Building the perception layer for autonomous robots.
I build systems that help robots understand where they are and what they see: from camera-LiDAR calibration pipelines to real-time visual SLAM and state estimation.
-19.49ms Offset
Camera–LiDAR Temporal Calibration lag estimated and compensated.
90% Alignment
Real-time ZED, LiDAR & IMU RTAB-Map tunnel mapping score.
31.42dB PSNR
3D Gaussian Splatting rendering accuracy trained on custom models.
21 FPS
Low-light enhancement deep learning inference speed on Jetson edge systems.
SYSTEM STATUS: OPT STEM ACTIVE
Bridging spatial AI prototype to real-world autonomy.
I’m Gautham Ramkumar, a Robotics Graduate Student at Northeastern University specializing in computer vision, sensor fusion, and spatial localization. I build the software layers that resolve sensor noise and time delays, enabling autonomous vehicles and AMRs to operate reliably in GPS-denied or adverse-lighting settings.
Currently concentrating in ECE during my Master’s, my work is driven by field deployments. I focus on developing robust state estimation structures (using GTSAM and Extended Kalman Filters) and deploying deep learning pipelines (like YOLO and U-Net) optimized on NVIDIA Jetson edge systems.
Master's concentration in Robot Sensing & Navigation, Autonomous Field Robotics, and 3D Perception.
Selected Projects
Perception pipelines, state estimation systems, and visual SLAM integrations grouped by core domains.
Technical Core
Software packages, developer tools, and sensor hardware configurations.
Experience & Education
Academic focus and internship history in autonomous engineering.
Get in Touch
Feel free to reach out for new-grad roles, collaborations, or discussions.
Let's build something worth navigating.
I am looking for new-grad roles in robotics: perception, localization, calibration, and state estimation. Available for roles in the US (F-1 STEM OPT eligible, no immediate sponsorship required).