Jason Gibson

Georgia Institute of Technology, United States Naval Research Laboratory

Papers

Total Citations

H-Index

About

Jason Gibson is a rising force in robotics and autonomous systems, whose work masterfully bridges the gap between perception, control, and multi-agent coordination. His research centers on three key areas: vision-based navigation, swarm intelligence, and optimal control theory. Gibson’s most influential contribution is the introduction of **deep optical flow (DOF) dynamics**, a novel framework that fuses optical flow with robot dynamics within a model predictive control (MPC) loop. This work, published in 2020 and garnering 37 citations, enables aggressive, perception-aware navigation by allowing drones to react to visual cues in real time—a critical step toward truly autonomous flight in cluttered environments. In parallel, his development of a **time-based A* path-planning method** for lighter-than-air multi-agent systems (26 citations) provides a scalable solution for synchronized swarm coordination, ensuring that multiple agents complete tasks simultaneously. Most recently, Gibson has pushed the boundaries of stochastic control theory by generalizing variational inference MPC using **Tsallis divergence** (17 citations), offering a more flexible framework for handling uncertainty in complex, nonlinear systems. With these contributions, Gibson is not only advancing the theoretical foundations of robot autonomy but also providing practical tools for the next generation of agile, cooperative robots.