Rishi Ramakrishnan

Papers

Total Citations

H-Index

About

Rishi Ramakrishnan’s research lies at the intersection of robotic perception and machine learning, with a focus on enabling autonomous systems to operate reliably in complex, real-world environments. His early work on shadow compensation for outdoor perception (2015, 16 citations) tackled a critical challenge for vision-based robots: handling illumination variations caused by occlusions. By developing a method to calculate lighting distributions in outdoor scenes, he provided a practical solution that enhances the robustness of perception modules—a foundational contribution for field robotics. More recently, Ramakrishnan has advanced the application of hyperspectral imaging in robotics, particularly for agriculture and mining. His 2017 study on hyperspectral CNN classification with limited training samples (4 citations) addresses a key bottleneck in deploying deep learning for material classification: the scarcity of labeled data. By demonstrating that convolutional neural networks can achieve strong performance with minimal training examples, he opened new possibilities for per-pixel thematic classification of materials. Though his citation counts are modest, his work reflects a thoughtful, problem-driven approach to making robotic perception more adaptive and data-efficient—an important step toward practical, deployable autonomous systems.