LTLDoG: Satisfying Temporally-Extended Symbolic Constraints for Safe Diffusion-based Planning
This work addresses safe robot deployment in complex environments by enabling flexible test-time adaptation to new constraints, representing an incremental improvement in constraint-satisfying planning methods.
The authors tackled the problem of generating long-horizon robot trajectories that adhere to novel static and temporally-extended constraints specified using linear temporal logic, resulting in a diffusion-based framework that successfully satisfies formulae for obstacle avoidance and visitation sequences in navigation and manipulation tasks.
Operating effectively in complex environments while complying with specified constraints is crucial for the safe and successful deployment of robots that interact with and operate around people. In this work, we focus on generating long-horizon trajectories that adhere to novel static and temporally-extended constraints/instructions at test time. We propose a data-driven diffusion-based framework, LTLDoG, that modifies the inference steps of the reverse process given an instruction specified using finite linear temporal logic ($\text{LTL}_f$). LTLDoG leverages a satisfaction value function on $\text{LTL}_f$ and guides the sampling steps using its gradient field. This value function can also be trained to generalize to new instructions not observed during training, enabling flexible test-time adaptability. Experiments in robot navigation and manipulation illustrate that the method is able to generate trajectories that satisfy formulae that specify obstacle avoidance and visitation sequences. Code and supplementary material are available online at https://github.com/clear-nus/ltldog.