Towards human-knee orthosis interaction based on adaptive impedance control through stiffness adjustment

Abstract

Rehabilitation interventions involving powered, wearable lower limb orthoses that can provide high-challenging locomotor tasks for repetitive training sessions, mainly when assist-as-needed strategies, such as adaptive impedance control, are designed. In this study, the adaptive behavior was ensured by software control of the robotic stiffness involved in the human-knee orthosis interaction in function of the gait cycle and speed. To estimate the stiffness, we analyzed the interaction torque-angle characteristics with experimental data. The speed-stiffness dependency was more evident when high stiffness values are demanded by the user's effort. Experimental evidence from five healthy subjects highlight that the adaptive control strategy provides a more comfortable, natural motion, and kinematic freedom as compared to the trajectory tracking control, allowing the user to contribute to the gait training. Future insights cover the implementation of gravitational compensation and real-time estimation and control of all inner dynamic properties of the impedance control law.

Knee Impedance Modulation to Control an Active Orthosis Using Insole Sensors

“…In our work, walking with ALLOR and the walker at 0.2 m/s implies an interaction torque of ±5 Nm, as shown in Figure 12 . In this sense, the new method for knee impedance modulation proposed here presents less knee torque than the torque presented in [ 19 ], with both position control and knee impedance modulation. Hence, the method based on FSR sensors for gait phase segmentation may be used to modulate knee impedance without demanding additional knee torque from the user during walking.…”

“…In [ 19 ], a gait analysis with an active knee orthosis without a walker was conducted, in which the torque with a position control at 0.28 m/s was approximately ±5 Nm. In addition, this study reported that the torque with an adaptive impedance control at 0.28 m/s and 0.44 m/s may have values of ±10 Nm.…”

“…Despite this, few studies have explored suitable and reliable methods to execute this modulation in gait applications, which are necessary in rehabilitation robots to guarantee a dynamic performance [ 18 ]. The literature provides information about impedance modulation for assist-as-needed control strategies based on interaction torque estimation methods and trajectory references [ 8 , 12 , 18 , 19 ]. In this case, a common limitation is the discontinuous model, just like to turn on or off the robotic assistance, rather than offering a seamless impedance tuning process [ 18 ].…”

Human-Robot Interaction in Rehabilitation and Assistance: a Review

“…To tackle this problem, Alquadi et al and Li et al proposed adaptive methods by incorporating biological signals and using neural networks to compensate for the unknown nonlinear robot dynamics [84,85]. Figueiredo et al used an online stiffness controller to ensure the adaptive behavior of the robot through analyzing the interaction torque-angle characteristics from experimental data [86]. In a research work by Li et al, a new adaptive impedance control was proposed in which an integral reinforcement learning (IRL) method was used to solve a linear quadratic regulation and minimize the position tracking errors.…”

Wearable Inertial Sensor System towards Daily Human Kinematic Gait Analysis: Benchmarking Analysis to MVN BIOMECH

“…Appendix A demonstrates the feasibility of the InertialLAB's integration into a powered exoskeleton, described in [16], for four-fold purposes. First, for real-time gait event detection, as described in [14], to adjust the human-orthosis dynamics for the adaptive impedance control strategy (proposed in [37]). The second purpose includes the real-time estimation of the hip and the knee angles for the powered knee exoskeleton and powered ankle exoskeleton, respectively, for the weight compensation of each exoskeleton through the gravitational compensation strategy.…”