Speed controller-based fuzzy logic for a biosignal-feedbacked cycloergometer

Abstract

Nowadays, fuzzy-logic systems are implemented to control machinery or processes that previously required human manipulation. The main objective of this research is to propose a controller based on fuzzy-logic that uses bio-signals for decision making. The study presents the implementation of a fuzzy-speed controller for a therapeutic machine called cycloergometer. It is used in patients who require rehabilitation therapy to improve their mobility in the lower body or to increase their relaxation or flexibility. Basic controllers have been developed where the speed is decided through a user interface, and the therapist must constantly increase or decrease the speed according to the condition of the patient. In this paper, the speed of the therapy equipment is adjusted using the heart rate of the patient. In this way, a bio-signal is used to determine whether a person is tired or relaxed. Therefore, a mechanism is obtained that is not subject to the visual criteria of the therapist. A detailed review of the literature illustrates that one of the main limitations of electroencephalography and electromyography recordings is the low signal-to-noise ratio and the fact that the signals captured at the electrodes are a mixture of sources that cannot be observed directly with noninvasive methods. Therefore, it was decided to work with electrocardiogram-based signals for better robustness of the proposed system. The controller output is a voltage signal in PWM, which is determined by the membership and error functions. The behavior of the implemented controller is validated by different experimental tests based on the increase and decrease of the simulated and real heart rate of a patient. Finally, the results obtained and the possible areas of opportunity for the proposed design are discussed.

Keywords:

• Fuzzy-logic control
• cycloergometer
• heart rate
• bio-signals
• biomedical applications
• physical therapist
• physiotherapy

Acknowledgment

The authors wish to thank Santiago López Hernández, student from the Facultad de Ingeniería of the Universidad Autónoma de Querétaro for allowing us to use the cycloergometer.

Disclosure statement

No potential conflict of interest was reported by the authors.

Funding

This research received funding from PRODEP and CONACYT.

Figure 1. Beats per minute membership functions.

Figure 2. Proposed control loop for the project.

Figure 3. Error membership functions.

Figure 4. Voltage membership functions.

Figure 5. bpm membership.

Figure 6. Error membership.

Figure 7. Voltage inference.

Figure 8. Connection diagram.