Abstract:
The Doctor Assistant Robot project addresses the growing demands of healthcare by integrating
advanced technologies such as WiFi fingerprinting, LiDAR, speech recognition, and natural
language processing into a robotic system. Designed to operate in complex healthcare environments,
the robot leverages WiFi fingerprinting for precise indoor localization and dynamic mapping, while
LiDAR technology enhances obstacle detection and path planning through detailed 3D maps.
Advanced algorithms, including Random Forest for obstacle classification for efficient pathfinding,
ensure accurate navigation. Additionally, the robot converts spoken prescriptions into electronic
records using speech recognition and integrates with Electronic Health Records (EHR) to improve
patient safety and compliance. Equipped with robust hardware, including WiFi sensors, LiDAR, and
a processing unit, the robot streamlines workflows, reduces errors, and alleviates the workload on
healthcare professionals. This project represents a significant step forward in healthcare delivery,
offering cutting-edge, robot-assisted solutions to enhance patient care and operational efficiency.
To further enhance the robot's capabilities, the project incorporates LightGBM, a high-performance
machine learning framework, for training models on datasets to improve obstacle classification,
speech recognition accuracy, and predictive maintenance. By leveraging LightGBM, the robot can
learn from historical data, optimize navigation paths, and dynamically adapt to changing
environments. This integration of machine learning ensures the robot operates with greater efficiency,
accuracy, and reliability, making it a transformative tool in modern healthcare settings.
