arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2607.02472cs.RO
收藏

利用可微四旋翼动力学学习敏捷入侵者拦截

Learning Agile Intruder Interception using Differentiable Quadrotor Dynamics

Carnegie Mellon University(卡内基梅隆大学);Artificial Intelligence Integration Center(人工智能集成中心)

Michael Anoruo, Xiaoyu Tian, Abhishek Rathod, Timothy Naudet, Thomas Canchola, Eric Sturzinger, Kshitij Goel, Wennie Tabib

AI总结:

提出一种基于可微四旋翼动力学的解析策略梯度方法,仅利用入侵者方向向量学习高速(10 m/s)敏捷拦截策略,相比简化质点动力学方法平均提升30%。

摘要:

本文提出了一种学习控制策略的方法,该方法利用指向入侵者的3D方向单位向量和拦截器状态来拦截入侵者。先前的深度强化学习方法假设相对位置或到入侵者的距离可用,但在使用被动单目相机传感器的实际应用中,这些信息不易获得。相反,我们提出了一种解决方案,利用可微四旋翼动力学的解析策略梯度方法,学习速度高达10 m/s的敏捷拦截。所提出的方法比利用简化质点动力学的基线方法平均提升30%。

英文摘要:

This paper presents a methodology for learning a control policy to intercept an intruder using the 3D direction unit vector to the intruder and the interceptor state. Prior deep reinforcement learning approaches assume either relative position or distance to the intruder is available, but this information is not readily accessible in real-world applications that employ passive, monocular camera sensors. Instead, we propose a solution that leverages an analytical policy gradient method using differentiable quadrotor dynamics to learn agile interception at speeds up to 10 m/s. The proposed approach outperforms baseline methods that utilize simplified point mass dynamics by an average of 30%.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