Mission Efficiency Optimization in Low-Altitude Economy: Adaptive Power Allocation for Coordinating Heterogeneous Aircraft Swarms

arXiv:2609.39776 · cs.IT, cs.SY, eess.SY, math.IT · Submitted 2026-09-30 · Read on arXiv

Listen

Transcript

Introduction to the show: ident: Robotics Radio. Generated commentary on the latest robotics and control papers.

Rosa: I'm Rosa, and with me are Dev and Taro, guest researcher.

Dev: Today's paper: "Mission Efficiency Optimization in Low-Altitude Economy".

Rosa: With the rapid development of low-altitude economy, complex missions requiring collaborative efforts among heterogeneous low-altitude aircraft (LAAs) necessitate efficient coordination through adaptive power allocation.

Dev: First, who's behind it and why it matters.

Paper summary: Rosa: So we're looking at the paper "Mission Efficiency Optimization in Low-Altitude Economy: Adaptive Power Allocation for Coordinating Heterogeneous Aircraft Swarms," and the main idea is that they introduce operational-capability entropy to figure out how to allocate power better when you have different types of low-altitude aircraft working together.

Dev: That's right, Rosa, the core thesis seems to be around using this operational-capability entropy as a way to quantify the effective work capability of these operational LAAs during a mission, and then using that along with channel conditions to solve for an adaptive power allocation problem aimed at minimizing the LQR cost.

Taro: I'm interested in how they define this operational-capability entropy, because it sounds like it's going beyond just looking at raw communication performance or individual aircraft capabilities; it seems to capture something about the physical limits of what a specific LAA can actually receive, decode, and use within one SC3 cycle.

Rosa: Exactly, Taro; they quantify this as "bits/SC3 cycle," which means a higher OCE value suggests the LAA can handle more complex tasks and demands more control information. This then sets up the constraint where R k E k (10g), linking the required data rate to this operational capability bound.

Dev: And that linkage is crucial because it ties directly into their formulation of problem (P1), which is the non-convex optimization problem they are trying to solve for power allocation vector p and auxiliary vector w. They state that minimizing the LQR cost function is mathematically equivalent to maximizing the mission-related data rate R.

Taro: Maximizing that data rate, R, seems like a very practical goal for a swarm operating in dynamic conditions, but I'm also looking at how they handle the channel heterogeneity mentioned in page one, where they model the channel gain h k using small-scale Rayleigh fading.

Rosa: They address that by incorporating the ergodic capability of each LAA channel into their formulation of C k, which is given by C k = E two one + p kh hk squared sigma squared (one). This shows they are directly accounting for both the inherent operational capability and how the specific channel conditions affect what each aircraft can actually achieve.

Paper summary: Dev: That's where I see a lot of engineering interest, Rosa; because they explicitly deal with heterogeneous OCE and channel conditions simultaneously in their formulation, it suggests a much more robust allocation strategy than just optimizing for one factor at a time. They then transform this non-convex problem (P1) into a max-min problem (P2) to make it solvable.

Taro: The transformation into the max-min problem (P2), which involves maximizing p while minimizing w, is interesting because it allows them to decompose the overall optimization into two separate convex subproblems, (P3) and (P4). That decomposition is a key step in making this complex problem tractable.

Rosa: And that decomposition leads to specific solutions for the power allocation vector p* and the auxiliary vector w*. The resulting optimal solution for p* k involves an upper bound term pupper k, which depends on both the OCE, channel noise variance, and other parameters like e k (page two).

Dev: I’m looking at those derived solutions now, and it seems that the optimal power allocation p* k is constrained by this calculated pupper k, which itself incorporates terms like sigma squared e w l squared k e E k two BkT - w k-one - one (Proposition two). This is quite detailed, showing how the power allocation is directly shaped by the operational constraints.

Taro: When we think about what this means in practice, especially if we consider scenarios where the world misbehaves, how does this scheme handle situations where one aircraft suddenly has a drastically reduced capability due to physical damage or environmental interference?

Rosa: That's a valid point for the real-world application; since their entire framework is built around quantifying that effective work capability via OCE, the system should naturally react by adjusting the power allocation to match that new lower bound on E k. If an LAA's capability drops, its allocated power p k will be constrained accordingly.

Dev: From a control engineer's view, I'm concerned about the loop rate and latency when we deploy this; since they use an iterative algorithm like the primal-dual steepest descent with global linear convergence, we need to make sure that iteration converges fast enough for real-time operation, especially with the channel dynamics.

