An Aircraft Upset Recovery System with Reinforcement Learning
For aircraft control engineers, this work provides an AI-driven alternative to conventional upset recovery systems, though it is an incremental improvement over existing methods.
The authors developed a reinforcement learning-based pilot activated recovery system for advanced jet trainers that outperforms conventional control methods in terms of behavior desirability as evaluated by domain experts.
This article explores the progress made in the creation of a pilot activated recovery system (PARS) for advanced jet trainers that utilizes artificial intelligence (AI) in an effort to enhance operational efficiency. The PARS model employs an advanced reinforcement learning (RL) architecture, incorporating a cutting-edge soft-actor critic (SAC) model and hyper-parameter optimization methods. Negative-g punishments and other handcrafted features remarked upon by control engineers and domain experts regarding PARS are also taken into account by the system. When evaluated by them, the AI model's behavior is deemed more desirable than that of conventional control methods.