Gaussian Processes for Improved Dynamic Modeling in the Predictive Control of an Arduino Temperature Control Lab
- Author(s)
- Idris Sadik, Armin Kuper, Steffen Waldherr
- Abstract
A practical application of Gaussian processes (GPs) as an alternative nonlinear system identification approach in model predictive control (MPC) is presented. By means of using an Arduino Temperature Control Lab, setpoint tracking accuracy for a Gaussian process-based MPC scheme is compared to state space MPC and a proportional-integral-derivative (PID) controller. Foregoing parameterized system identification, GPs are proven to offer superior accuracy, while eliminating the tedious developments plaguing first-principle nonlinear alternatives. By further utilizing GPs probabilistic framework, estimates for variance are interpreted as system-specific uncertainty and used to better select control solutions that remain in training regions. Including variance within the optimal control problem (as opposed to its exclusion), improved overall setpoint tracking and affords a more cautious controller.
- Organisation(s)
- External organisation(s)
- Katholieke Universiteit Leuven
- Pages
- 1757-1763
- No. of pages
- 7
- DOI
- https://doi.org/10.23919/ECC54610.2021.9655164
- Publication date
- 2021
- Peer reviewed
- Yes
- Austrian Fields of Science 2012
- 202034 Control engineering
- ASJC Scopus subject areas
- Control and Optimization, Artificial Intelligence, Decision Sciences (miscellaneous), Control and Systems Engineering, Mechanical Engineering, Computational Mathematics
- Portal url
- https://ucris.univie.ac.at/portal/en/publications/gaussian-processes-for-improved-dynamic-modeling-in-the-predictive-control-of-an-arduino-temperature-control-lab(7c598453-2c5e-43f4-9fd9-91d6d1f3e555).html