Autonomous Perception for Bi-Manual Mobile Manipulator Nursing Robots
Abstract: My presentation focuses on autonomous perception for mobile manipulator nursing robots in indoor healthcare environments. It presents a system that combines three main capabilities: goal-directed navigation using visual SLAM, human action recognition for safe interaction, and marker-less object manipulation with human-in-the-loop control. The robot is designed to navigate hospital-like settings, detect and respond appropriately to human activities, and perform pick-and-place tasks for 18 common medical and food items, including shelf organization. I also discuss the system architecture, perception pipeline, and experimental evaluation. The results demonstrate improved navigation reliability, interaction safety, and manipulation performance, highlighting the feasibility of deploying assistive robots in real-world nursing scenarios.
Advisor: Professor Jane Li
Committee: Professor Constantinos Chamzas, Professor Fiona Yuan
Zoom Link: https://wpi.zoom.us/j/93426165210