From CAD to POMDP: Probabilistic Planning for Robotic Disassembly of End-of-Life Products

2025-11-28

AI Summary

Decommissioned products often exhibit structural degradation (e.g., wear, corrosion), causing CAD model inaccuracies and introducing uncertainty into robotic disassembly planning. Method: This work formulates robotic disassembly planning as a Partially Observable Markov Decision Process (POMDP) for the first time, incorporating latent variables to represent unknown structural states. We integrate CAD-based geometric reasoning, multimodal sensor perception (e.g., vision, force/tactile), and robot motion capabilities into a task-and-motion co-planning framework. Bayesian filtering enables online belief-state updates, while reinforcement learning synthesizes robust disassembly policies. Results: Evaluated on three representative end-of-life products across two distinct robotic platforms, our approach reduces average disassembly time by 23.6% and execution variance (standard deviation) by 41.2% compared to deterministic methods. It demonstrates strong cross-platform generalizability and robustness under realistic degradation conditions.

Technology Category

• Intelligent Robots: State Estimation
• Planning, Routing, and Scheduling: Planning with Markov Models (MDPs, POMDPs)
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

To support the circular economy, robotic systems must not only assemble new products but also disassemble end-of-life (EOL) ones for reuse, recycling, or safe disposal. Existing approaches to disassembly sequence planning often assume deterministic and fully observable product models, yet real EOL products frequently deviate from their initial designs due to wear, corrosion, or undocumented repairs. We argue that disassembly should therefore be formulated as a Partially Observable Markov Decision Process (POMDP), which naturally captures uncertainty about the product's internal state. We present a mathematical formulation of disassembly as a POMDP, in which hidden variables represent uncertain structural or physical properties. Building on this formulation, we propose a task and motion planning framework that automatically derives specific POMDP models from CAD data, robot capabilities, and inspection results. To obtain tractable policies, we approximate this formulation with a reinforcement-learning approach that operates on stochastic action outcomes informed by inspection priors, while a Bayesian filter continuously maintains beliefs over latent EOL conditions during execution. Using three products on two robotic systems, we demonstrate that this probabilistic planning framework outperforms deterministic baselines in terms of average disassembly time and variance, generalizes across different robot setups, and successfully adapts to deviations from the CAD model, such as missing or stuck parts.

Problem

• Planning robotic disassembly under uncertainty for end-of-life products
• Handling deviations from CAD models due to wear and undocumented repairs
• Formulating disassembly as POMDP with hidden structural properties

Innovation

• Formulates robotic disassembly as POMDP problem
• Derives POMDP models automatically from CAD data
• Uses reinforcement learning with Bayesian filtering