Fractional Order Optimal Controller Design for Cyber Physical Cardiac Pacemaker using a Selection Hyper-Heuristic
2026 10th International Artificial Intelligence and Data Processing Symposium (IDAP), Malatya, Türkiye, 6 - 07 Eylül 2026, ss.1-5, (Tam Metin Bildiri)
- Yayın Türü: Bildiri / Tam Metin Bildiri
- Doi Numarası: 10.1109/idap71869.2026.11728067
- Basıldığı Şehir: Malatya
- Basıldığı Ülke: Türkiye
- Sayfa Sayıları: ss.1-5
- İnönü Üniversitesi Adresli: Evet
Özet
The cardiac pacemaker is a safety-critical cyberphysical device that requires robust and precise regulation of its response. In this study, the fractional order PID (FOPID) controller parameters were optimized with a hyper-heuristic approach to the robust control of the cardiac pacemaker. A selection hyper-heuristic method was employed, which manages a set of low-level optimization algorithms and dynamically applies the chosen algorithm during the search process, instead of a single optimization algorithm. Atom Search Optimization (ASO), Geometric Mean Optimizer (GMO), Multi-Verse Optimizer (MVO), Flow Direction Algorithm (FDA), and Sine Cosine Optimization (SCO) algorithms, inspired by different physical laws, were utilized under the selection hyper-heuristic structure. This is one of the first studies investigating a selection hyper-heuristic to solve a multi-objective cyber-physical control problem, an important contribution to the literature. The goal is to detect the best FOPID parameters of a cardiac pacemaker minimizing the rising time, settling time, and overshoot values in the system response, simultaneously. The selection hyper-heuristic maintains a score forming the probability of a low-level optimization algorithm being chosen and applied. The softmax function is used as the algorithm selection method. The performance comparison of the selection hyper-heuristic to the previously proposed algorithm indicates its potential, providing better settling time, rising time, and overshoot values from the FOPID controller. The statistical tests revealed that the ASO algorithm had a high frequency of usage, and the controller parameters yielding the best control performance were obtained when this algorithm was in operation. This further confirms the effectiveness of the proposed framework.