Journal Article10.48550/arxiv.2404.00232
Efficient Automatic Tuning for Data-driven Model Predictive Control via Meta-Learning
1
TL;DR: AutoMPC's efficiency and stability are improved using meta-learning-based Portfolio optimization, which stabilizes the tuning process and optimizes initial designs for BO.
read more
Abstract: AutoMPC is a Python package that automates and optimizes data-driven model predictive control. However, it can be computationally expensive and unstable when exploring large search spaces using pure Bayesian Optimization (BO). To address these issues, this paper proposes to employ a meta-learning approach called Portfolio that improves AutoMPC's efficiency and stability by warmstarting BO. Portfolio optimizes initial designs for BO using a diverse set of configurations from previous tasks and stabilizes the tuning process by fixing initial configurations instead of selecting them randomly. Experimental results demonstrate that Portfolio outperforms the pure BO in finding desirable solutions for AutoMPC within limited computational resources on 11 nonlinear control simulation benchmarks and 1 physical underwater soft robot dataset.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Enhancing motion planning efficiency in dynamic environments through advanced algorithms for mobile robots
TL;DR: This thesis introduces three novel motion planning approaches for mobile robots in dynamic environments, leveraging reactive navigation, Deep Reinforcement Learning, and Nonlinear Model Predictive Control to enhance efficiency and adaptability in complex tasks.
References
Sequential model-based optimization for general algorithm configuration
Frank Hutter,Holger H. Hoos,Kevin Leyton-Brown +2 more
- 17 Jan 2011
TL;DR: This paper extends the explicit regression models paradigm for the first time to general algorithm configuration problems, allowing many categorical parameters and optimization for sets of instances, and yields state-of-the-art performance.
•Proceedings Article
Efficient and robust automated machine learning
Matthias Feurer,Aaron Klein,Katharina Eggensperger,Jost Tobias Springenberg,Manuel Blum,Frank Hutter +5 more
- 07 Dec 2015
TL;DR: This work introduces a robust new AutoML system based on scikit-learn, which improves on existing AutoML methods by automatically taking into account past performance on similar datasets, and by constructing ensembles from the models evaluated during the optimization.
•Posted Content
Benchmark Environments for Multitask Learning in Continuous Domains
TL;DR: This work describes a benchmark set of tasks that is developed in an extendable framework based on OpenAI Gym and releases the framework publicly to be expanded and used for the systematic comparison of multitask, transfer, and lifelong learning in continuous domains.
36
Automatic Tuning for Data-driven Model Predictive Control
William R. Edwards,Gao Tang,Giorgos Mamakoukas,Todd D. Murphey,Kris Hauser +4 more
- 30 May 2021
TL;DR: In this article, the authors present a method to jointly optimize the data-driven system identification, task specification, and control synthesis of unknown dynamical systems, and they use their method to develop AutoMPC3, a software package designed to automate and optimize datadriven MPC.
36
Automatically-Tuned Model Predictive Control for an Underwater Soft Robot
W. D. Null,William Edwards,Dohun Jeong,Teodor Tchalakov,James Menezes,Kris Hauser,Y. Z +6 more
TL;DR: This study develops an automatically-tuned model predictive control (MPC) method, AutoMPC, for controlling an underwater soft robot, achieving higher accuracy and reliability than state-of-the-art baselines in- and out-of-distribution of training data.
7