Deployment-Time Reliability of Learned Robot Policies

2026-03-11

AI Summary

This work addresses the insufficient reliability of learned robotic policies during deployment, which stems from distribution shifts, error accumulation, and task complexity. To mitigate these challenges, the authors propose a suite of deployment-time enhancement mechanisms: a runtime behavioral consistency monitor that enables failure prediction without requiring failure data, an influence-function-based, data-centric explainability framework that traces policy performance back to individual training samples, and a language-specified task execution module integrating success-rate estimation with feasibility-aware planning. Together, these components substantially improve robustness and success rates in long-horizon and language-guided tasks, while also enabling interpretable policy diagnostics and targeted dataset refinement.

Technology Category

• Intelligent Robots: Learning & Optimization for ROB
• Natural Language Processing: Safety and Robustness
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Recent advances in learning-based robot manipulation have produced policies with remarkable capabilities. Yet, reliability at deployment remains a fundamental barrier to real-world use, where distribution shift, compounding errors, and complex task dependencies collectively undermine system performance. This dissertation investigates how the reliability of learned robot policies can be improved at deployment time through mechanisms that operate around them. We develop three complementary classes of deployment-time mechanisms. First, we introduce runtime monitoring methods that detect impending failures by identifying inconsistencies in closed-loop policy behavior and deviations in task progress, without requiring failure data or task-specific supervision. Second, we propose a data-centric framework for policy interpretability that traces deployment-time successes and failures to influential training demonstrations using influence functions, enabling principled diagnosis and dataset curation. Third, we address reliable long-horizon task execution by formulating policy coordination as the problem of estimating and maximizing the success probability of behavior sequences, and we extend this formulation to open-ended, language-specified tasks through feasibility-aware task planning. By centering on core challenges of deployment, these contributions advance practical foundations for the reliable, real-world use of learned robot policies. Continued progress on these foundations will be essential for enabling trustworthy and scalable robot autonomy in the future.

Problem

• deployment-time reliability
• distribution shift
• compounding errors
• task dependencies
• learned robot policies

Innovation

• deployment-time reliability
• runtime monitoring
• influence functions
• policy interpretability
• feasibility-aware planning