Dongsik Chang

Georgia Institute of Technology, Oregon State University

Papers

Total Citations

H-Index

About

Dongsik Chang is a robotics researcher whose work lies at the intersection of motion planning, control theory, and autonomous manipulation, with a particular emphasis on challenging underwater environments. His research addresses fundamental challenges in enabling robots to operate efficiently and robustly in unstructured, perception-limited settings. Chang’s most cited work, “Glider CT: Analysis and Experimental Validation” (2016, 9 citations), provides foundational insights into the dynamics and control of underwater gliders, a key platform for long-duration ocean monitoring. More recently, he has pioneered approaches to kinodynamic motion planning, as demonstrated in “Parallelized Control-Aware Motion Planning With Learned Controller Proxies” (2023, 7 citations), which introduces a method for robots to find energy-efficient, hazard-free paths while accounting for real-world controller deviations from planned trajectories. His work on “Real-Time Generative Grasping with Spatio-temporal Sparse Convolution” (2023, 5 citations) tackles the critical problem of fast, robust grasp identification for mobile manipulators in noisy, dynamic environments like the underwater domain. Through these contributions, Chang is advancing the frontier of autonomous robotics, making systems more capable of performing complex tasks in some of the most difficult operational conditions on Earth.