Arun L. Bishop

Papers

Total Citations

115

H-Index

About

Arun L. Bishop is redefining how robots interact with the physical world through contact-rich manipulation and locomotion. His pioneering work on **Contact-Implicit Model Predictive Control (CI-MPC)** — already garnering 81 citations since 2024 — introduces a bilevel planning framework that allows robots to anticipate and exploit contacts, enabling fluid, dynamic behaviors in tasks from walking to object manipulation. To make such complex control computationally tractable, Bishop developed **ReLU-QP**, a GPU-accelerated quadratic programming solver that reformulates ADMM as a deep neural network, achieving real-time performance for high-dimensional problems (17 citations). Beyond control theory, his **SLoMo** system (17 citations) bridges computer vision and robotics by transferring skilled motions from casual videos of humans and animals to legged robots, synthesizing physically plausible keypoint trajectories from monocular footage. This work opens doors to learning complex behaviors without expensive motion capture. Bishop’s contributions sit at the intersection of optimization, learning, and dynamics, offering practical tools for agile, contact-aware autonomy. His research is essential reading for anyone building robots that must push, pull, step, or climb in the real world.