Guided Multi-Fidelity Bayesian Optimization for Data-driven Controller Tuning with Digital Twins

2025-09-22

AI Summary

Data-driven controller tuning suffers from low efficiency due to limited fidelity of digital twin (DT) simulations and the high cost of acquiring real-world measurements. Method: We propose a guided multi-fidelity Bayesian optimization (BO) framework. It employs a dynamically adaptive multi-fidelity surrogate model that fuses calibrated DT simulations with online real-world measurements. A cost-aware acquisition function is designed, leveraging Gaussian process regression and online learning to continuously correct DT bias while jointly estimating cross-source correlations and adapting sampling strategies. Results: Experiments on robotic actuator hardware and simulation platforms demonstrate that our method significantly improves tuning efficiency and convergence speed under limited sample budgets, outperforming standard BO and conventional multi-fidelity approaches. It achieves a favorable trade-off between accuracy and cost-effectiveness.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Intelligent Robots: Learning & Optimization for ROB
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

We propose a extit{guided multi-fidelity Bayesian optimization} framework for data-efficient controller tuning that integrates corrected digital twin (DT) simulations with real-world measurements. The method targets closed-loop systems with limited-fidelity simulations or inexpensive approximations. To address model mismatch, we build a multi-fidelity surrogate with a learned correction model that refines DT estimates from real data. An adaptive cost-aware acquisition function balances expected improvement, fidelity, and sampling cost. Our method ensures adaptability as new measurements arrive. The accuracy of DTs is re-estimated, dynamically adapting both cross-source correlations and the acquisition function. This ensures that accurate DTs are used more frequently, while inaccurate DTs are appropriately downweighted. Experiments on robotic drive hardware and supporting numerical studies demonstrate that our method enhances tuning efficiency compared to standard Bayesian optimization (BO) and multi-fidelity methods.

Problem

• Tuning controllers efficiently with limited real-world data availability
• Addressing simulation inaccuracies in digital twin-based controller optimization
• Balancing computational cost and performance improvement in controller tuning

Innovation

• Multi-fidelity Bayesian optimization with corrected digital twins
• Adaptive cost-aware acquisition function balancing improvement and cost
• Dynamic re-estimation of digital twin accuracy using new measurements