantaneous and perfect measurement of real-time wind information, which might be challenging in practice due to sensor limitations and collection delays.
  • Simplified channel model in online phase: While the offline phase uses a probabilistic LoS model, the online phase relies on a practical channel model, but the paper notes that the achievable rate is slightly lower than the offline design due to trajectory deviations.
  • Computational complexity: While the OBOA design is designed for low complexity per time slot, the overall computational burden for the offline phase can still be significant.

Future work could explore more realistic wind measurement models, investigate the impact of communication constraints on the online adaptation, and further optimize the computational efficiency of the algorithms.

Summary updated

BibTeX
@article{yuan2024energy,
  author = {Xiaojun Yuan and Xinhong Dai and Bin Duo and Xiaojun Yuan and Marco Di Renzo},
  title = {{Energy-Efficient UAV Communications in the Presence of Wind: 3D Modeling and Trajectory Design}},
  journal = {IEEE Trans. Wireless Communications},
  year = {2024},
  doi = {10.1109/twc.2023.3292290},
}
Figure