Human-Centric Validation of Reinforcement Learning–Based Control in Fluid Mechatronics: An Experimental Case Study

DOI

10.2991/978-94-6463-940-7_28How to use a DOI?

Keywords

Reinforcement Learning; Human-in-the-Loop Control; Experimental Validation; Fluid Mechatronics; Case Study

Abstract

Reinforcement Learning (RL) has emerged as a promising paradigm for process automation, yet its experimental validation with human operators in real-time control scenarios remains limited. This work presents a human-in-the-loop case study that evaluates the performance and usability of an RL based controller implemented on a laboratory-scale fluid mechatronics system. A user-friendly web interface was developed to allow participants to monitor and manipulate the system by toggling an RL agent switch, adjusting pump speed, and responding to disturbances introduced by the operator. Experiments were conducted with graduate and undergraduate students, each tasked with maintaining the system pressure within a specified setpoint range. Detailed logs and sensor measurements, including pressure, flow, and tank level, were collected alongside user actions. Initial experiments revealed logging errors that produced misleading spikes in recorded data; corrective modifications to callback functions were introduced, resulting in more accurate representation of user interventions. The results demonstrate that RL assistance improved system recovery under disturbances, while human operators contributed critical corrective actions during unexpected dynamics. The study highlights the synergy between intelligent automation and human adaptability, providing insights into the robustness, usability, and educational value of RL based control systems.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Ponnada Swaroop Chandra
AU  - Gandroju Mahalakshmi Sree
AU  - Mantena Sireesha
AU  - Purushottama Rao Dasari
PY  - 2025
DA  - 2025/12/31
TI  - Human-Centric Validation of Reinforcement Learning–Based Control in Fluid Mechatronics: An Experimental Case Study
BT  - Proceedings of the Conference on Social and Sustainable Innovation in Technology & Engineering (SASI-ITE 2025)
PB  - Atlantis Press
SP  - 374
EP  - 385
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-940-7_28
DO  - 10.2991/978-94-6463-940-7_28
ID  - Chandra2025
ER  -