Max Robotics — Global Service Robot Directory + US Certification

About

Avinash Ummadisingu is a roboticist whose research focuses on intelligent manipulation, grasping, and perception for autonomous systems operating in unstructured environments. His work bridges hardware design and learning-based control, with a particular emphasis on food handling, transparent object perception, and mobile manipulation. Ummadisingu’s most cited work, “Uncertainty-aware Self-supervised Target-mass Grasping of Granular Foods” (26 citations), introduces a novel approach enabling robots to learn to grasp precise amounts of granular food items—a critical challenge for automating the diverse and rapidly changing menus in food packing industries. He further advanced perception for manipulation with “SAID-NeRF” (8 citations), a segmentation-aided neural radiance field method that achieves depth completion for transparent objects, a notoriously difficult problem in computer vision and robotics. His contributions extend to hardware innovation, including the design of the Two-Fingered Hand with Gear-Type Synchronization Mechanism (F2 Hand) and the Four-Axis Adaptive Fingers Hand (FAAF Hand), which improve grasping of small, offset, and misaligned objects. Ummadisingu also pioneered the first reinforcement learning-based system for targeted grasping with active vision on mobile manipulators. His work demonstrates a rare integration of mechanical design, perception, and learning, making significant strides toward practical, adaptable robotic systems for real-world applications.