# Waypoint Optimization Using Bayesian Optimization: A Case Study in Airborne Wind Energy Systems

arXiv: https://arxiv.org/abs/1910.12901 · HTML: https://mlsanity.com/paper/1910.12901/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 28 Oct 2019.

## Built on by (most cited first, 6 of 6)

- [VABO: Violation-Aware Bayesian Optimization for Closed-Loop Control Performance Optimization with Unmodeled Constraints](https://mlsanity.com/paper/2110.07479/) (2021)
- [Violation-Aware Contextual Bayesian Optimization for Controller Performance Optimization with Unmodeled Constraints](https://mlsanity.com/paper/2301.12099/) (2023)
- [Falsification of Learning-Based Controllers through Multi-Fidelity Bayesian Optimization](https://mlsanity.com/paper/2212.14118/) (2022)
- [CONFIG: Constrained Efficient Global Optimization for Closed-Loop Control System Optimization with Unmodeled Constraints](https://mlsanity.com/paper/2211.11822/) (2022)
- [Contextual Causal Bayesian Optimisation](https://mlsanity.com/paper/2301.12412/) (2023)
- [Concurrent Learning of Policy and Unknown Safety Constraints in Reinforcement Learning](https://mlsanity.com/paper/2402.15893/) (2024)

## Abstract (arXiv)

We present a data-driven optimization framework that aims to address online adaptation of the flight path shape for an airborne wind energy system (AWE) that follows a repetitive path to generate power. Specifically, Bayesian optimization, which is a data-driven algorithm for finding the optimum of an unknown objective function, is utilized to solve the waypoint adaptation. To form a computationally efficient optimization framework, we describe each figure-8 flight via a compact set of parameters, termed as basis parameters. We model the underlying objective function by a Gaussian Process (GP). Bayesian optimization utilizes the predictive uncertainty information from the GP to determine the best subsequent basis parameters. Once a path is generated using Bayesian optimization, a path following mechanism is used to track the generated figure-8 flight. The proposed framework is validated on a simplified 2-dimensional model that mimics the key behaviors of a 3-dimensional AWE system. We demonstrate the capability of the proposed framework in a simulation environment for a simplified 2-dimensional AWE system model.