A Fuzzy Neural Network Controller Using Compromise Features for Timeliness Problem

Abstract

W HEN the control rules of traditional fuzzy controller are determined, it comes to be time-consuming and laborious to adjust for different usage conditions. Therefore, the timeliness cannot be guaranteed to solve the timeliness problem, and a fuzzy controller with modifiable factors is designed. While the entire control table is affected by the modifiable factors selection table, all previous control parameters need to be reset. In light of above problems, this paper firstly proposes new fuzzy controller design methods, which retain strengths of traditional controller and controller with modifiable factors. It effectively overcomes the shortcomings of the two controllers mentioned above, and only need to adjust the compromise factor for different working conditions of proposed controller, which is more convenient and efficient. Secondly, the proposed fuzzy controller also adopts a four-layer neural network to optimize the control rules of compromise to improve control precision and system robustness. Finally, the excellent characteristics of proposed controller are verified through simulation research, and the simulation result proves the proposed fuzzy controller has the advantages of higher control precision and smaller transition.

Simulation-Based Study of Control Strategies for Beating of Human Cardiomyocyte Cultures

“…S1 (a) and (c). As this simpler controller requires a smaller number of tuning parameters, it could significantly shorten the time-consuming determination of the control rules [8]. With this we examined, how well the simpler controller performs compared to one with a more complex logic in different simulation conditions.…”

“…Furthermore, there is typically a high variation in biological systems; for example, the spontaneous beating rate of hiPSC-CMs has been reported to range from 21 to 52 beats per minute [6], resulting in significant challenges to the traditional controllers. Therefore, researchers have studied fuzzy logic to control biological systems, as this approach does not require a precise mathematical model of the system [7], [8], [9], [10], [11]. Furthermore, fuzzy control allows for an easy way to eliminate the control response to the small transient-level changes by implementing a dead zone on the controller [12, p. 458].…”

A Soft Actor-Critic Approach for a Blind Walking Hexapod Robot with Obstacle Avoidance

“…Motion control of hexapod robots is a combination of various sensor information for joint control and coordination the between legs. The main current control methods are bio-inspired control [18], engineering-based control [19], machine learning-based control [20,21], and combinations of two or more. Bio-inspired control is typified by CPG control, which is derived from bionics inspired by insects and allows hexapod robots to execute gaits without any loss of equilibrium or increased delay [22,23].…”