Avik Jain

Papers

Total Citations

H-Index

About

Avik Jain is a rising force in robotics and control theory, whose work centers on advancing model predictive control (MPC) for complex, real-world planning. His research tackles a fundamental challenge: bridging the gap between short, computationally feasible MPC horizons and the long-horizon tasks that robots must actually solve. In his highly cited 2021 paper, “Optimal Cost Design for Model Predictive Control,” Jain introduced a novel framework that systematically learns cost functions to align short-sighted MPC policies with long-term objectives. This contribution, which has already garnered 8 citations, provides a principled way to design cost terms that make nonconvex trajectory optimization both tractable and effective for tasks like autonomous navigation and manipulation. By enabling MPC to implicitly reason about future consequences without expanding its horizon, Jain’s work has significant implications for agile robotics, where computational speed is paramount. His approach offers a powerful alternative to traditional reward shaping, promising more robust and efficient planning in dynamic environments. As his research continues to evolve, Avik Jain is establishing himself as a key innovator in the intersection of optimal control and machine learning.