Neurologic injuries, such as stroke, spinal cord and weaknesses of skeletal muscles with elderly people, may considerably limit the ability this population to achieve main daily living activities. Recently, there has been an increasing interest in development wearable devices, so-called exoskeletons, assist well patients limb pathologies, for movement assistance rehabilitation. In paper, we review discuss state art lower exoskeletons that are mainly used physical An overview commonly...

Recognition of the wearer's motion intention plays an important role in study power-assist robots. In this paper, intention-guided control strategy is proposed and applied to upper-limb exoskeleton. Meanwhile, a human-robot interface comprised force-sensing resistors (FSRs) designed estimate upper limb real time. Moreover, new concept called "intentional reaching direction (IRD)" quantitatively describe intention. Both state model observation IRD are obtained by studying behavior modes...

Using supervised machine learning approaches to recognize human activities from on-body wearable accelerometers generally requires a large amount of labelled data. When ground truth information is not available, too expensive, time consuming or difficult collect, one has rely on unsupervised approaches. This paper presents new approach for activity recognition raw acceleration data measured using inertial sensors. The proposed method based upon joint segmentation multidimensional series...

This article presents a machine learning methodology for diagnosing Parkinson's disease (PD) based on the use of vertical Ground Reaction Forces (vGRFs) data collected from gait cycle. A classification engine assigns subjects to healthy or Parkinsonian classes. The diagnosis process involves four steps: pre-processing, feature extraction and selection, performance evaluation. selected features are used as inputs each classifier. Feature selection is achieved through wrapper approach...

Abstract Recent technological advances made necessary the use of robots in various types applications. Unlike traditional robot-like scenarios dedicated to industrial applications with repetitive tasks, current focus attention is on that require close human interactions. One main fields such concerns assisting and rehabilitating dependent/elderly persons. In this study, state-of-the-art research lower-limbs assistance presented. This includes a review covering mainly lower-limb actuated...

Gait modes, such as level walking, stair ascent/descent, and ramp show different lower-limb kinematic kinetic characteristics. Therefore, an accurate detection of these modes is critical for a wearable robot to provide appropriate power assistance. In this paper, fast gait-mode-detection method based on body sensor system proposed. A fuzzy logic algorithm used estimate the likelihoods gait in real time. Since proposed mode makes it possible select models each mode, assistive torques required...

This paper presents a bounded control strategy of knee-joint exoskeleton that ensures knee flexion/extension movements during the assistance and rehabilitation processes. The is easy to implement, since it does not require any prior knowledge human effort. proposed law based on nested saturations with main advantage explicitly avoiding actuator's saturation, guaranteeing enhancing wearer's security. Lyapunov-based analysis stated prove: 1) asymptotic stability shank-foot-exoskeleton in...

As an important movement of the daily living activities, sit-to-stand (STS) is usually a difficult task facing elderly and dependent people. To provide appropriate power assistance for movement, novel intention-based Active Impedance Control (AIC) strategy applied on lower limb exoskeleton proposed in this paper. The AIC able to adapt mechanical impedance human-exoskeleton system towards desired one using exoskeleton's assistance. In structure, human joint torque observer designed estimate...

Reducing human lower limb effort during walking is still a challenge for wearable exoskeletons. From the wearer's viewpoint, joint impedance increases while wearing exoskeleton due to fact that exoskeleton's mass, friction, and gravity are added those of limbs. Active required from compensate not only mechanical but also limb. This paper presents nonlinear reduction control (IRC) approach applied knee flexion/extension movements. The IRC ensures adaptation human-exoskeleton system toward...

In this article, a proxy-based force control method is proposed for three important human-robot interaction modes: zero-impedance mode, assistive and large mode. A two-mass dynamic-model-based nonlinear disturbance observer used to meet the output accurate tracking requirements with respect disturbances from wearer environment. Additionally, significant compliance can be achieved guarantee wearer's safety when torque large. The evaluated via experiments by comparison conventional...

As an important movement of the daily living activities, sit-to-stand (STS) is usually a difficult task facing elderly and dependent people. In this article, novel impedance modulation strategy lower-limb exoskeleton proposed to provide appropriate power balance assistance during STS movements while preserving wearer's control priority. The ensures adaptation mechanical human–exoskeleton system toward desired one requiring less effect reinforcing ability movements. A human joint torque...

Gesture recognition becomes a thriving research area in modern human motion systems. The intensification of demands on efficient interactive human-machine-interface systems, commercial objectives, and many other factors contributes to fuel this revival dynamics. Understanding gestures essential for prevention health monitoring applications. In particular, analyzing hand is paramount importance personalized healthcare-related applications help practitioners providing more qualitative...

Loss of mobility and/or balance resulting from neural trauma is a critical public health issue. Robotic exoskeletons hold great potential for rehabilitation and assisted movement. However, the synergy robot operation with human effort remains problem. In particular, optimal assist-as-needed (AAN) control unresolved given pathological variance among patients. We introduce model predictive (MPC) architecture lower limb that achieves on-the-fly transitions between modes assistance. The...

The paper concerns the control of a lower limb orthosis acting on knee joint level. Therefore, model shank-orthosis system is given considering human effort as an external torque system. A reference adaptive law developed and applied to in order make (shank-orthosis) track desired trajectory predefined by rehabilitation doctor. main advantage this on-line parameters regulation allowing ensure best performance Lyapunov-based analysis performed prove input-to-state stability with respect...