TY - GEN
T1 - JUST TELL ME
T2 - 9th International Conference on Automation, Robotics and Applications, ICARA 2023
AU - Coutras, Alexander
AU - Obidat, Omar
AU - Zhu, Michelle
AU - Wang, Weitian
N1 - Funding Information:
ACKNOWLEDGMENT This work is supported in part by the National Science Foundation under Grant CMMI-2138351 and in part by the National Science Foundation under Grant CNS-2117308.
Publisher Copyright:
© 2023 IEEE.
PY - 2023
Y1 - 2023
N2 - Robotics technology has been increasingly applied to healthcare contexts to enhance efficiency and safety in healthcare processes in recent years. People with mobility impairments and disabilities often require caretakers in their lives. Fortunately, robots can provide attention and assistance consistently for them instead of human caretakers. Motivated by this, we develop a robot-assisted e-health solution to empower the patients' daily lives and improve their wellbeing in this study. A transfer learning-based approach is proposed to train the robot to understand and identify patients' needs through a small dataset. Using the proposed approach, the robot is able to understand the patient's needs through speech recognition and recognize objects that the patient has requested. The proposed solution is experimentally implemented in real-world human-robot interactive healthcare contexts. Results and analysis indicate the success and accuracy of our approaches.
AB - Robotics technology has been increasingly applied to healthcare contexts to enhance efficiency and safety in healthcare processes in recent years. People with mobility impairments and disabilities often require caretakers in their lives. Fortunately, robots can provide attention and assistance consistently for them instead of human caretakers. Motivated by this, we develop a robot-assisted e-health solution to empower the patients' daily lives and improve their wellbeing in this study. A transfer learning-based approach is proposed to train the robot to understand and identify patients' needs through a small dataset. Using the proposed approach, the robot is able to understand the patient's needs through speech recognition and recognize objects that the patient has requested. The proposed solution is experimentally implemented in real-world human-robot interactive healthcare contexts. Results and analysis indicate the success and accuracy of our approaches.
KW - computer vision
KW - E-health
KW - healthcare informatics systems
KW - human-robot interaction
KW - mobility impairments
KW - Robotics
UR - http://www.scopus.com/inward/record.url?scp=85161352061&partnerID=8YFLogxK
U2 - 10.1109/ICARA56516.2023.10125947
DO - 10.1109/ICARA56516.2023.10125947
M3 - Conference contribution
AN - SCOPUS:85161352061
T3 - 2023 9th International Conference on Automation, Robotics and Applications, ICARA 2023
SP - 139
EP - 144
BT - 2023 9th International Conference on Automation, Robotics and Applications, ICARA 2023
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 10 February 2023 through 12 February 2023
ER -