MORN: Metacognitive Object-Goal Regulation for Resource-Rational Long-Horizon Navigation

2026-05-16

AI Summary

This work addresses the failure of existing zero-shot object navigation agents in partially observable environments, where a lack of global resource awareness leads them to persistently pursue infeasible goals until resources are exhausted. Inspired by dual-process theory in cognitive science, we propose MORN—a novel architecture that introduces metacognitive mechanisms into object navigation for the first time. MORN employs a meta-controller that dynamically integrates online progress estimates with perceptual uncertainty to instantiate three neurocognitive states: potential indexing, persistence gating, and evidence accumulation. This enables resource-rational goal scheduling, effectively avoiding the sunk-cost fallacy by facilitating early abandonment of unattainable (“zombie”) targets and focused pursuit of viable ones. Evaluated on the HM3D dataset, our method improves the success rate from 0.23 to 0.30 and reduces the proportion of ineffective steps from 0.90 to 0.70, significantly enhancing navigation efficiency under resource constraints.

Abstract

Robots deployed in unstructured human environments must frequently execute long-horizon missions, such as find the mug, then the chair, then the printer, under strict operational constraints. While contemporary zero-shot Object Navigation (ObjectNav) agents leverage Vision-Language Models (VLMs) to effectively localize semantic targets, they operate as purely reactive systems that inherently lack global resource awareness. Consequently, these agents inadvertently exhaust critical budgets, including time and battery, on infeasible subgoals due to partial observability, failing to balance local exploration with global mission viability. To bridge this gap by injecting resource-rationality into the navigation loop, we present MORN (Metacognitive Object-goal Regulation Navigation), an executive architecture inspired by Dual-Process Theory in cognitive science. MORN augments frozen navigation backbones with a System 2 meta-controller that continuously monitors the System 1 locomotor. By formalizing three neuro-cognitive states, Potentiality Index, Persistence Gating, and Evidence Accumulation, MORN dynamically regulates the mission schedule based on online estimates of progress velocity and perceptual uncertainty. This mechanism effectively neutralizes the Sunk Cost Fallacy, enabling agents to abort zombie goals early and decisively commit to achievable ones. Extensive experiments on the HM3D dataset demonstrate that MORN improves Goal Completion Rate (CR) from 0.23 to 0.30 and reduces Wasted Step Fraction (WSF) from 0.90 to 0.70, establishing that in resource-constrained autonomy, the metacognitive awareness of global resources is as critical as the reactive ability to navigate.

Problem

• long-horizon navigation
• resource-rationality
• object-goal regulation
• metacognitive control
• zero-shot ObjectNav

Innovation

• Metacognitive Regulation
• Resource-Rational Navigation
• Object Goal Abandonment
• Dual-Process Architecture
• Long-Horizon Navigation