Bayesian Optimization Parameter Tuning Framework for a Lyapunov Based Path Following Controller

2025-12-14

AI Summary

Manual tuning of nonlinear geometric path-following controllers under hardware constraints suffers from strong gain coupling, low efficiency, and high experimental cost. Method: This paper proposes a Bayesian optimization–based automated parameter tuning framework: the closed-loop system is treated as a black box; a Gaussian process surrogate model enables data-efficient, uncertainty-aware search over high-dimensional, coupled controller gains; and—uniquely—the optimization is tightly integrated with a Lyapunov-based geometric controller. Contribution/Results: Evaluated on a real Honda AI-Formula robotic platform, the method achieves significant improvements in tracking accuracy and closed-loop stability using only 32 physical trials (including 15 warm-up runs), successfully completing a full-lap track mission. Results demonstrate the approach’s applicability, practicality, and robustness for strongly nonlinear, highly coupled systems.

Technology Category

• Intelligent Robots: Learning & Optimization for ROB
• Search and Optimization: Sampling/Simulation-based Search
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Parameter tuning in real-world experiments is constrained by the limited evaluation budget available on hardware. The path-following controller studied in this paper reflects a typical situation in nonlinear geometric controller, where multiple gains influence the dynamics through coupled nonlinear terms. Such interdependence makes manual tuning inefficient and unlikely to yield satisfactory performance within a practical number of trials. To address this challenge, we propose a Bayesian optimization (BO) framework that treats the closed-loop system as a black box and selects controller gains using a Gaussian-process surrogate. BO offers model-free exploration, quantified uncertainty, and data-efficient search, making it well suited for tuning tasks where each evaluation is costly. The framework is implemented on Honda's AI-Formula three-wheeled robot and assessed through repeated full-lap experiments on a fixed test track. The results show that BO improves controller performance within 32 trials, including 15 warm-start initial evaluations, indicating that it can efficiently locate high-performing regions of the parameter space under real-world conditions. These findings demonstrate that BO provides a practical, reliable, and data-efficient tuning approach for nonlinear path-following controllers on real robotic platforms.

Problem

• Bayesian optimization tunes nonlinear path-following controller gains efficiently
• It addresses costly manual tuning under limited real-world evaluation budgets
• The framework improves performance on a physical robot within few trials

Innovation

• Bayesian optimization for black-box controller tuning
• Gaussian-process surrogate for data-efficient parameter search
• Real-world implementation on robotic platform with limited trials