Human muscle rigidity identification by human-robot approximation characteristics framework on internet of things platform

Research output: Contribution to journalArticlepeer-review

Abstract

In the health care system and Internet of Things (IoT) platform, medical care robotics is becoming one of the quickest expanding areas of robot technology. The integration of robotics and human knowledge identifies human muscle rigidity from the healthcare data obtained from the wearable sensor. In an IoT platform, Electromyography is a method used for evaluating and tracking the electrical activity of muscles. The transferring of human muscle rigidity to a robot facilitates the robot to obtain resistive management initiatives in a useful and effective way while carrying out physical interaction activities in unstructured surroundings. The major challenges to overcome the unpredictability during physical interaction allow a robot to realize the individual behaviour with adaptability and versatility of muscles. Therefore, in this article, Human-Robot Approximation Characteristics Framework (HRACF) has been proposed for developing physiological communication between humans and robots. HRACF permits robots to understand differential resistive abilities of muscles from human presentations. The pulses collected from Electromyography are used to retrieve human arm muscle rigidity during activity presentation. The characteristics of motion and rigidity are concurrently modelled using an estimation and approximation model with a logistic regression obtained by IoT devices. The analysed human arm muscle rigidity is then connected to the robot impedance regulator. HR model uses an optimized resistive approximator to measure the creative variables of the robot and continue driving to monitor the quoted pathways at the time of interaction. The relationship between motion data and rigidity data is systematically coded in the HR model. HRACF makes it possible to detect uncertainties through space and time that facilitates the robot to meet rigidity specification to 98[Nm/Rad] and error rate to 0.15% during physical interaction.

Original languageEnglish
JournalExpert Systems
DOIs
Publication statusAccepted/In press - 2021

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Theoretical Computer Science
  • Computational Theory and Mathematics
  • Artificial Intelligence

Fingerprint

Dive into the research topics of 'Human muscle rigidity identification by human-robot approximation characteristics framework on internet of things platform'. Together they form a unique fingerprint.

Cite this