Skip to content
View N-Raghav's full-sized avatar
🤖
Aspiring Robotics Engineer
🤖
Aspiring Robotics Engineer
32 followers · 124 following

Achievements

Achievement: Pair ExtraordinaireAchievement: Pull Shark

Achievements

Achievement: Pair ExtraordinaireAchievement: Pull Shark

Highlights

  • Pro

Block or report N-Raghav

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
N-Raghav/README.md

Hi, I'm Raghav Nallaperumal

MS Robotics Engineering @ WPI | Perception • Deep Learning • Sensor Fusion

📧 raghavnallaperumal753@gmail.com
🌐 Portfolio | LinkedIn

I build intelligent robotic systems that perceive, learn, and act in the real world. My work spans vision-based perception (MonoSense, VIO, NeRF), learning-based manipulation (deep RL, imitation learning), and real hardware deployment (Franka Panda, UR5e, Crazyflie).

Recent Projects

🚗 MonoSense – Tesla-inspired autonomous driving perception pipeline (YOLOv6, depth estimation, 3D pose, ego-motion)
🛸 Deep VIO – 6 visual-inertial odometry approaches for UAVs with transformer-based sensor fusion
🤖 Imitation Learning – BC-Transformer vs Diffusion Policy for UR5e robotic stacking
🎮 Deep RL for Picking – A3C, Actor-Critic from scratch for robotic grasping
📐 SfM + NeRF – 3D reconstruction from scratch (27.3 dB PSNR)
🚁 Quadrotor Control – Sim-to-real transfer with 83% RMSE reduction

Tech Stack

Languages: Python, C++, MATLAB
Robotics: ROS 2 (Humble), MoveIt 2, Nav2, Gazebo, PyBullet, robosuite
Machine Learning: PyTorch, Deep RL (A3C, Actor-Critic), Imitation Learning, Diffusion Policy, NeRF
Perception: OpenCV, YOLOv6, VIO, SLAM, Structure from Motion, Bundle Adjustment, Sensor Fusion
Controls: LQR, PID, MPC, System Identification, Trajectory Optimization
Hardware: Franka Emika Panda, UR5e, Intel RealSense D435, Crazyflie 2.0, Beckhoff TwinCAT PLC
Tools: Docker, Git, SLURM/HPC, CUDA, Blender

📊 Seeking Full-Time Opportunities (Aug 2027) in perception engineering, robotics software, or ML for robotics.

Pinned Loading

  1. Mono-Sense Mono-Sense Public

    Tesla-inspired autonomous driving visualization using monocular camera perception — lane detection, 3D object placement, pedestrian pose estimation, and depth-based scene reconstruction rendered in…

    Jupyter Notebook

  2. Reinforcement-Learning-for-Picking-Task Reinforcement-Learning-for-Picking-Task Public

    Deep reinforcement learning algorithms(A2C and Reinforce) for LunarLanderv2 and A3C for Kuka Pick and Place Manipulation Task

    Python

  3. Hybrid-Push-Dynamics-Learning-Methods Hybrid-Push-Dynamics-Learning-Methods Public

    A comparative implementation and analysis of physics-based, neural network, and hybrid models for predicting object motion in robotic pushing tasks.

    Python

  4. SfM-and-NeRF SfM-and-NeRF Public

    Forked from prithvi-raj-b/CV-SfM-and-NeRF

    3-D reconstruction using images by implementing Structure from Motion and Neural Radiance Fields

    Python

  5. Imitation-Learning-Agent Imitation-Learning-Agent Public

    Imitation learning with BC-Transformer and Diffusion Policy for robotic stacking using the UR5e arm

    Shell

  6. Deep-Visual-Inertial-Odometry Deep-Visual-Inertial-Odometry Public

    Deep visual-inertial odometry stack for UAVs: vision-only, AirIMU+AirIO, PRGFlow baselines, and EKF/cross-attention fusion on Blender-rendered trajectories.

    Python