Falling accidents are costly due to their prevalence in the workplace. Slipping has been known be main cause of falling. Understanding motor response used regain balance after slipping is crucial developing intervention strategies for effective recovery. Interestingly, studies on spinalized animals and subjected electrical microstimulation have provided major evidence that Central Nervous System (CNS) uses primitives, such as muscle synergies, control tasks. Muscle synergies thought a...

Exploiting hand gestures for non-verbal communication has extraordinary potential in HCI. A data glove is an apparatus widely used to recognize gestures. To improve the functionality of glove, a highly stretchable and reliable signal-to-noise ratio sensor indispensable. do this, study focused on development soft silicone microchannel sensors using Eutectic Gallium-Indium (EGaIn) liquid metal alloy gesture recognition system via proposed sensor. The EGaIn-silicone was uniquely designed...

Neurological disorders are the leading causes of poor balance. Previous studies have shown that biofeedback can compensate for weak or missing sensory information in people with deficits. These inputs be easily recognized and converted into proper by central nervous system (CNS), which integrates appropriate sensorimotor stabilizes human posture. In this study, we proposed a form cutaneous feedback stretches fingertip pad rotational contactor, so-called skin stretch. Skin stretch at simply...

The present study aimed to compare the effects of a deep learning-based digital application with physical therapy (DPT) and those conventional (CPT) on back pain intensity, limited functional ability, lower extremity weakness, radicular symptoms, range motion (ROM), movement, quality life, cost-effectiveness, postintervention questionnaires for perceived transmission risk COVID-19 satisfaction results in 100 participants low (LBP).One hundred LBP were randomized into either DPT or CPT...

Facial expressions are a nonverbal way to express our intent or feelings others. Human-like android robots should also be able make facial for people natural behavior. We generate the of robot, but there is no rule expressing it naturally. So, we have created by controlling motors one one. This very repetitive and inefficient procedure, resulting will vary depending on developer's skill. In this paper, propose method based action unit defined Ekman. used OpenFace recognize face robot...

This paper proposes a subplanner algorithm, which will be referred to as attractive force rotation and restoration, compensate the local minima problem under artificial potential based planner.The key role of proposed is build navigating path for escaping from by rotating restoring force.A mobile manipulator was adopted our robotic application substantiate capability various simulation scenarios in finds target object without collision against stationary/moving obstacles.The results...

Cutaneous sensory feedback can be used to provide additional cues a person performing motor task where vision is dominant signal. A haptic joystick has been widely guide user by providing force feedback. However, the benefit of still debatable due performance dependency on factors such as user's skill-level, difficulty. Meanwhile, recent studies have shown feasibility improving skin-stretch Therefore, combination two aforementioned types deemed promising promote synergistic effects...

Because the internal structure, degree of freedom, skin control position and range android face are different, it is very difficult to generate facial expressions by applying existing expression generation methods. In addition, differ among robots because they designed subjectively. To address these problems, we developed a system that can automatically robot combining an android, recognizer capable classifying genetic algorithm. We have two types (older men young women) simulate human...

This paper presents the bio-mimetic design approach, dynamic model, and potential applications for a hybrid soft actuator. The proposed actuator consists of two main parts: cylinder-shaped rigid core silicone spikes wrapped around core’s surface. key idea approach is to mimic movement grass-spike at functional level by converting vibration force generated small electric motor with counterweight in into propulsion produced elastic restoration spikes. One advantage this that does not need be...

A catadioptric vision system combines a camera and mirror to achieve wide field of view imaging system. This type has many potential applications in mobile robotics. paper is concerned with the design robust image-based control scheme using mounted on robot. We exploit fact that decoupling property contributes robustness method. More precisely, from image point, we propose minimal decoupled set features measurable any Using set, classical method proved be presence point range errors....

Android robots for arts performances are required to perform human-like actions. Recently, the most popular method is reconstruct action of robot by video-recording choreographer's directed a performance director in studio with 3D motion capture equipment. In this paper, we propose create data based on video-recorded human demonstration. The proposed aims at following two objectives: 1) can be made through existing video clips demonstrated actor, and 2) partial corrections instantaneously...

This paper proposes a soft linear actuator imitating foxtail movement and introduces two applications. The proposed biomimetic mainly consists of three parts: core containing vibration motor with silicone spikes on its surface, tubular passage to guide the core, switching mechanism change direction when an excessive tensile force is applied. When generated, moves along toward one depended leaning spikes; namely, it can be utilized as manipulate object if wired onto it. By inserting spring at...

This paper presents an approach for modeling user's driving-characteristics in a steering task, and determining the parameters of virtual fixture to assist user-control on basis his/her task-performances. First, we briefly introduce our assistive human-robot interaction (HRI) interface as backgrounds related this research. The designed HRI provides assistance by actively constraining with fixture. Second, discuss way model task. In driving-characteristics, use techniques from inverse optimal...

In the field of human rehabilitation, user satisfaction is a vital criterion in assessing efficacy assistive devices. A wheelchair popularly researched device with more than 3.5 million users US [1]. Through subsequent innovations, beginnings have undergone tremendous transformations to result today's powered-wheelchair. This technological advancement has been bolstered studies that prove powered-wheelchair's positive impact on user's mentality and social cost savings per [2].

사람은 태스크의 속성을 파악한 뒤 해당 태스크를 해결하기 위한 피드백 정책을 생성하는 일련의 과정을 통해 주어진 해결한다. 이러한 메타 인지 개념을 강화학습에 접목한 것이 메타-강화학습으로, 환경이 학습이 진행되었던 환경과 달라졌을 때에도 에이전트가 적절한 대응 생성하도록 한다. 본 논문에서는 시각-운동 일종인 1-구멍 1-공 geduld-spiele(GS) 큐브를 사람처럼 푸는 에이전트를 구현하기 메타-강화학습 기반 방법론을 제안한다. 먼저, 다양한 곡률의 GS 큐브 태스크(주어진 판의 곡률)를 프레임워크로 정의하였다. 다음으로, probabilistic embeddings for actor-critic RL을 사용하여, 훈련하였다. 마지막으로, 학습된 에이전트의 성능을, 특정 곡률에 대해서만 강화학습 에이전트와 비교분석하였다. 실험 결과, 메타-강화학습으로 학습한 에이전트는 경험해 보지 못한 GS큐브 문제에 있어, 곡률만을 에이전트 대비 91%에 상응하는 성능을 보였다....

This paper presents the strategy for searching a hidden object in an unknown area using by multiple distributed autonomous robotic systems (DARS). To search target Markovian space, DARS should recognize th ε ir surrounding at where they are located and generate some rules to act upon themselves. First of all, obtain 6-distances from itself environment infrared sensor which hexagonally allocated around itself. Second, it calculates 6-areas with those distances then take action, i.e., turn...

This paper proposes a systemic approach to upper arm gym-workout classification according spatio-temporal features depicted by biopotential as well joint kinematics. The key idea of the proposed is impute biopotential-kinematic relationship merging kinematic data into multichannel electromyography signal and visualizing merged an image. Under this approach, can be imputed counting on functionality convolutional neural network: automatic feature extractor followed classifier. First, while...

When we develop wearable assistive devices, comfort and support are two main issues that need to be considered. In conventional design approaches, the degree of freedom wearer's joint movements tends oversimplified. Accordingly, motion becomes restrained bone/ligament injuries might occur in case an unexpected fall. To mitigate these issues, this paper proposes a novel link mechanism inspired by human spine structure as well functionalities. The key feature proposed spine-like is...