UniPlan: Vision-Language Task Planning for Mobile Manipulation with Unified PDDL Formulation

TL;DR

UniPlan is a comprehensive vision-language planning system that extends symbolic manipulation to large-scale indoor environments, enabling mobile robot tasks with improved success rates and efficiency.

Contribution

It introduces a unified PDDL framework for long-horizon mobile manipulation, integrating scene topology, visuals, and robot capabilities for complex indoor tasks.

Findings

Outperforms existing methods in success rate and plan quality.

Operates efficiently on large-scale real-world maps.

Successfully handles navigation, door traversal, and bimanual tasks.

Abstract

Integration of VLM reasoning with symbolic planning has proven to be a promising approach to real-world robot task planning. Existing work like UniDomain effectively learns symbolic manipulation domains from real-world demonstrations, described in Planning Domain Definition Language (PDDL), and has successfully applied them to real-world tasks. These domains, however, are restricted to tabletop manipulation. We propose UniPlan, a vision-language task planning system for long-horizon mobile-manipulation in large-scale indoor environments, that unifies scene topology, visuals, and robot capabilities into a holistic PDDL representation. UniPlan programmatically extends learned tabletop domains from UniDomain to support navigation, door traversal, and bimanual coordination. It operates on a visual-topological map, comprising navigation landmarks anchored with scene images. Given a language instruction, UniPlan retrieves task-relevant nodes from the map and uses a VLM to ground the anchored image into task-relevant objects and their PDDL states; next, it reconnects these nodes to a compressed, densely-connected topological map, also represented in PDDL, with connectivity and costs derived from the original map; Finally, a mobile-manipulation plan is generated using off-the-shelf PDDL solvers. Evaluated on human-raised tasks in a large-scale map with real-world imagery, UniPlan significantly outperforms VLM and LLM+PDDL planning in success rate, plan quality, and computational efficiency.