Anshumali Shrivastava

Papers

Total Citations

153

H-Index

About

Anshumali Shrivastava is a prominent researcher working at the intersection of machine learning and robotics, with a particular focus on motion planning, task and motion planning (TMP), and sampling-based algorithms. His work addresses one of robotics' most persistent challenges: enabling robots to efficiently navigate complex, high-dimensional environments by intelligently combining learned experience with classical planning frameworks. Shrivastava's most influential contributions demonstrate a consistent drive to bridge discrete and continuous planning. His 2019 work on learning feasibility for task and motion planning (76 citations) showed how continuous geometric information could be meaningfully incorporated into discrete search to accelerate plan discovery. Complementing this, his research on leveraging local experiences for global motion planning (46 citations) introduced principled ways to guide sampling-based planners through geometrically difficult regions. His subsequent 2020 work (31 citations) extended these ideas into high-dimensional configuration spaces, improving generalization across diverse environments — a notoriously difficult problem in robotics. Collectively, these papers have accumulated over 150 citations, reflecting meaningful influence within the robotics planning community. Shrivastava's research offers a compelling vision for smarter, experience-driven robots capable of tackling real-world manipulation and navigation tasks with greater efficiency and adaptability.