Fuzzy Adaptive Passive Control Strategy Design for Upper-Limb End-Effector Rehabilitation Robot

Abstract

Robot-assisted rehabilitation therapy has been proven to effectively improve upper-limb motor function in stroke patients. However, most current rehabilitation robotic controllers will provide too much assistance force and focus only on the patient's position tracking performance while ignoring the patient's interactive force situation, resulting in the inability to accurately assess the patient's true motor intention and difficulty stimulating the patient's initiative, thus negatively affecting the patient's rehabilitation outcome. Therefore, this paper proposes a fuzzy adaptive passive (FAP) control strategy based on subjects' task performance and impulse. To ensure the safety of subjects, a passive controller based on the potential field is designed to guide and assist patients in their movements, and the stability of the controller is demonstrated in a passive formalism. Then, using the subject's task performance and impulse as evaluation indicators, fuzzy logic rules were designed and used as an evaluation algorithm to quantitively assess the subject's motor ability and to adaptively modify the stiffness coefficient of the potential field and thus change the magnitude of the assistance force to stimulate the subject's initiative. Through experiments, this control strategy has been shown to not only improve the subject's initiative during the training process and ensure their safety during training but also enhance the subject's motor learning ability.

Keywords: assist-as-needed; end-effector rehabilitation robot; fuzzy logic; human–robot interaction; potential field.

Figure 1

The design of the EBULRR platform.

Figure 2

Diagram of the control system hardware.

Figure 3

Diagram of the coordinate system of each linkage of the robot arm.

Figure 4

The potential energy field is designed by using a circle as the desired trajectory; the red line is the desired trajectory. The higher the potential energy at the current point, the greater the normal force provided.

Figure 5

Fuzzy inference system structure diagram.

Figure 6

Membership functions for the input and output.

Figure 7

Input and output surface diagram of fuzzy logic.

Figure 8

The effect of the potential field stiffness coefficient  $K$ on the potential field. (a) $K=6$ (b) $K=12$. Higher potential energy provides greater assistance force.

Figure 9

The simplified control block diagram of FAP.

Figure 10

Comparison chart of security simulation tests between the proposed FAP controller and the conventional impedance controller.

Figure 11

The subject was participating in an experiment with a circle drawing task.

Figure 12

Partial task trajectory diagram of S1 in three modes.

Figure 13

S1’s task trajectory diagram in FAP mode: (a) task1, (b) task7, and (c) task19.

Figure 14

Evolution of $Score,I_a$, and $K$ during S1’s task in FAP mode.

Figure 15

(a) Comparison of robot end force sensor values for eight subjects in different training modes. (b) Group average of force sensor values for eight subjects in different modes.

Figure 16

(a) Comparison chart of actual trajectory of S1 in training stage. (b) Comparison chart of actual trajectory of S1 in the evaluation stage.

Figure 17

The evolution of the $Score$ of S1 in different stages. The top half of the image shows the comparison of $Score$ during the training stage and the bottom half of the image shows the comparison of scores during the evaluation stage.

Figure 18

Graph of the mean scores of the eight subjects at different stages. The top half of the image shows the comparison of mean scores during the training stage and the bottom half of the image shows the comparison of mean scores during the evaluation stage.

Figure 19

(a) Comparison of K values of S1 in different modes. (b) Comparison of K values of eight subjects in different modes.