NAMOUnc: Navigation Among Movable Obstacles with Decision Making on Uncertainty Interval

2025-09-16

AI Summary

Existing NAMO (Navigation Among Movable Obstacles) approaches often rely on idealized assumptions—ignoring realistic uncertainties such as sensor noise, model inaccuracies, action failures, and partial observability—leading to insufficient decision robustness. This paper introduces, for the first time, uncertainty interval modeling into the NAMO decision framework, jointly optimizing lower and upper bounds of time cost alongside task success probability to enable risk-aware path planning and obstacle manipulation co-decision. We propose a four-stage algorithm: uncertainty estimation, interval comparison, success-probability optimization, and decision fusion, validated in both simulation and real-robot experiments. Results show that our method improves navigation success rate by 23.6% and reduces average task completion time by 17.4% compared to state-of-the-art NAMO methods, significantly enhancing safety and adaptability in complex, dynamic environments.

Technology Category

• Planning, Routing, and Scheduling: Planning under Uncertainty
• Intelligent Robots: Motion and Path Planning
• Reasoning under Uncertainty: Sequential Decision Making

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

Navigation among movable obstacles (NAMO) is a critical task in robotics, often challenged by real-world uncertainties such as observation noise, model approximations, action failures, and partial observability. Existing solutions frequently assume ideal conditions, leading to suboptimal or risky decisions. This paper introduces NAMOUnc, a novel framework designed to address these uncertainties by integrating them into the decision-making process. We first estimate them and compare the corresponding time cost intervals for removing and bypassing obstacles, optimizing both the success rate and time efficiency, ensuring safer and more efficient navigation. We validate our method through extensive simulations and real-world experiments, demonstrating significant improvements over existing NAMO frameworks. More details can be found in our website: https://kai-zhang-er.github.io/namo-uncertainty/

Problem

• Addressing navigation uncertainties in movable obstacle environments
• Optimizing decision-making between obstacle removal and bypassing
• Improving success rate and time efficiency in robotics navigation

Innovation

• Estimates uncertainties for obstacle removal and bypassing
• Compares time cost intervals to optimize decisions
• Integrates uncertainty into decision-making for safer navigation