Abhinav Bhamidipati Portfolio

Abhinav Bhamidipati

I'm

Perception, Path Planning, Computer Vision, Machine Learning

About

My name is Abhinav Bhamidipati, 23-year-old pursuing Master's Degree in Robotics at the University of Maryland, College Park. I'm currently working at Inception Robotics contributing to develop autonomy stacks and enhancing robot capabilities through visual odometry and sensor fusion techniques.
My passion lies in the fields of Computer Vision and Machine Learning, and I have experience in areas such as Perception, Path Planning, SLAM, and Reinforcement Learning. My strong analytical thinking and problem-solving abilities enable me to navigate complex technical issues effectively. I've led teams in pioneering projects that have made a positive impact. I'm a lifelong learner who thrives in dynamic environments and embraces new challenges with enthusiasm.
Beyond engineering, I'm creative, indulging in sketching and painting. I'm driven by innovation, eager to collaborate, and excited to make meaningful contributions to the world of robotics and beyond.

Education

Master of Engineering, Robotics

Aug 2023 - May 2025

University of Maryland, College Park, MD

CGPA : 3.84/4

Bachelor of Engineering, Electronics and Telecommunication

Aug 2019 - May 2023

University of Mumbai, Atharva College of Engineering, Mumbai, India

CGPA : 8.97/10

Experience

Robotics Software Intern Inception Robotics, College Park

Sept 2024 - Present

Designed and deployed full‑stack autonomy pipelines for Autonomous UGVs using C++, Python, ROS2, and CUDA in Unity and Isaac Sim simulations and real‑world testing; integrated LiDAR and visual odometry modules with 3D ICP scan matching and ORB feature tracking, enhancing localization robustness through adaptive outlier filtering.
Built and scaled reinforcement learning–based terrain navigation models, importing environment from AWS and automating end‑to‑end CI/CD pipelines with Docker, and GitLab, achieving 90% navigation success across dynamic terrains.
Extended the autonomy stack with a DWA‑based local planner and an LLM‑powered natural language interface, leveraging SITL pipelines for multi‑run validation, enabling real‑time dynamic obstacle avoidance and voice‑commanded mission execution.
Developed a CLIP‑based vision‑model for final‑stage object damage assessment, capturing multi‑angle imagery of assigned objects & evaluating condition against predefined guidelines to detect surface deformation, cracks, & structural anomalies.

Robotics Software Intern Onki Robotics, New York City

Jun 2024 - Aug 2024

Optimized SLAM and localization pipelines for UGVs on Jetson platforms using ROS2, reducing pose update latency by 25% to enable real‑time tracking and navigation in dynamic, crowded environments.
Enhanced environmental perception, with sensor‑fusion of LiDAR and stereo camera data, generating denser occupancy maps and improving robustness to occlusions and cluttered scenes for consistent person‑following behavior.
Developed ML‑based human feature extraction and tracking models, enabling the UGV to detect, lock onto, and continuously follow a selected individual through crowds, maintaining robust identity tracking across partial occlusions and reappearances.

Research Assistant University of Maryland, College Park

Jan 2024 - May 2024

Integrated multi‑beam sonar and depth sensors in simulations for UUVs, enhancing decision‑making capabilities.
Improved environmental perception and navigation precision for UUVs through advanced sensor fusion techniques.
Developed control algorithms to optimize UUV path planning and maneuverability in complex underwater environments.

Research Assistant University of Toronto

Jun 2022 - Nov 2022

Performed in‑depth analysis of a large‑scale database with over 1 million patient records daily, uncovering critical trends.
Developed algorithms for data cleaning using Isolation Forest, and also created real‑time tracking dashboards.

Web Developer N. Thakkar & Associates

Aug 2021 - Sept 2021

Experion, New York, NY

Optimized the throughput of a web application by 10% by using a load balancer and optimizing the code.
Created a user interface design for a website using wireframes and mockups.