This paper introduces a real-time Human–Robot Interaction (HRI) system that establishes a gesture-based interface between a patient and a healthcare robot. The system detects patient motion through image processing, extracting geometric features from the patient’s hand movements. The system also incorporates real-time environmental monitoring using an Android camera, facilitated by an attendant in situations where a patient’s condition is too severe for the self-controlling of the robot. In scenarios necessitating social distancing, such as the COVID-19 pandemic, the proposed system holds the potential to control a healthcare robot. This system, incorporating computer vision technology and an Android camera, empowers the robot to securely grasp objects like medication, hand sanitizer, monitoring equipment, or food through a gripper control mechanism. The gesture recognition system accurately identifies right-hand gestures 93.75% of the time and left-hand gestures 91.25% of the time.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

A Gesture-Based HRI System for Health Care

  • Vibha Gaur,
  • Pushkar Baranwal,
  • Ravneet Kaur

摘要

This paper introduces a real-time Human–Robot Interaction (HRI) system that establishes a gesture-based interface between a patient and a healthcare robot. The system detects patient motion through image processing, extracting geometric features from the patient’s hand movements. The system also incorporates real-time environmental monitoring using an Android camera, facilitated by an attendant in situations where a patient’s condition is too severe for the self-controlling of the robot. In scenarios necessitating social distancing, such as the COVID-19 pandemic, the proposed system holds the potential to control a healthcare robot. This system, incorporating computer vision technology and an Android camera, empowers the robot to securely grasp objects like medication, hand sanitizer, monitoring equipment, or food through a gripper control mechanism. The gesture recognition system accurately identifies right-hand gestures 93.75% of the time and left-hand gestures 91.25% of the time.