外观
论文arXiv 2609.146471 个月前
腿式机器人强化学习的技能组合
原文链接：Skill Composition for Legged Robot Reinforcement Learning（arXiv 2609.14647 · 发布日期：2026-09-13）
Skill Composition for Legged Robot Reinforcement Learning
本文聚焦腿式机器人强化学习中专用技能间脆弱切换，主张研究独立子策略的可靠组合，摘要未报告实验验证。
核心要点
- 单项行为通常由专用控制器训练，训练快、收敛可靠且可单独验证。
- 核心问题是技能之间的转换仍然脆弱，不能只作为临时实现细节处理。
- 作者主张将独立子策略的可靠组合视为研究问题，使控制可安全传递。
- 若可随时在策略间切换，下一步行为选择可交给规划器、自动机或符号控制器。
英文摘要 (Abstract)
Robots, and humanoid robots in particular, are increasingly competent at individual behaviors, each obtained by training a specialized controller. A specialized skill is quick to train, converges reliably because the problem it faces is narrow, and can be validated on its own, none of which is true of a single end-to-end policy asked to cover everything. What remains fragile is the transition between them. We argue that the composition of independent sub-policies deserves to be treated as a research problem in its own right, rather than as an implementation detail left to whatever mechanism happens to be at hand. Reliable composition is what turns a collection of separate skills into a repertoire that can be used, extended and shared. More fundamentally, if control can be passed between specialized policies safely, and at any moment, the choice of what the robot should do next can be delegated to a component of an entirely different nature, such as a planner, an automaton or a symbolic controller, whose behavior can be inspected in advance. The policies would then only ever have to act, and what the robot can be trusted to do would become verifiable.
作者：Daniel Gigliotti, Flavio Maiorana, Fabio Patrizi, Luca Iocchi
来源
本页事实对应的公开页面，以原文为准。