Tanvir Parhar

Papers

Total Citations

H-Index

About

Tanvir Parhar is a researcher at the intersection of agricultural robotics and deep learning, dedicated to automating high-throughput plant phenotyping and precision agriculture. His work focuses on developing intelligent perception and control systems that enable robots to interact with complex, unstructured plant environments. Parhar’s major contributions include the creation of StalkNet, a pioneering deep learning pipeline for the high-throughput measurement of sorghum stalk count and width, which has garnered 62 citations and set a benchmark for automated plant trait analysis. He further advanced this field with a deep learning-based stalk grasping pipeline, integrating detection and manipulation for robotic harvesting. Expanding into viticulture, Parhar developed a deep reinforcement learning policy for a 7-degree-of-freedom robot to precisely reach pruning locations within dormant grapevine canopies, addressing the challenge of navigating occluded, natural growth. His work demonstrates a clear trajectory from perception to autonomous action, with cumulative citations exceeding 87. By combining convolutional neural networks with robotic planning, Parhar is helping to replace labor-intensive manual phenotyping with scalable, data-driven solutions, making him a notable figure in the push toward fully autonomous agricultural systems.