Jay Patrikar

Carnegie Mellon University, Indian Institute of Technology Kanpur

Papers

Total Citations

H-Index

About

Jay Patrikar is a leading researcher in aerial robotics and autonomous navigation, whose work bridges simulation, perception, and socially-aware motion planning. His key contributions span three critical areas: high-fidelity simulation for multi-rotor systems, deep learning-based robot relocalization, and long-horizon navigation in crowded environments. His most impactful work, the "Pegasus Simulator" (2024, 29 citations), provides an Isaac Sim-based framework for testing novel control and motion planning algorithms across diverse conditions—a vital tool for the robotics community. Earlier, his research on "Convolutional Neural Network Based Sensors for Mobile Robot Relocalization" (2018, 13 citations) pioneered lightweight CNN architectures for camera pose estimation, enabling efficient deployment on resource-constrained mobile robots. Most recently, his "SoRTS: Learned Tree Search for Long Horizon Social Robot Navigation" (2024, 5 citations) introduces a novel approach that integrates learned motion prediction with tree search, allowing robots to navigate crowded spaces safely and seamlessly. This work addresses the growing demand for trustworthy autonomous agents in human-shared environments. Patrikar’s research is characterized by its practical impact—developing tools and algorithms that directly advance the capabilities of autonomous systems in real-world settings.