Unidirectional Human-Robot-Human Physical Interaction for Gait Training

TL;DR

A novel rehabilitation framework designed for a therapist, wearing an inertial measurement unit (IMU) suit, to virtually interact with a lower-limb exoskeleton worn by a patient with motor impairments, which aims to harmonize the skills and knowledge of the therapist with the capabilities of the exoskeleton.

Abstract

This work presents a novel rehabilitation framework designed for a therapist, wearing an inertial measurement unit (IMU) suit, to virtually interact with a lower-limb exoskeleton worn by a patient with motor impairments. This framework aims to harmonize the skills and knowledge of the therapist with the capabilities of the exoskeleton. The therapist can guide the patient's movements by moving their own joints and making real-time adjustments to meet the patient's needs, while reducing the physical effort of the therapist. This eliminates the need for a predefined trajectory for the patient to follow, as in conventional robotic gait training. For the virtual interaction medium between the therapist and patient, we propose an impedance profile that is stiff at low frequencies and less stiff at high frequencies, that can be tailored to individual patient needs and different stages of rehabilitation. The desired interaction torque from this medium is commanded to a whole-exoskeleton closed-loop compensation controller. The proposed virtual interaction framework was evaluated with a pair of unimpaired individuals in different teacher-student gait training exercises. Results show the proposed interaction control effectively transmits haptic cues, informing future applications in rehabilitation scenarios.

Figures (7)

• Figure 1: Framework structure: A therapist, on the left, can virtually interact with the patient, on the right, through unidirectional impedance interaction control while performing a rehabilitation exercise. The therapist wears the IMU suit, while the patient wears the Exo-H3 exoskeleton.
• Figure 2: Block diagram of the Exo-H3 controller. On the top-left portion, the interaction control computes the desired torque ($\tau_{\text{des}}$) using the impedance control transfer function $Z(s)$. The reference trajectory is defined by the partner selector between pre-recorded trajectories and the IMU-suit angles. In case the interaction control is not selected, it is possible to switch the control modality to transparent control ($\tau_{\text{des}}=0$). On the bottom-right portion, the WECC is responsible for compensating for the exoskeleton's gravity depending on the detected gait state. Then, it compensates for friction and implements the torque controller to drive the robotic joint.
• Figure 3: Bode plot of different $Z(s)$ changing the $k_{\text{high\_f}}$ values ([$Z_1$, $Z_2$, $Z_3$, $Z_4$, $Z_5$, $Z_6$]=[150, 100, 50, 30, 15, 0] Nm/rad), while $k_{\text{low\_f}} = 150$ Nm/rad and $f_{\text{cut}}=1$ Hz are kept constant. $Z_1$ represents the constant stiffness case, while $Z_6$ represents the low-pass filter case.
• Figure 4: Transparency results of the WECC while walking at 0.5 km/h, 1.1 km/h, and 1.5 km/h. The left column shows the mean interaction torque normalized by the subject's body mass, while the right column shows the average joint angles. The shaded areas represent +/- one standard deviation.
• Figure 5: Mean impedance variation during the treadmill walking test interacting with a virtual partner at different speeds (from left to right, 0.5 km/h, 1.1 km/h, and 1.5 km/h). On the $y$-axis, the desired interaction torque (impedance controller output) is reported over the tracking error (impedance controller input) on the $x$-axis. For each speed, the $Z_{\text{constant}}$ and the $Z_{\text{designed}}$ are shown together. In dotted line, the average $K$ +/- standard deviation is reported.