KAIST DRCD Lab

Dynamically-Consistent Trajectory Optimization for Legged Robots via Contact Point Decomposition

KAIST Humanoid v0.5

[RA-L] A Modular Residual Learning Framework to Enhance Model-Based Approach for Robust Locomotion

A Single Humanoid Leg That Balances, Hops, and Flips

[RA-L] Online Friction Coefficient Identification for Legged Robots Using Smoothed Contact Gradients

[ICRA] A Learning Framework for Diverse Legged Robot Locomotion Using Barrier-Based Style Rewards

Winter Hiking with KAIST Hound!

[IJRR] Contact-Implicit MPC: Controlling Diverse Quadruped Motions Without Pre-Planned Contact Modes

KAIST Hound, a Quadruped Robot 100m World Record

Seamless Reaction Strategy for Bipedal Locomotion Exploiting Real-Time Nonlinear MPC

Invariant Smoother for Legged Robot State Estimation with Dynamic Contact Event Information

Agile and versatile climbing on ferromagnetic surfaces with a quadrupedal robot

Contact-Implicit Differential Dynamic Programming for Model Predictive Control(2022 IROS)

Legged Robot State Estimation With Dynamic Contact Event Information(2021 IROS)