Yu Laboratry

千葉大学
  • JAPANESE

Prosthetic hand

Rehabilitation device

Development of a soft hand rehabilitation device for a wide variety of excersises

Problem declaration Interest in soft robotics for rehabilitation and power-assist has been increasing recently as an alternative for rigid systems.
However, so far, the developed soft glove devices do not enable support to all the various motions of the hand joints, including the abduction-adduction of the fingers' metacarpophalangeal joints.
Goal We aim to enable a wide variety of upper limb rehabilitation exercises by developing a cheap, portable, and intuitive home-use rehabilitation glove that combines soft fiber-reinforced actuators and functional electrical stimulation (FES).
Progress We have established fabrication methods for soft pneumatic fiber-reinforced actuators, and have been developing methods for making multi-pocket structures to enable higher degree of control over the human joints.
Currently we are assembling the actuators into a full glove, and developing its control and user interface. We are also collaborating with rehabilitation experts and a hemiparetic patient to improve the intuitiveness and ease use of the device.

BCI Optimization for 3D Continuous Hand Movement Reconstruction from EEG and EMG recordings

Purpose  The main objective of this research is to create the control for an upper-limb prosthetic device using non-invasive methods such as EEG and EMG. Right now the only reliable control system use invasive methods that require dangerous surgery. By using a complex predictors and biofeedback we aim to reach a control with similar precision to the invasive methods.
Goal  To be able to use this system to ease up daily living reaching and grasping activities using both the normal arm and a prosthetic arm.
Progress  Vision sensors shows that it is possible to estimate the 3D trajectory. However the current results have not been successfully tested against a normal hand and a prosthetic hand. Vision sensors also demonstrate advantages of their use against other methods such as bio-signals that do not detect and estimate objects and hands postures.
 

Estimation of Bimanual Coordinated Movement using biomedical signals

Problem declaration Shoulder amputee has a low number of remaining muscles, so it’s difficult to control Trans-humeral Prostheses with multiple degrees of freedom
Goal Estimate the Bimanual Coordinated Movement and Trans-humeral Prostheses is operated automatically.
Progress We are estimating the bimanual coordination using physical quantity.

Research on Vision Based 3D Trajectory Estimation for controlling bimanual coordination between a normal hand and a prosthetic hand

目的  The main objective of this research theme is to develop a non-invasive prosthetic hand 3D trajectory using 3D RGB-D Vision sensors. Such trajectory should be able to control the bimanual coordination between a normal hand and a prosthetic hand. We also intend to explore the optimal usage of vision sensor cameras in objects identification and obstacle avoidance in a prosthetic hand reaching and grasping space especially in a clustered environment.
最終応用目標  To be able to use this system to ease up daily living reaching and grasping activities using both the normal arm and a prosthetic arm.
現在までの結果  Vision sensors shows that it is possible to estimate the 3D trajectory. However the current results have not been successfully tested against a normal hand and a prosthetic hand. Vision sensors also demonstrate advantages of their use against other methods such as bio-signals that do not detect and estimate objects and hands postures.

Auditory based Sensory Function Substitution for Prosthetic Applications

目的  To develop a non-invasive sensory function substitution scheme for prosthetic applications using an auditory display. Also, to explore how the usage of this sensory substitution scheme is affecting the human-robot interaction from a point of view of performance, workload, attention, and emotions.
最終応用目標  To use the developed system as a training or support tool to dynamically couple an amputee’s sensory system with the prostheses, in order to improve its motor control for daily reaching and grasping activities.
現在までの結果  The results showed that an auditory feedback scheme could be effectively used to improve performance greatly during reaching and grasping motions and that a higher mental effort is needed when the subjects rely only on their vision, and that this effort seems to be reduced when auditory feedback is added to the human-machine interaction. Also, it shows the advantages and disadvantages of using psychophysiological techniques and self-assessment questionnaires to assess mental workload, attentional demands and emotions in prosthetic applications.

Study on reflex reaction of upper limbs aiming at application to artificial hands

Purpose Measurement and examination of reflection of upper limbs against multidirectional disturbance during reaching motion (myoelectric potential · acceleration · motion trajectory)
Goal  We investigate the reflection reaction useful for artificial limb control and aim to improve the safety and stability of the prosthetic hand by adding reflection function.
 
Progress It was suggested that the reflection has a characteristic that the intensity varies depending on the direction of the disturbance.

Development of multi-piece parallel link type shoulder artificial hand using pneumatic actuator

Purpose In order to support independence of upper limb amputees, we develop practical myoelectric shoulder artificial hands.
We aim to reduce flexibility by arranging pneumatic actuators in parallel for each link and to improve flexibility and wider range of motion than to make backbone of structure into rubber material.
Goal We analyze the stress and deformation of the backbone using CAE and verify whether it is effective to use rubber material to widen the movable range.
Progress Based on the analysis result of the backbone of the 1st link, since the area widened in the movable range when the backbone is made into the rubber material, in the future we will conduct analysis of the 2nd link and selection taking multiple materials and loading force into account.
⇒⇒

Development of biosignal selection and sensor harness for upper limb movement identification

Purpose Enable upper limb amputee to use motorized artificial hands regardless of the degree of cutting, to help improve quality of life (QOL).
Goal Select biomedical signals useful for discrimination of upper limb movements in shoulder and back.
In addition, we apply the selected bio signal to the control of the electric prosthesis, and fabricate a wearable sensor harness.
Progress The undulations of the forearm's body surface were measured when hands and wrists were moved.
At that time, a piezoelectric film (following figure) was attached to the forearm and used.
In the future, confirm the difference of measurement data between the movements and measure with shoulder and back.

Study on online learning method in myoelectric prosthesis

Purpose We aim to construct a discriminator to discriminate motion of upper arm / forearm from surface myoelectric potential by online learning.
Goal We aim to apply to the myoelectric arms by constructing a system that can update the classifier in real time according to the change in the myoelectric potential of the user.
Progress We succeeded in constructing a discriminator by off line learning using the myoelectric potential in the vicinity of the shoulder and discriminating motion in real time using the constructed discriminator.