We show that autonomous, high-speed trail navigation can be achieved through a single idea: encoding the trail as a semantic layer of the terrain.
We show that autonomous, high-speed trail navigation can be achieved through a single idea: encoding the trail as a semantic layer of the terrain.
Research teams at the RAI Institute had several papers and workshops accepted for CoRL 2026, covering topics such as handheld data collection, loco-manipulation, robot mapping...
We introduce Sceniris, a highly efficient procedural scene generation framework for rapidly generating large-scale, collision-free scene variations.
We propose Bi-modal Routing for Imitation Data via Gated Experts (BRIDGE), a mixture of diffusion policy experts that routes between specialist heads conditioned on the...
Current policy evaluation protocols often conflate policy and execution errors by treating the robot controller as a fixed implementation detail. By testing various controllers across...
We present the Koala Gripper system, a data capture device and robotic gripper platform that improves dexterity and grasp capability compared to parallel jaw grippers...
We present Eucalyptus, an extension of UMI designed for sustained, force-aware data collection. Eucalyptus combines a robustified and ergonomics-oriented mechanical design with a high-fidelity multimodal...
By replacing external motion capture with onboard depth-camera perception and local policy evaluation, we can achieved full autonomy and untethered outdoor execution to perceive, locate,...
Spot performs dynamic manipulation outside the lab using onboard depth cameras and a computer.
To perform tasks that require forceful manipulation and delicacy with a single robot design requires not only a powerful robot but also smart control. Providing...
We present RoMo-GS, an online 3DGS system. Instead of averaging N discrete sub-frames as in prior motion blur-aware 3DGS, we analytically integrate each projected Gaussian...
Integrating locomotion and manipulation is essential for robot autonomy, but scaling standard Reinforcement Learning (RL) to complex tasks is severely bottlenecked by the slow, manual...