Taro: The authors do mention that their solution involves an iterative algorithm, which is good because it implies a way to handle the complexity of (P2), but I wonder about the computational load this puts on the operational LAAs themselves when they have to solve these subproblems in real-time.

Paper summary: Rosa: It does put a load on them, but they argue that by framing it as an adaptive power allocation problem that minimizes LQR cost, they are optimizing for mission efficiency, which should justify the computational overhead compared to less coordinated methods.

Dev: The simulation results support this idea, showing that this proposed scheme achieves the lowest LQR cost for any given P max, and interestingly, when P max is below 7dBW, conventional schemes actually lead to system instability where the cost approaches infinity. That's a significant finding regarding stability.

Taro: That instability at lower power limits suggests that this OCE-based approach provides a much better safety margin or constraint adherence compared to simpler power allocation methods when resources are tight. It addresses the robustness aspect of autonomy in adverse conditions.

Rosa: So, to wrap up this discussion on "Mission Efficiency Optimization in Low-Altitude Economy: Adaptive Power Allocation for Coordinating Heterogeneous Aircraft Swarms," the paper's main contribution is using operational-capability entropy to jointly consider heterogeneous OCE and channel conditions to formulate a power allocation problem that minimizes LQR cost.

Dev: And the implication, especially from an engineering standpoint, is that this approach offers a way to achieve better mission efficiency by explicitly modeling how different aircraft capabilities and fading channels interact in real-time.

Taro: For the broader implications, this work suggests a viable path for highly coordinated low-altitude swarms where individual aircraft roles are diverse and their operational limits vary significantly, moving us closer to truly autonomous collaborative operations.

Rosa: I think that's what excites me most; it moves the research from isolated communication studies to designing holistic, mission-oriented coordination strategies for complex multi-agent systems operating in the low altitude environment.

Dev: I agree, and the way they handle the non-convex nature by transforming it into a max-min structure shows a solid mathematical foundation for implementing this kind of adaptive control loop reliably.

Taro: Moving forward, I think future work could explore how this OCE concept generalizes to even more complex mission scenarios involving unexpected system failures or dynamic topology changes within the swarm.

Rosa: That would be an interesting direction; testing the limits of this scheme under extreme, unexpected operational stress would really validate its practical utility beyond the controlled simulation environment.

Dev: It sounds like a solid piece of work that connects theoretical constraints with practical mission performance metrics through this adaptive power allocation framework.

Conclusion: Taro: I've been thinking about the core idea of operational-capability entropy—how it quantifies that effective work capability in bits per SC3 cycle—and it seems like a really clever way to put a hard physical limit on what an aircraft can actually do.

Rosa: Exactly, Taro; it moves beyond just looking at communication links and incorporates the physical limits of how much information each aircraft can handle during a mission cycle.

Dev: And from my side, I’m really focused on how they managed to make that non-convex problem solvable by decomposing it into two convex subproblems, (P3) and (P4), which is something I find pretty impressive for a real-time control loop.

Taro: That decomposition is key because it lets them solve the maximization of the data rate, R, while keeping the power allocation constraints manageable through those iterative algorithms they used.

Rosa: It really shows how they connect abstract concepts like entropy and optimization directly to practical constraints like power limits and channel conditions in a way that seems very applicable outside of just a textbook setting.

Dev: I'm still wondering about the hardware aspect; Rosa, if we take these results from the simulation and try to deploy this on actual LAAs, how long do you think the system could reliably run before latency becomes an issue?

Rosa: Well, based on their simulation showing stability even when P max is low, I'd say it has a good chance of working in real-world scenarios if we can keep the iteration speed high enough to meet tight control deadlines.

Taro: That brings up my concern about failure modes; what happens to this adaptive power allocation strategy when the environment suddenly shifts drastically, like an aircraft losing its sensor or communication link completely?

Dev: That’s a critical question, Taro; they do flag that the framework is designed to react by adjusting constraints based on the new capability bounds defined by OCE, which should handle sudden drops in performance.

Rosa: So it seems the main implication here is that we're moving toward a more resilient swarm coordination system where each aircraft dynamically adjusts its power usage not just for communication, but for mission execution capabilities.

Taro: The bigger picture is that this kind of joint consideration of operational limits and channel conditions could make truly autonomous, complex collaborative operations feasible in the low-altitude domain.

