发表机构
Università di Siena; Uninettuno University; École Polytechnique Fédérale de Lausanne (EPFL)（锡耶纳大学; 尤尼内图诺大学; 洛桑联邦理工学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出JESSI，一种端到端多任务强化学习框架，直接利用LiDAR进行安全社交导航，通过概率感知与Dirichlet动作空间提升安全性与社交行为，并在仿真和实际中验证其优于现有基线。
AI 中文摘要
自主社交导航需要在效率、物理安全和社会合规性之间取得平衡。强化学习（RL）方法提供了一种可行且有效的解决方案，但往往依赖于不切实际的假设，例如对行人位置和速度的已知。在本文中，我们介绍了JESSI（基于JAX的端到端安全社交可解释导航），一个轻量级的端到端RL框架，直接将原始LiDAR扫描映射到运动学可行的控制命令。JESSI通过Dirichlet参数化的连续动作空间和确定性边界增强安全性，同时集成的基于注意力的感知模块提取概率性行人状态，以实现可解释的、具有社交意识的决策。通过大量仿真和在差速驱动机器人上的实际部署，我们证明在多任务范式中联合优化RL策略与监督感知信号能增强社交行为。最终，与最先进的基线相比，JESSI能够平衡高导航成功率与优越的社交行为。
英文摘要
Autonomous social navigation requires balancing efficiency, physical safety, and social compliance. Reinforcement Learning (RL) methods provide a viable and effective solution but often rely on unrealistic assumptions, such as the knowledge of humans' position and velocity. In this paper, we introduce JESSI (JAX-based E2E Safe Social Interpretable navigation), a lightweight end-to-end RL framework that maps raw LiDAR scans directly to kinematically feasible control commands. JESSI enhances safety via Dirichlet-parameterized continuous action spaces and deterministic bounding, while an integrated attention-based perception module extracts probabilistic human states for interpretable, socially aware decision-making. Through extensive simulations and real-world deployment on a differential-drive robot, we demonstrate that jointly optimizing the RL policy with a supervised perception signal in a multi-task paradigm enhances social behavior. Ultimately, JESSI is able to balance high navigation success rates and superior social behaviors compared to state-of-the-art baselines.
CommentsAccepted to the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2026
