Design and development of an upper extremity motion capture system for a rehabilitation robot

2013 IEEE 13th International Conference on Rehabilitation Robotics (ICORR)

et al. 2013

In recent years, robot-assisted rehabilitation has gained momentum as a viable means for improving outcomes for therapeutic interventions. Such therapy experiences allow controlled and repeatable trials and quantitative evaluation of mobility metrics. Typically though these robotic devices have been focused on rehabilitation within a clinical setting. In these traditional robot-assisted rehabilitation studies, participants are required to perform goal-directed movements with the robot during a therapy session. This requires physical contact between the participant and the robot to enable precise control of the task, as well as a means to collect relevant performance data. On the other hand, non-contact means of robot interaction can provide a safe methodology for extracting the control data needed for in-home rehabilitation. As such, in this paper we discuss a contact and non-contact based method for upper-arm rehabilitation exercises that enables quantification of upper-arm movements. We evaluate our methodology on upper-arm abduction/adduction movements and discuss the advantages and limitations of each approach as applied to an in-home rehabilitation scenario.

Section: B Hidden Markov Modelmentioning

confidence: 82%

Non-contact versus contact-based sensing methodologies for in-home upper arm robotic rehabilitation

Howard

Brooks

2013 IEEE 13th International Conference on Rehabilitation Robotics (ICORR)

et al. 2013

“…In this project, the Upper Extremity Motion Capture System was built to perform this operation. This motion capture system was designed, developed, and constructed in the Biomechatronic Learning Laboratory [15]. The system consists of a platform, which ensures the stability of the device, and a brace for the user's arm.…”

Section: B Data Acquisition Systemmentioning

confidence: 99%

Using Hidden Markov Models to track upper extremity arm motions for surface electromyographic based robot teleoperation

Gebregiorgis

5th ISSNIP-IEEE Biosignals and Biorobotics Conference (2014): Biosignals and Robotics for Better and Safer Living (BRC)

2014

A Hidden Markov Model (HMM) is used to predict and characterize a stochastic process that is not easily identifiable; i.e., it is hidden from the observer. This process can only be identified through an additional set of stochastic events that is not only observable, but is also responsible for producing the original hidden stochastic process mentioned above. The goal of this project is to use a HMM to track upperextremity arm motions performed in the sagittal plane (representing the hidden states) by means of the surface electromyographic (sEMG) activity associated with these arm motions (representing the observed states). After which, we intend to use the characterized sEMG signals to teleoperate a robotic manipulator. We wish to create a rehabilitative robotic platform for people who have suffered from progressive muscular degenerative disorders and neurological deficits. This platform will take advantage of any residual physiological information that is still available (non-invasively) within these individuals. The ultimate goal is to create more intelligent orthotics and wearable robotic systems for people having these types of disabilities. It is hoped that this kind of device could assist a disabled user in performing daily living tasks that require reaching for an object. The HMM algorithm presented here is implemented and tested offline in Matlab with five healthy participants. It was successful in tracking two degrees of freedom on the human arm (representing the elbow and shoulder joints) with less than 15° of error.

“…All of the data presented in the work for the TDNN based myoelectric control scheme was collected using the Upper Extremity Motion Capture System shown in Figs 4 through 6 [24]. The developed system captures EMG and joint angle data from the elbow and shoulder.…”

Section: Myoelectric Control Scheme Based On a Time Delayed Neural Nementioning

confidence: 99%

Myoelectric Control Techniques for a Rehabilitation Robot

Smith

Applied Bionics and Biomechanics

2011

This work examines two different types of myoelectric control schemes for the purpose of rehabilitation robot applications. The first is a commonly used technique based on a Gaussian classifier. It is implemented in real time for healthy subjects in addition to a subject with Central Cord Syndrome (CCS). The myoelectric control scheme is used to control three degrees of freedom (DOF) on a robot manipulator which corresponded to the robot's elbow joint, wrist joint, and gripper. The classes of motion controlled include elbow flexion and extension, wrist pronation and supination, hand grasping and releasing, and rest. Healthy subjects were able to achieve 90% accuracy. Single DOF controllers were first tested on the subject with CCS and he achieved 100%, 96%, and 85% accuracy for the elbow, gripper, and wrist controllers respectively. Secondly, he was able to control the three DOF controller at 68% accuracy. The potential applications for this scheme are rehabilitation and teleoperation. To overcome limitations in the pattern recognition based scheme, a second myoelectric control scheme is also presented which is trained using electromyographic (EMG) data derived from natural reaching motions in the sagittal plane. This second scheme is based on a time delayed neural network (TDNN) which has the ability to control multiple DOF at once. The controller tracked a subject's elbow and shoulder joints in the sagittal plane. Results showed an average error of 19° for the two joints. This myoelectric control scheme has the potential of being used in the development of exoskeleton and orthotic rehabilitation applications.