面向火箭远距离自主着陆的课程式深度强化学习方法
Curriculum-Driven Deep Reinforcement Learning for Autonomous Far-Range Rocket Landing
AI 导读
一篇 arXiv 论文提出课程驱动的深度强化学习框架，用于可复用运载火箭的远距离导航与精确着陆。该框架采用三阶段课程、critic 预热和防遗忘的重演机制，智能体从 1,000 至 1,500 单位的初始水平距离飞行并软着陆。在 100,000 回合评估中，所提策略成功率达 99.98%（基线 68.31%），剩余燃料 69.1%，落点距台中心平均 7.07 单位（基线 13.10 单位）。
正文 · 原文
Abstract:Autonomous vertical landing of reusable launch vehicles is a critical yet formidable challenge in aerospace engineering, primarily due to highly nonlinear dynamics, stringent precision require- ments, and inherent environmental uncertainties. While reinforcement learning (RL) has emerged as a promising alternative to traditional model-based control, existing approaches frequently rely on flat training paradigms and highly simplified initial conditions, drastically limiting their real-world applicability. This paper proposes a comprehensive curriculum-driven deep reinforcement learning framework designed to master both far-range navigation and precision landing. By employing a three-stage curriculum augmented with a critic warm-up procedure and a rehearsal mechanism to prevent catastrophic forgetting, alongside a high-dimensional observation vector utilizing a full 3 X 3 rotation-matrix, the agent is trained to navigate from an initial horizontal distance of 1,000 to 1,500 units vastly farther than prior benchmarks before executing a soft touchdown. Crucially, we intro- duce a redesigned reward function that reduces unnecessary computational overhead and addresses limitations of earlier formulations, including poorly calibrated reward components and potentially conflicting optimization objectives. The redesigned function bounds the per-step reward magnitude by (O(1)), while its progress term accumulates the total distance closed over the trajectory. The available 100,000-episode results report a 99.98% success rate for the proposed policy and 68.31% for the baseline under the evaluation conditions. In the evaluation, the proposed policy retained 69.1% of its initial fuel capacity and achieved a mean final distance from the pad centre of 7.07 units, compared with 13.10 units for the baseline. Landing visualizations are available in Video 1 and Video 2.
| Comments: | This is the extension of our previous paper that got accepted in ICAART 2026 |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.35954 [cs.LG] |
| (or arXiv:2610.35954v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.35954 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Atin Chowdhury Mr. [view email]
[v1]
Sat, 24 Oct 2026 19:57:44 UTC (3,862 KB)
来源：arXiv cs.AI · arxiv.org