Aug 2026· Aerospace· Vol 13, pp. 721· 0 citations· 20 references
TL;DR
This study proposes an explainable, data-driven framework integrating active-reward proximal policy optimization (AR-PPO), which successfully distills black-box AI strategies into verifiable, physics-informed standard operating procedures (SOPs), providing a highly transparent and robust solution for autonomous windshear escape and future competency-based flight training.
Abstract
Low-altitude micro downbursts pose a severe threat to aviation safety, yet conventional control approaches and standard deep reinforcement learning (DRL) often fail due to explicit modeling difficulties and sparse reward constraints. To address these challenges, this study proposes an explainable, data-driven framework integrating active-reward proximal policy optimization (AR-PPO). A bilevel optimization architecture driven by meta-gradients is developed to dynamically discover optimal reward functions without human intervention. Furthermore, a policy distillation pipeline utilizing wavelet-multivariate singular spectrum analysis (W-MSSA) and classification and regression trees (CART) is proposed to translate high-frequency continuous neural outputs into discrete, pilot-readable rules. Simulation results on a B737-800 model demonstrate that AR-PPO effectively overcomes the “stall trap” by autonomously learning to trade altitude for airspeed, outperforming static-reward baselines and empirical human pilots in extreme, zero-shot windshear encounters (22.0 m/s downdraft). Ultimately, the proposed framework successfully distills black-box AI strategies into verifiable, physics-informed standard operating procedures (SOPs), providing a highly transparent and robust solution for autonomous windshear escape and future competency-based flight training.
Comparative control theory analysis reveals a fundamental trade-off: transitioning from continuous to discrete rule-based control induces high-frequency Bang-Bang actuation and a stable bimodal limit cycle.
Multi-agent reinforcement learning (MARL) has emerged as a promising approach for traffic signal control. However, standard MARL policies typically optimize for expected returns under nominal conditions, leaving them highly vulnerable to spatial-temporal demand shifts and catastrophic congestion under adverse scenarios. To address this critical limitation, this paper proposes an algorithm-agnostic Distributionally Robust (DR) MARL framework integrating an adaptive Contextual-Bandit Worst-Case Estimator (CB-WCE). Operating on a slower timescale, the CB-WCE co-evolves with the traffic controllers by dynamically generating adversarial demand mixtures during training. This steers the learning process to fortify policies against bottleneck scenarios without requiring modifications to the underlying MARL architectures. The framework is evaluated across value-based, actor-critic, and policy-gradient methods on both a synthetic 5x5 grid and a heterogeneous Monaco City network. Empirical results demonstrate that the DR framework prevents unbounded queue growth and profoundly enhances both worst-case robustness and average-case efficiency. Notably, for the Proximal Policy Optimization (PPO) architecture in the Monaco environment, on average, robust retraining reduced the worst-case queue length by 74.39% and improved the average-case network-wide queue length by 75.45%. Furthermore, the retrained policies exhibit strong zero-shot generalization to unseen traffic distributions, highlighting the framework's scalability and potential for resilient real-world urban deployment.
Shuwei Pei, Joran Borger, Arda Kosay et al.· 0 citations
The problems of partial observability and sensor shortage pose a significant challenge for autonomous Unmanned Aerial Vehicles (UAVs) as they prove to be challenging for conventional Deep Reinforcement Learning (DRL) methods to undertake well under such conditions. In this paper, a memory-augmented Proximal Policy Optimization (PPO) model extended using a Long Short-Term Memory (LSTM) network is proposed as a solution to such challenges. The observation space is constructed from 2D LiDAR and Inertial Measurement Unit (IMU) data to sense simultaneously external observation and internal state of motion, whereas the action space consists of continuous velocity commands. A shaped reward function is optimized for encouraging safe target approaching, obstacle avoidance, and convergence speed. Experimental outcomes show that the PPO-LSTM described herein achieves smoother paths, more robust reward convergence, and a much lower rate of collision than regular PPO. It also generalizes to new environments with movable obstacles. Qualitatively, the success rate increased from 64.5% to 83.9%, collision frequency reduced by over 70%, and path efficiency increased from 0.60 to 0.85, without suffering from unstable training behavior
M. Haddad, Dhayaa Khudher· Kufa journal of Engineering· 0 citations
This work proposes Single-rollout Autoregressive Policy Optimization (SAPO), a low-memory and compute-efficient framework in which the policy and value functions share a single autoregressive backbone, and introduces a trajectory-level generalized advantage estimator that combines lambda-returns with batch normalization.
Deploying reinforcement learning (RL) agents for heating, ventilation, and air-conditioning (HVAC) control across climatically diverse buildings is challenging, as policies trained in one climate often fail to generalize to others. We investigate transfer and meta-reinforcement learning for building climate control using Sinergym with a custom EnergyPlus model of a real room, comparing Double Deep Q-Network (DDQN) and Model-Agnostic Meta Learning DDQN (MAML-DDQN) across three transfer scenarios on a target very hot climate. Fine-tuned DDQN reduces temperature violations by a factor of 35 over the from-scratch baseline and improves reward by 86% over zero-shot transfer, while MAML-DDQN converges in less than half the fine-tuning episodes and exhibits substantially lower variance across seeds. Results highlight a key trade-off: MAML-DDQN is preferable at adaptation-focused deployments, while fine-tuned DDQN is superior when maximum performance is the objective.
Ulaş Filiz, Sajjad Baghaee, Mubeen Ahmed Fayyaz et al.· Signal Processing and Commun...· 0 citations
Reinforcement learning (RL) with verifiable rewards constructs trajectory-level advantage estimates, yet it often fails to credit the few pivotal decisions that determine outcomes in long-horizon, multi-turn agentic tasks. Recent work introduces privileged self-distillation for credit assignment, providing denser supervision, but it remains unclear how such local signals should represent sequential credit. We propose AgentOPSD, a critic-free, recursive method for turn-level credit assignment in agentic reinforcement learning. AgentOPSD aggregates token-level teacher-student log-probability gaps into turn-level evidence and recursively updates a Bayesian belief state in log-odds space. This yields a principled reweighting scheme that converts sparse outcome supervision into turn-level credit signals and identifies pivotal turns through the marginal belief revision between consecutive states. The method is fully compatible with standard policy optimization and requires neither an additional critic nor extra rollouts. We evaluate AgentOPSD on ALFWorld, WebShop, and Search-QA using Qwen2.5 models at two scales (3B and 7B). AgentOPSD outperforms GRPO and strong self-distillation baselines, achieving 89.1% success on ALFWorld with Qwen2.5-7B. Ablation studies attribute the gains to turn-level aggregation and history-dependent recursive belief updates.
Zi-Han Wang, Zhengxi Lu, Zhiyuan Yao et al.· 0 citations
