Temporal Logic Guidance for Action-Only Diffusion Policies with World Models
基于时序逻辑的动作扩散策略与世界模型引导
机构 * German Academic Exchange Service (DAAD)(德国学术交流中心(DAAD))
AI总结 提出一种利用世界模型实现时序逻辑鲁棒性可微评估的引导方法,在不重新训练的情况下改善动作扩散策略的约束满足,在Robomimic任务中将违规率从80%降至4%。
Comments Accepted at the ICRA 2026 Workshop on Bridging the Gap between Robot Learning and Human-Robot Interaction. 3 pages, 2 figures, 1 table