Çetin Meriç
Department of Computer Engineering, Pamukkale University, Kinikli Campus, 20070 Denizli, Turkey.
Cogn Neurodyn. 2020 Feb;14(1):51-67. doi: 10.1007/s11571-019-09555-8. Epub 2019 Sep 22.
Uncontrolled seizures may lead to irreversible damages in the brain and various limitations in the patient's life. There exist experimental studies to stabilize the patient seizures. However, the experimental setups have many sensory devices to measure the dynamics of the brain cortex. These equipments prevent to produce small portable stabilizers for patients in everyday life. Recently, a comprehensive cortex model is introduced to apply model-based observers and controllers. However, this cortex model can be uncertain and have time-varying parameters. Therefore, in this paper, a robust Takagi-Sugeno (TS) controller and observer are designed to suppress the epileptic seizures without sensory measurements. The unavailable sensory measurements are provided by the designed nonlinear observer. The exponential convergence of the observer and controller is satisfied by the feedback parameter design using linear matrix inequalities. In addition, TS fuzzy observer-controller design has been compared with the conventional PID method in terms of control performance and design problem. The numerical computations show that the epileptic seizures are more effectively suppressed by the TS fuzzy observer-based controller under uncertain membrane potential dynamics.
癫痫发作不受控制可能会导致大脑不可逆转的损伤以及患者生活中的各种限制。存在一些稳定患者癫痫发作的实验研究。然而，实验装置有许多用于测量大脑皮层动态的传感设备。这些设备阻碍了为患者在日常生活中生产小型便携式稳定器。最近，引入了一个综合皮层模型来应用基于模型的观测器和控制器。然而，这个皮层模型可能是不确定的且具有时变参数。因此，在本文中，设计了一种鲁棒的高木-菅野（TS）控制器和观测器，以在无需传感测量的情况下抑制癫痫发作。所设计的非线性观测器提供了无法获取的传感测量值。通过使用线性矩阵不等式的反馈参数设计，满足了观测器和控制器的指数收敛。此外，在控制性能和设计问题方面，将TS模糊观测器-控制器设计与传统的PID方法进行了比较。数值计算表明，在不确定膜电位动态的情况下，基于TS模糊观测器的控制器能更有效地抑制癫痫发作。
