Hyun-Sub Park

A Neural Network-Based Gait Phase Classification Method Using Sensors Equipped on Lower Limb Exoskeleton Robots

An exact classification of different gait phases is essential to enable the control of exoskeleton robots and detect the intentions of users. We propose a gait phase classification method based on neural networks using sensor signals from lower limb exoskeleton robots. In such robots, foot sensors with force sensing registers are commonly used to classify gait phases. We describe classifiers that use the orientation of each lower limb segment and the angular velocities of the joints to output the current gait phase. Experiments to obtain the input signals and desired outputs for the learning and validation process are conducted, and two neural network methods (a multilayer perceptron and nonlinear autoregressive with external inputs (NARX)) are used to develop an optimal classifier. Offline and online evaluations using four criteria are used to compare the performance of the classifiers. The proposed NARX-based method exhibits sufficiently good performance to replace foot sensors as a means of classifying gait phases.

Three-layered hybrid architecture for emotional reactive system

In this paper, we present the design of an emotional reactive system based on three-layered hybrid architecture. The purpose of emotional reactive system is to make robot software system that has its autonomous emotion in interaction with human. The three-layered hybrid architecture is composed of Hardware abstraction layer, Reactive Layer and Deliberative Layer and developed for autonomous mobile robots. We applied this architecture to make android EveR-2psilas emotional reactive system. With this architecture, EveR-1 could make robot emotion generated by external stimulus and play its emotional behavior in robot task execution.

Brief biomechanical analysis on the walking of spinal cord injury patients with a lower limb exoskeleton robot

This paper presents a brief biomechanical analysis on the walking behavior of spinal cord injury (SCI) patients. It is known that SCI patients who have serious injuries to their spines cannot walk, and hence, several walking assistance lower limb exoskeleton robots have been proposed whose assistance abilities are shown to be well customized. However, these robots are not yet fully helpful to all SCI patients for several reasons. To overcome these problems, an exact analysis and evaluation of the restored walking function while the exoskeleton is worn is important. In this work, walking behavior of SCI patients wearing the rehabilitation of brain injuries (ROBIN) lower-limb walking assistant exoskeleton was analyzed in comparison to that of normal unassisted walking. The analysis method and results presented herein can be used by other researchers to improve their robots.

Gait Phases Classification using Joint angle and Ground Reaction Force: Application of Backpropagation Neural Networks

This paper proposes the gait phase classifier using backpropagation neural networks method which uses the angle of lower body`s joints and ground reaction force as input signals. The classification of a gait phase is useful to understand the gait characteristics of pathologic gait and to control the gait rehabilitation systems. The classifier categorizes a gait cycle as 7 phases which are commonly used to classify the sub-phases of the gait in the literature. We verify the efficiency of the proposed method through experiments.

Organization of Sensor System and User's Intent Detection Algorithm for Rehabilitation Robot

In this paper, we propose the organization of a sensor system and user’s intent detection algorithm for walking assist rehabilitation robots. The main purpose of walking assist rehabilitation robots is assisting SCI patients to walk in normal environment. To use walking assist rehabilitation robot in normal environment, it is needed to consider various factors about user’s safety and detection of user’s intent and so on. For these purposes, we have analyzed the use case of rehabilitation robots and organized the system of sensors for walking assist rehabilitation robots and finally, we have developed the algorithm which is used to detect user’s intent for those. We applied our proposal method in the rehabilitation robot, ROBIN, and verified their effectiveness by normal, not patient.

A hybrid control method of an exoskeleton robot for intention-driven walking rehabilitation of stroke patients

When the robotic exoskeleton is used for gait rehabilitation, the robot should recognize the intention of patients before moving. We can consider recognition of intentions for the robot as recognition of a time (when the patient wants to move) and a behavior (what the patient wants to do). In this paper, the intent-detection method consisting of a gait phase recognizer and a gait pattern generator using neural networks is introduced. The effectiveness of the proposed method is verified briefly using recorded gait patterns.

Development of Gait Assisting Rehabilitation Robot for SCI (Spinal Cord Injury) Patient

This paper deals with gait assisting rehabilitation robot which helps SCI patient walk again. We propose new concept of orthotic for robot considering motions of Hip and Knee Joints, and how to fit the robot to a user in terms of weight balance and comfortable standing. Then we describe our first engineering sample being designed based on the passive orthotic and show how to make the robot work for SCI patient in basic operation.This paper deals with gait assisting rehabilitation robot which helps SCI patient walk again. We propose new concept of orthotic for robot considering motions of Hip and Knee Joints, and how to fit the robot to a user in terms of weight balance and comfortable standing. Then we describe our first engineering sample being designed based on the passive orthotic and show how to make the robot work for SCI patient in basic operation.