Lydia E. Kavraki

Fibration-RRT can tackle high-dimensional multi-robot motion planning by seamlessly integrating projection and decomposition strategies, proving to be both efficient and robust.

Kinodynamic motion planning just got a whole lot faster: AkinoPDF achieves microsecond-level planning times by exploiting differential flatness for analytical solutions.

LLMs can generate performant Python bindings for large C++ robotics libraries, but watch out for shared pointers and trampolines.

High-DoF robots can now navigate complex environments while actively maintaining visual contact with multiple objects, thanks to a new motion planner that bakes perception directly into the planning process.