POrTAL: Plan-Orchestrated Tree Assembly for Lookahead

2025/12/02 by Conway, Evan, Porfirio, David, Chan, David +2
Computer Science · Engineering · #AI-based Problem Solving and Planning #Artificial Intelligence (cs.AI) #Baseline (sea) #Computational complexity theory #FOS: Computer and information sciences #Modular Robots and Swarm Intelligence #Plan (archaeology) #Probabilistic logic #Robot #Robotic Path Planning Algorithms #Robotics (cs.RO) #Series (stratigraphy) #Tree (set theory)

paper · open access · doi:10.48550/arxiv.2512.06002

published in arXiv (Cornell University) (Cornell University)

openalex publication_date 2025/12/02 · openalex created_date 2025/12/10 · openalex updated_date 2026/07/28

Abstract

When tasking robots in partially observable environments, these robots must efficiently and robustly plan to achieve task goals under uncertainty. Although many probabilistic planning algorithms exist for this purpose, these algorithms can be inefficient if executed with the robot's limited computational resources, or may produce policies that take more steps than expected to achieve the goal. We therefore created a new, lightweight, probabilistic planning algorithm, Plan-Orchestrated Tree Assembly for Lookahead (POrTAL), that combines the strengths of two baseline planning algorithms, FF-Replan and POMCP. We demonstrate that POrTAL is an anytime algorithm that generally outperforms these baselines in terms of the final executed plan length given bounded computation time, especially for problems with only moderate levels of uncertainty.

Citations

• Learning-Augmented Model-Based Multi-Robot Planning for Time-Critical Search and Inspection Under Uncertainty
• Uncertainty Expression for Human-Robot Task Communication
• Human-Agent Coordination in Games under Incomplete Information via Multi-Step Intent
• Task-Directed Exploration in Continuous POMDPs for Robotic Manipulation of Articulated Objects
• Habitat-Web: Learning Embodied Object-Search Strategies from Human Demonstrations at Scale
• Online algorithms for POMDPs with continuous state, action, and observation spaces
• QMDP-Net: Deep Learning for Planning under Partial Observability
• Planning and acting in partially observable stochastic domains

Related

• Tru-POMDP: Task Planning Under Uncertainty via Tree of Hypotheses and Open-Ended POMDPs
• Probabilistic roadmaps for path planning in high-dimensional configuration spaces
• Vote-Tree-Planner: Optimizing Execution Order in LLM-based Task Planning Pipeline via Voting
• Sampling-based path planning under temporal logic constraints with real-time adaptation
• Combining Large Language Models and Symbolic Reasoning for Multi-Robot Temporal Planning through Explainable Knowledge Bases