NEXAR

Implementation of an AI Powered Autonomous Mobile Robot
with Vision-Based Navigation
Sricharan Srikrishna
22BEC0948
SENSE
Vellore Institute of Technology, Vellore
Tamil Nadu, India
sricharan.s2022a@vitstudent.ac.in
ABSTRACT
in this paper we present an AI-based autonomous robot
with the capabilities of real-time perception, action, and
reasoning. This paper uses computer vision OpenCV and
YOLO for object identification and tracking, ESP32
microcontrollers, and a decision intelligence layer that
makes use of large language model reasoning patterns.
The robot takes video input, tracks objects in real time,
spatial awareness is implemented by the relative position
of the object to the camera frame, and control commands
are generated to help the robot navigate. So basically
processing unit connects to the ESP32 controller through
HTTP based wireless communication. Based on the
results, the system can track and navigate objects under
various circumstances.