Care Companion

The Home-Based Geriatric Care Robot is an innovative solution designed to support elderly individuals in living independently and aging gracefully in the comfort of their own homes. Equipped with advanced sensors, artificial intelligence, and compassionate care algorithms, this robot provides personalized assistance with daily activities, health monitoring, and social interaction.

Researched, explored and implemented multiple features for the Home Based Geriatric Care Robot, enabling indiviuals to age gracefully

Optimized LSTM + CNN based fall detection pipeline; reducing frame processing time from 1.5s to 0.1s; replacing OpenPifPaf pose estimation with TRT_Pose, using Tensorflow, deployed on Jetson Nano

Developed vision-based spillage detection system; predicting 5 class labels with 95% accuracy in real time; using YoloV4 network trained on custom data optimized using TensorRT, deployed on Jetson Nano

Collaborated with health and geriatric experts to develop a novel mobility index score; using features such as gait speed, step length, stride width, flexion angle etc.; able to accurately predict and model changes in ambulatory movement (Patent Pending)

Developed a web-app to generate synthetic data; enabling experts to model a hypothetical subject with declining mobility; using streamlit.

Machine Learning, Data Analisys, REST/RESTful APIs, AWS Glue, Flask, Python, Docker, Modular APIs, Azure, ETL, Pandas, Numpy, Git