Personalized Elder Care Robot

Oro Robot

Vision Classifier Test & Validation

For my senior thesis work at Rice University, I worked with a team of computer engineering and computer science students to build a personalized elder care robot. In our work, sponsored by Intel, we designed, built, and deployed a mobile base robot. Our work aimed to create a system that could assist geriatric patients in living in their homes longer by automating nursing tasks. We wanted the robot to stay clear of people, require little to no modification of its environment, and passively monitor vitals and localize the patient.

Our robot, named Elderado, was built on the Robotic Operating System (ROS) and featured a LiDAR scanner, depth camera, touchscreen interface, actuated pan/tilt head, long-lasting battery, and charging dock. These sensors allowed us to use simultaneous localization and mapping (SLAM) algorithms to enable the robot to navigate. We used SE(2) mobile base planning algorithms aimed at differential drive designs with our environment model to move autonomously. Furthermore, as our robot had depth camera information, we demonstrated state-of-the-art segmentation and analysis models developed by the University of Toronto to classify if a patient was prone on the ground and in distress.

In addition to the robot, we built a low-power bracelet worn by the patient that – in tandem with low-energy Bluetooth beacons placed at various locations – could track the approximate location of patients and report vitals to the robot. Armed with this information, we built a state machine model of the patient’s typical location at different points in the day. When the patient deviated from the model’s expectation, the robot would automatically navigate to their location and assess if medical intervention was necessary.

We wanted the Elderado robot to integrate easily with the existing telemedicine infrastructure. As such, we consulted physicians at the Texas Medical Center and built a proof of concept allowing a remote user to teleoperate the robot with a two-way call.

Unfortunately, due to the COVID-19 pandemic, we could not present our work at the end-of-year Engineering Design Showcase. But all participants, myself included, received an honors grade for our work.