Harsh Muriki
I am a 1st-year BS/MS student in Robotics at Georgia Tech advised by Dr. Sonia Chernova. I am also working as a Graduate Research Assistant with Dr. Yongsheng Chen.
✉️ vmuriki3@gatech.edu
Research Interest: Human-Robot Interaction, Robot Learning, Understanding Unseen and Dynamic Environments, AgTech Robotics
Publications
Robotic Phenotyping for Small-Scale Urban Farms
H. Muriki, R. Teo, V. Sengupta, and A. Hu
2025 IEEE International Conference on Robotics and Automation (ICRA) in Atlanta, USA
V(HS)IT: Using Vision Transformers to Enhance Prospects for Rapid Lettuce Tissue Nutrient Detection and Precision Nutrient Control
A. Cohen, H. Muriki, Y. Sun, S. Narayanan, L. Gan, and Y. Chen
Computers and Electronics in Agriculture 2025
Integrating ORB-SLAM2 and Khronos for Real-Time Object-Aware Semantic Mapping with Mobile Manipulators
H. Muriki, N. Nguyen, R. Rustagi, M. Patel, K. Ramachandruni, and S. Chernova
Hello-Robot Newsletter April 2025
Research Projects
Semantic Proactive Assistance for Households
3-D Spatio-Temporal mapping with Hello-Robot-Stretch to understand and track dynamic indoor environments.
View Project
Deep-Learning-Driven Lettuce Analysis
Collected the largest hyperspectral lettuce dataset and trained vision-transformer models to predict nutrient content and growth.
Robotic Phenotyping for Small Scale Urban Farms
3D Point-cloud processing and pose-estimation pipelines for accurate plant monitoring and robotic pollination.
View ProjectCourse Projects
IRL Project: HINTeract-Hint-Driven Hierarchical Learning for Robotic Furniture Assembly
Built a modular HRL framework where a Sawyer arm requests on-demand human hints to grasp, align, and insert parts, achieving sample-efficient toy-table assembly in Robosuite with 260 expert demos and open-sourced code.
View Project
MRS Project: CableCrop RL-Multi-Cable Robot Planning for Agricultural Data Capture
Designed a coordinated RGB+hyperspectral cable-robot pair and trained PPO/MAPPO agents in simulation to maximize crop coverage and data richness while dodging collisions for scalable precision-ag monitoring.
View Project
HRI Project: LM-NavGrasp-Language-Guided Navigation & Grasping on Hello Robot Stretch
Developed a pipeline that fuses large vision-language models (CLIP + SAM) with ROS-Nav so Stretch can parse a natural-language object description, autonomously navigate to the correct table, and execute a grasp on the specified item in cluttered lab scenes.
View Project