研究人员开发了新的方法来提高扩散策略在机器人操作中的安全性和性能。PACT是一个训练后框架，通过将策略投影到约束可行的区域来增强安全性，将违规行为减少31%，同时将任务成功率提高30.7%。潜在扩散策略（LDP）通过在成形潜在空间中分离场景理解和轨迹生成来简化学习，在复杂的协调任务上表现优于先前的方法。此外，WorldDP将以物体为中心的世界模型与扩散策略集成，以实现多阶段机器人任务的层次化规划，在性能上优于现有基线。 AI
影响 这些机器人操作AI的进步可能导致更安全、更强大的机器人在复杂的现实世界任务中执行任务。
排序理由 arXiv上发表了多篇研究论文，详细介绍了机器人操作AI的新方法。
AI 生成摘要 · Google Gemini · 来自 7 个来源。 我们如何撰写摘要 →
arXiv:2606.10025v1 Announce Type: cross Abstract: We present GHOST, a framework for learning visuomotor manipulation policies that generalize beyond the training distribution. GHOST factorizes control into (i) a high-level policy that predicts the next sub-goal as a distribution …
arXiv:2606.08414v1 Announce Type: cross Abstract: Diffusion policies have achieved remarkable success in robotic manipulation, yet they often fail to satisfy strict physical constraints required for safe deployment. Existing approaches impose safety either prematurely during trai…
arXiv:2606.08657v1 Announce Type: cross Abstract: Diffusion-based visuomotor policies operating directly in raw action spaces conflate scene comprehension with trajectory generation within a single denoising process. The resulting velocity field must simultaneously encode scene i…
arXiv:2606.08775v1 Announce Type: cross Abstract: Visual world models have shown great potential in learning complex system dynamics. Recent advancements leverage these models as transition functions within Model Predictive Control (MPC) frameworks to solve various control tasks.…
Diffusion policies have achieved remarkable success in robotic manipulation, yet they often fail to satisfy strict physical constraints required for safe deployment. Existing approaches impose safety either prematurely during training or reactively via external guardrails at test…
arXiv:2606.05660v1 Announce Type: cross Abstract: Embodied AI systems are increasingly expected to reason and act over extended horizons in physical environments. This growing capability brings safety to the foreground, because failures in the physical world can harm people, dama…
arXiv:2512.06628v3 Announce Type: replace-cross Abstract: Scalable embodied intelligence is constrained by the scarcity of diverse, long-horizon robotic manipulation data. Existing video world models in this domain are limited to synthesizing short clips of simple actions and oft…