Adithya Rajendran

Building intelligent robotic systems that bridge perception and action.

About Me

I'm a Robotics & AI Engineer passionate about building intelligent systems that can perceive, reason, and interact with the physical world. Currently pursuing my Master's in Robotics at Northeastern University, I focus on the intersection of machine learning and robotic manipulation.

My research interests lie in Physical AI — developing robots that learn from vision-language models to perform complex manipulation tasks. I work extensively with frameworks like SmolVLA, ACT policies, and reinforcement learning to create adaptive robotic behaviors.

I believe the future of robotics is in systems that can understand human intent through natural language and visual cues, then translate that understanding into precise physical actions.

Research Focus

Physical AI & Embodied Intelligence
Vision-Language Models for Robotics
Robot Learning & Manipulation
Motion Planning & Control

Technical Skills

Physical AI

SmolVLA ACT Reinforcement Learning PyTorch TensorFlow OpenAI Gym HuggingFace W&B

Robotics Simulation

ROS 2 MoveIt Nav2 Gazebo PyBullet CARLA URDF

Perception & Navigation

SLAM Sensor Fusion Path Planning Depth Vision Computer Vision OpenCV

Languages & Tools

Python C++ Linux Git CMake MATLAB SolidWorks

Hardware

SO-101 Robot Intel NUC Raspberry Pi Depth Cameras Arduino

Experience

Research Assistant

Silicon Synapsis Lab, Northeastern University

January 2025 - Present Boston, MA

Engineered and validated control policies for a hybrid multi-modal robot (legged and aerial) in Gazebo, ensuring seamless mode transitions and collision-free navigation across >600 graph nodes
Accelerated robot path-planning by 40% by porting Probabilistic Roadmaps (PRM) and A* algorithms from MATLAB to optimized ROS 2 C++
Developed custom cost functions within the navigation stack to prioritize energy efficiency, reducing onboard CPU usage and power consumption

Gazebo ROS 2 C++ PRM A* Motion Planning

Research Intern

Indian Institute of Technology, Delhi · Prof. SK Saha's Lab

November 2023 - July 2024 · 9 months Delhi, India

Architected an LLM-driven agent on edge (Raspberry Pi) using OpenAI Whisper and ChatGPT, achieving 2x lower latency for real-time natural language command execution
Reduced collision incidents by 90% by implementing a LiDAR-based perception pipeline and tuning Nav2 parameters (Behavior Trees, Costmaps, AMCL)
Developed a wavefront-frontier exploration algorithm integrated with SLAM, achieving 95% automated map coverage without human teleoperation

LLM Whisper SLAM Nav2 LiDAR ROS

Research And Development Intern

SeiAnmai Technologies · IIT Delhi Campus

November 2023 - July 2024 · 9 months Delhi, India

Developed TOTOCE robot for autonomous navigation in crowded environments, showcased at Robocon 2024
Implemented LiDAR-based perception pipeline with Nav2 tuning (Behavior Trees, Costmaps, AMCL), reducing collision incidents by 90%
Built wavefront-frontier exploration algorithm integrated with SLAM, achieving 95% automated map coverage

TOTOCE SLAM Nav2 LiDAR ROS AMCL

Project Intern

RobotoAI Technologies

August 2022 - November 2022 · 4 months Coimbatore, India

Developed robot localization and navigation techniques for autonomous mobile robots
Implemented SLAM and AMCL algorithms for precise robot positioning and mapping
Integrated multiple sensors and built path planning systems for collision-free navigation

SLAM AMCL Navigation ROS

Student Intern - Center of Excellence

Titan Engineering & Automation Limited

June 2022 - July 2022 · 2 months Hosur, Tamil Nadu, India

Conducted process analysis and optimization in manufacturing environments
Implemented Industrial Internet of Things (IIoT) solutions using IOT2050 devices
Utilized FlexSim for manufacturing process simulation and workflow optimization

IIoT FlexSim IOT2050 Process Analysis

Industrial Trainee

AGIIT

May 2019 - June 2019 · 2 months Coimbatore, Tamil Nadu, India

Completed hands-on training in PLC programming and industrial automation systems
Gained practical experience with relay logic and control system fundamentals

PLC Relay Logic Automation

Featured Projects

In-Hand Dice Reorientation with RL

February 2026 - Present

PPO-based dexterous manipulation with Unitree Dex3-1 hand (7 DOF), finger gaiting reward, and 7-phase curriculum learning in MuJoCo/MJX. 100% eval SR on all 6 faces.

PPO MuJoCo MJX PyTorch Dexterous Manipulation