Jiarui Zhang, Wei Feng, Chao Dong, Ning Ge, Qihui Wu

Department of Electronic Engineering, State Key Laboratory of Space Network and Communications, Tsinghua University · Key Laboratory of Dynamic Cognitive System of Electromagnetic Spectrum Space, Nanjing University of Aeronautics and Astronautics

cs.IT, cs.SY, eess.SY, math.IT

Submitted: 2026-09-30

Updated: 2026-09-30

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 69/100

The gist: With the rapid development of low-altitude economy, complex missions requiring collaborative efforts among heterogeneous low-altitude aircraft (LAAs) necessitate efficient coordination through

Key concepts

Operational-Capability Entropy (OCE): OCE measures the effective work capability of an aircraft in 'bits per SC3 cycle.' It quantifies the physical limit on how much control information an LAA can successfully receive, decode, and use within one complete sensing-communication-computing loop. Higher OCE means the aircraft can handle more complex tasks.
SC3 Closed Loop: This refers to the multi-aircraft coordination structure in low-altitude operations involving sensing, communication, and computing. The operational LAAs are crucial for this loop, but their communication resources are limited and they vary significantly in both their operational skills and the quality of the wireless channel they experience.
Linear Quadratic Regulator (LQR) Cost: This cost function is used to mathematically measure mission efficiency. Minimizing this cost directly optimizes how power is allocated across the swarm, ensuring that resources are used optimally to achieve the desired mission objectives while respecting physical constraints.

Terminology

Core — 4× in this paper · explained in 2 episodes
Decode — 3× in this paper · explained in 2 episodes
Linear Quadratic Regulator — 2× in this paper
Closed-Loop Control — 1× in this paper · explained in 2 episodes
Computational overhead — 1× in this paper · explained in 2 episodes
Deep Reinforcement Learning — 1× in this paper · explained in 2 episodes
Deep Reinforcement Learning (DRL) — 1× in this paper · explained in 2 episodes
Digital Twin — 1× in this paper · explained in 4 episodes
Failure Modes — 1× in this paper · explained in 3 episodes
Mobile Edge Computing — 1× in this paper · explained in 2 episodes
Multi-Agent System — 1× in this paper · explained in 6 episodes
Reinforcement Learning — 1× in this paper · explained in 4 episodes
Robustness — 1× in this paper · explained in 5 episodes

Summary

With the rapid development of low-altitude economy, complex missions requiring collaborative efforts among heterogeneous low-altitude aircraft (LAAs) necessitate efficient coordination through adaptive power allocation. The proposed scheme introduces operational-capability entropy (OCE) to quantify effective work capability and form a mission-oriented adaptive power allocation problem that minimizes the Linear Quadratic Regulator (LQR) cost, thereby optimizing mission efficiency by jointly considering heterogeneous OCE and channel conditions.

The gist: The proposed scheme effectively exploits limited resources by jointly accounting for OCE and channel heterogeneity, significantly outperforming traditional approaches and validating the value of using OCE in closed loop system design.

Introduction to the Problem

Low-altitude operations rely on multi-LAA coordination, forming a sensing-communication-computing control (SC3) closed loop. This involves LAAs with distinct roles: sensing, communication, computing, and mission execution. The operational LAAs are critical to the SC3 closed loop; however, their communication resources are constrained by limited payloads and they exhibit significant heterogeneity in both operational capabilities and channel conditions. Existing studies often focus only on communication performance or model multi-actuator missions as independent loops, failing to capture the collaborative nature of coordinated unmanned operations or the hardware limitations of operational LAAs.

Quantifying Operational Capability with OCE

The paper introduces the concept of operational-capability entropy (OCE) to quantify the effective work capability of operational LAAs during mission execution, measured in bits/SC3 cycle. This metric captures the physical limit on how much control information an LAA can effectively receive, decode, and utilize within one SC3 cycle. An LAA with a large OCE is assigned more complex tasks, while a small OCE implies fewer commands for simpler tasks. The constraint imposed by this capability is given by Rk ≤ Ek, where Rk is the rate of LAA k's data rate, and Ek is its operational capability bound.

Formulation of the Mission-Oriented Optimization

The core objective is to minimize the LQR cost function, which serves as a metric for mission efficiency. The problem (P1) is formulated as:

(P1) min p,w l (10a) s.t. l ≥ nN(v) detM/n squared 2n (R−log2 det A) − 1 + tr(ΣvS) (10b), R > log2 det A (10c), R ≤ XK k=1 Rk (10d), XK k=1 pk ≤ Pmax (10e), Rk ≤ BkT Ck,ap, k = 1, 2, · · ·, K (10f), and Rk ≤ Ek, k = 1, 2, · · ·, K (10g).

