Jul 2026
Vision-TL-Action: Neuro-Symbolic Trajectory Generation from Visual Observations and Temporal Logic
Vision-TL-Action is introduced, which generates action trajectories from multi-view images, a coordinate-free TL syntax graph, and the robot initial state, demonstrating a direct mapping from visual observations and structured TL goals to action trajectories without requiring object geometry at inference.
Zezhi Liu, Zhiwei Zheng, Hanqian Luo et al.
· arXiv.org · 2 citations