Arvind Car

AI & Robotics Engineer

ABOUT ME

Hi! I am a Master's student at Carnegie Mellon University, specializing in AI and Robotics, with a focus on Deep Learning for Robotics, Motion Planning, and Controls. I work as a Graduate Student Researcher at the Mechanical and AI Lab at CMU, where my research focuses on deep learning methods for robotic manipulation. My academic journey has been shaped by an unwavering passion for physics, which has evolved into a keen interest in modeling complex phenomena.

My interests are driven by a broader goal – to predict, control, and make well-informed decisions across diverse domains. Over the course of exploring system modeling, controls, and decision-making, I have equipped myself with powerful tools such as deep learning, control theory, and robotics. As I continue to immerse myself in these dynamic fields, I am eager to leverage my academic foundation to contribute to the evolving landscape of AI and Robotics.

PROJECTS

Quadrotor Landing on an Unstable Mobile Platform

Designing and Implementing Model Predictive Control to make a quadrotor land on an unstable moving platform

ML-EECBS - Estimating better MAPF heuristics

Using ML methods to improve Explicit Estimation Conflict Based Search, a S.O.T.A. Multi-Agent Path Finding Algorithm

IROS 2024

SculptDiff: Diffusion policy for robotic sculpting

Developed a goal-conditioned diffusion-policy based imitation learning framework for deformable object manipulation

Deep RL for Aerobraking
Maneuvers

Trained a Proximal Policy Optimization Agent to perform aerobraking maneuvers faster and with reduced fuel consumption

Multi-Agent Pathfinding Algorithms

Implemented S.O.T.A. Multi-Agent Path Finding Algorithms (PP, CBS, PBS, EECBS) and evaluated their performances

Sq_Flatten_the_dough_and_stab_it-ezgif.com-optimize.gif

Planning with LLMs for Robotic Manipulation

Designed an Agentic LLM Framework to identify object locations within a scene, generate a high-level plan, and execute this plan through a series of low-level actions

VIEW ALL