Matthew Sivaprakasam
Papers
2
Total Citations
48
H-Index
2
About
Matthew Sivaprakasam is a leading researcher in off-road autonomous navigation, specializing in self-supervised learning, inverse reinforcement learning, and large-scale robotics datasets. His most impactful contributions center on advancing how autonomous vehicles perceive and navigate unstructured, rugged terrain. He is the driving force behind the TartanDrive series—TartanDrive 2.0 (2024, 25 citations) stands as one of the largest and most comprehensive off-road driving datasets, featuring multi-modal sensor data at speeds up to 15 m/s, designed to fuel self-supervised learning research. This work builds on his foundational TartanDrive 1.0, which set new standards for off-road data collection. In parallel, his paper "Learning Risk-Aware Costmaps via Inverse Reinforcement Learning for Off-Road Navigation" (2023, 23 citations) tackles the engineering-intensive challenge of costmap design, using expert driving data to train deep neural networks that predict risk-aware navigation costs. With a growing citation footprint, Sivaprakasam’s work bridges the gap between data-driven learning and practical off-road autonomy, providing both the datasets and algorithms that enable safer, more capable robotic exploration in challenging environments.
Research Focus
Key Achievements
Top Papers
- 1
- 2