Max Robotics — Global Service Robot Directory + US Certification

About

Venkatraman Narayanan is a robotics researcher whose work spans computer vision, motion planning, and autonomous navigation, with particular emphasis on enabling robots to perceive and interact intelligently with complex real-world environments. He is perhaps best known for PoseCNN, a landmark convolutional neural network architecture for 6D object pose estimation in cluttered scenes, which has garnered over 2,000 citations since its publication and remains a foundational reference in robotic manipulation and perception research. This work addressed the notoriously difficult challenge of identifying object orientations and positions despite occlusions and environmental clutter. Beyond perception, Narayanan has made significant contributions to heuristic search-based motion planning, developing the Multi-Heuristic A* (MHA*) family of algorithms, which cleverly leverage multiple, potentially inadmissible heuristics to guide efficient high-dimensional planning — a persistent challenge in robotics. His subsequent work on Dynamic MHA* and improved uncalibrated heuristic variants reflects a sustained commitment to refining these foundations. He has also explored probabilistic edge-existence planning and topological path constraints, demonstrating breadth across planning paradigms. More recently, his work on EWareNet introduces emotion-aware pedestrian intent prediction for social robot navigation, signaling an expanding research vision toward human-robot coexistence. Collectively, his contributions have meaningfully advanced both the theoretical and practical frontiers of autonomous robotics.