This non-convex problem is transformed into a max-min problem (P2) to maximize the mission-related data rate R. The transformation leads to two convex subproblems:

(P3) max p XK k=1 Rk(pk) s.t. (10e),(10h),(11b).

(P4) min w XK k=1 Rk(wk) s.t. (10i).

Decomposition and Solution Strategy

The problem (P2) is shown to be strongly concave in p and strongly convex in w, allowing it to be decomposed into two convex optimization subproblems, (P3) and (P4). The solution involves an iterative algorithm, such as the primal-dual steepest descent (PDSD) algorithm with global linear convergence.

(P3) is a convex optimization subproblem in p with w fixed.

(P4) is a convex optimization subproblem in w with p fixed.

The optimal solution for power allocation vector p and auxiliary vector w are derived:

(Proposition 2) The optimal solution to (P3) is given by:

(p∗k = (pupperk, if PK k=1 pupperk ≤ Pmax, min n pupperk, max n 0, BkTλ − σ2 l squared k oo,otherwise), where pupperk ≜ σ squared e w l squared k e Ek ln 2 BkT −wk−e−wk +1 − 1).

(Proposition 3) The optimal solution to (P4) is given by:

(w∗k = ln (1 + q 1 + 4pkl squared k σ squared / 2)).

Simulation Results and Discussion

Simulation results demonstrate that the proposed scheme achieves the lowest LQR cost for any given Pmax. When Pmax is less than 7dBW, conventional schemes lead to system instability, causing the cost to approach infinity.

Improvements for AI systems

Based on this scientific paper, here are the specific improvements that can be made to AI systems by implementing the proposed methodology:

Improve coordination in multi-agent robotic/drone swarms operating in complex, time-critical environments (e.g., disaster response, search and rescue).
Enable autonomous decision-making for heterogeneous aircraft where different agents have varying sensing, communication, computing, and execution capabilities (the SC3 loop).
Optimize the power allocation strategy for communication links between an Edge Information Hub (EIH) and operational aircraft to maximize mission efficiency rather than just raw data rate.
Develop AI control schemes that inherently respect the physical processing limits of individual agents, preventing overloading which can lead to system instability or command failure in real-world scenarios.
Design intelligent resource allocation systems that dynamically balance the uncertainty of wireless channel conditions (fading) with the inherent capability constraints (OCE) of each agent to ensure robust mission success under strict power budgets.

This improved AI system can:

Perform coordinated, autonomous missions in low-altitude economies by intelligently assigning sensing, communication, computing, and execution roles to heterogeneous aircraft.
Act as a highly efficient nerve center (EIH) that delivers control commands to operational agents while simultaneously optimizing the power budget across all links based on the calculated operational capability entropy (OCE) of each receiving agent and current channel conditions.
Achieve superior mission efficiency by minimizing a Linear Quadratic Regulator (LQR) cost function, which directly measures mission success metrics rather than just communication throughput.
Ensure that control commands delivered to an aircraft are always within its processing capacity, preventing time-outs and system destabilization caused by exceeding its operational capability entropy.
Dynamically adjust power allocation across the swarm to ensure that agents with high operational capacity receive appropriate task loads while mitigating the risk of allocating too much power to agents whose limited capability (low OCE) cannot effectively utilize it, leading to maximum overall mission performance under resource constraints.

Abstract

With the rapid development of the low-altitude economy, low-altitude operations are booming, where complex missions require collaborative efforts among multiple heterogeneous low-altitude aircrafts (LAAs). Specifically, different LAAs assume distinct roles: some for sensing, some for communication, some for computing, and others for mission execution, together forming a sensing-communication-computing-control (SC3) closed loop, akin to a reflex arc. To enable efficient coordination in such multi-LAA swarms, we introduce the concept of operational-capability entropy (OCE) to quantify the effective work capability of operational LAAs. Accordingly, by jointly considering heterogeneous OCE and channel conditions among LAAs, we formulate the power allocation problem with the goal of minimizing the linear quadratic regulator (LQR) cost, which serves as a metric for mission efficiency. The resulting complex optimization problem is decomposed into two convex subproblems that are solved iteratively, with closed-form solutions derived for each. Simulation results demonstrate that the proposed mission-oriented adaptive power allocation scheme significantly outperforms traditional ones.

Sources

Shared Sky, Shared Spectrum: Coordinated Satellite-5G Networks for Low-Altitude Economy