INTELLIGENT CONTROL ENGINEERING BENCHMARK: AN EDUCATIONAL EXPERIENCE

Gilberto Reynoso-Meza; Leandro dos Santos Coelho; Emerson Donaisky; Roberto Zanetti Freire

doi:10.20906/SBAI-SBSE-2023/4036

Gilberto Reynoso-Meza Industrial and Systems Engineering Graduate Program (PPGEPS), Pontifical Catholic University of Paraná, Curitiba, PR
Leandro dos Santos Coelho Industrial and Systems Engineering Graduate Program (PPGEPS), Pontifical Catholic University of Paraná, Curitiba, PR / Department of Electrical Engineering, Federal University of Paraná
Emerson Donaisky Polytechnical School, Pontifical Catholic University of Paraná (PUCPR), Curitiba, PR
Roberto Zanetti Freire Department of Electrical Engineering, Technological Federal University of Paraná, Curitiba, PR

DOI: https://doi.org/10.20906/SBAI-SBSE-2023/4036

Keywords: Controller tuning, multi-objective optimization, intelligent control

Abstract

This teaching testimony will share some experiences developing and designing an engineering control problem to assist our learning outcomes in the Mechatronics and Automatic Control curricula. The primary motivation is to bring a contextualized problem suitable to generate a learning environment where students can link the importance of abstracting a problem, solving it, and concretizing the real-world implications of their solutions. Its context is oriented to the automatic supply of propofol in children. Via simulation, the dynamics of different patients and their response to propofol infusion are analyzed to induce a state of unconsciousness. The general objective is to tune a minimum set of controllers to deal with inter-patient variability and build a regression model to identify control parameters for a new instance. Preliminary results with a pilot group showed the viability of using this benchmark for control engineering and machine learning education.