Problem
Research questions and friction points this paper is trying to address.
Linear Temporal Logic
Robot Learning
Inference-Time Guidance
Safety Constraints
Short-Horizon Policies
hint$^2$: Hierarchical World Models for Inference-Time Temporal Logic Guidance
🤖 AI Summary
This study addresses the challenge of satisfying complex temporal logic and safety constraints during robotic policy inference by proposing a dual-objective guidance mechanism based on hierarchical world models. The approach decouples long-horizon task progress from short-term local safety through high- and low-level abstractions, leveraging Linear Temporal Logic and atomic proposition predictions to guide diffusion policies toward generating compliant actions. Experimental evaluations on both the CALVIN benchmark and a real-world UR5e manipulator demonstrate that this method significantly outperforms existing baselines. It effectively ensures reliable execution under complex liveness and safety constraints, thereby substantially enhancing the policy's capacity to satisfy temporal logic specifications in practical robotic applications.
📝 Abstract
A central goal of robot learning is to enable robots to execute rich instructions specified at runtime. Large-scale language-conditioned policies have made substantial progress toward this goal, yet still struggle with temporal structure and safety constraints. Linear Temporal Logic (LTL) provides a powerful language to express complex, non-Markovian instructions. However, guiding learned manipulation policies toward LTL satisfaction remains challenging because modern policies generate short-horizon action chunks and replan in closed loop, while almost all LTL specifications are evaluated over long-horizon trajectories. In this paper, we introduce hint$^2$, a method for guiding short-horizon policies toward satisfying complex LTL specifications at inference time using hierarchical world models. Our key idea is to derive two separate guidance objectives using each world model's abstraction level. A high-level model predicts future action-induced transitions in task-relevant atomic propositions to guide progress through the LTL automaton, while a low-level dynamics model predicts immediate state evolution for accurate local safety guidance. Our results show that hint$^2$ overcomes the limitations of current LTL-guided diffusion methods, outperforms existing inference-time steering methods in CALVIN, and successfully completes instructions with complex liveness and safety constraints more elegantly than language-conditioned alternatives. Finally, we demonstrate that hint$^2$ can handle complex instructions on a real UR5e manipulator.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Hierarchical World Models
Linear Temporal Logic
Inference-Time Guidance
Short-Horizon Policies
Robot Manipulation
🔎 Similar Papers
No similar papers found.
