Dhruv A. Thanki

University of Delaware, Florida Institute for Human and Machine Cognition

Papers

Total Citations

H-Index

About

Dhruv A. Thanki is pioneering the frontier of agile and stable humanoid robotics, with a primary focus on reactive motion planning and multi-contact teleoperation. His most impactful work introduces a sequential Model Predictive Control (MPC) approach that leverages "safe corridors"—directed convex decompositions of free space—to enable bipedal robots to navigate highly cluttered environments with moving obstacles. This contribution, published in 2022 and garnering 31 citations, addresses a critical bottleneck in real-world robot mobility by ensuring dynamic stability amid dense, unpredictable surroundings. Thanki further advances the field by tackling the challenge of humanoid teleoperation during multi-contact tasks, such as using hand contacts on non-coplanar surfaces. His 2025 work on stability-aware retargeting directly confronts the risks of motor torque saturation and loss of stability—common failure modes when operators command complex, contact-rich motions. By developing algorithms that preserve equilibrium while translating human intent to robot action, Thanki is bridging the gap between high-level human commands and low-level robot stability. His research is essential reading for anyone working on legged locomotion, human-robot interaction, or control in unstructured environments.