Safe Reinforcement Learning With Dual Robustness via Robust Control Barrier Functions

稳健性(进化) 控制理论(社会学) 鲁棒控制 计算机科学 对偶(语法数字) 控制工程 强化学习 控制系统 控制(管理) 工程类 夹持器 自适应控制
作者
Lifei Dai,Yiqun Liu,Peiyu Cui,Hao Zhang,Zhuping Wang,Changzhu Zhang
出处
期刊:IEEE Transactions on Automation Science and Engineering [Institute of Electrical and Electronics Engineers]
卷期号:23: 18331-18343
标识
DOI:10.1109/tase.2026.3741869
摘要

In this paper, we propose a novel framework for dual-robust safe reinforcement learning via robust control barrier functions. This framework eliminates the gradient conflicts caused by the inconsistency between the adversarial attack terms in the performance and safety objectives, thereby enabling the use of the Lagrange multiplier method for robust safe reinforcement learning. The key idea is to reformulate the robust safety constraint using a robust discrete-time control barrier function (RDTCBF), which replaces the safety-related adversarial term with an adversary-independent worst-case safety margin by leveraging the Lipschitz continuity of the RDTCBF. We then analyze the conservatism induced by the Lipschitz relaxation, and further develop a local Lipschitz bound, leading to a less conservative safety constraint while preserving robustness Finally, we employ a state-dependent Lagrange multiplier to transform the constrained problem into an unconstrained minimax formulation, and theoretically show that the state-wise safety constraints are not relaxed into a global average constraint. Experiments on safety-critical benchmarks demonstrate that the proposed method improves performance by an average of 10.40% over state-of-the-art baselines while maintaining robust safety. Real-world quadrotor experiments further validate its effectiveness. Note to Practitioners—Safe reinforcement learning provides an effective approach to safety-critical control problems. However, policies trained in simulation may experience performance degradation and safety violations when deployed in real-world systems due to model errors and external disturbances. To address this challenge, this paper presents a robust control barrier function-based safe reinforcement learning method that enhances the robustness of both task performance and safety within a unified adversarial learning framework, thereby facilitating practical deployment in autonomous systems.

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
1秒前
陆陆完成签到,获得积分10
1秒前
Alex_yw完成签到,获得积分10
1秒前
单薄傲易发布了新的文献求助30
1秒前
Savananh_Xie的应助被zxw采纳,获得10
2秒前
Nole的应助被明亮的迎松采纳,获得10
2秒前
Catea发布了新的文献求助10
3秒前
3秒前
4秒前
小圆发布了新的文献求助10
4秒前
星辰大海的应助被白白白采纳,获得10
4秒前
wangyongfang完成签到,获得积分10
6秒前
无花果的应助被木土采纳,获得10
6秒前
安静半双发布了新的文献求助10
7秒前
胡萝卜完成签到 ,获得积分10
7秒前
7秒前
酷波er的应助被平淡的画板采纳,获得50
8秒前
9秒前
研友_VZG7GZ的应助被执笔绘朝暮采纳,获得10
9秒前
小鱼发布了新的文献求助10
10秒前
cadr完成签到,获得积分10
10秒前
wanci的应助被耳东采纳,获得10
10秒前
小蘑菇的应助被mate采纳,获得10
11秒前
传奇3的应助被大果冻采纳,获得10
11秒前
外科老白发布了新的文献求助10
11秒前
13秒前
13秒前
Carmen发布了新的文献求助30
13秒前
14秒前
CipherSage的应助被蛋黄蟹采纳,获得30
14秒前
XZQ发布了新的文献求助10
15秒前
Rr发布了新的文献求助10
15秒前
yjt发布了新的文献求助10
16秒前
姜姜姜姜发布了新的文献求助10
17秒前
jing发布了新的文献求助10
17秒前
17秒前
箭剑尖发布了新的文献求助10
17秒前
17秒前
一十六发布了新的文献求助20
18秒前
18秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1700
The World Atlas of Wine 9th edition 1000
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
八旗制度の研究 600
高分子基功能界面水蒸发材料与应用研究 500
外场增强的水汽发电研究 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8184240
求助须知:如何正确求助?哪些是违规求助? 9573297
关于积分的说明 21830585
捐赠科研通 7781300
什么是DOI,文献DOI怎么找? 3434285
关于科研通互助平台的介绍 2558389
邀请新用户注册赠送积分活动 2480814