In this video, we present KIMLAB's recent work toward hardware-based reinforcement learning (RL) and the development of character-like bipedal robots. We explore the use of HybridLeg as a platform for expressive character robots such as Olaf or Snogie, equipped with a lantern-shaped, sensorized mechanical cover that can safely support whole-body contacts.Humanoid robots are inherently unstable, making fall management a critical challenge, especially for on-hardware RL, where falls are frequent and unavoidable. To address this, we introduce a protective mechanical design that mitigates impact damage during falls while enabling the robot to autonomously recover to a standing posture. This capability allows the robot to self-reset and reinitialize after each trial without human intervention, which is essential for scalable real-world hardware RL. The system integrates a multimodal fall detection framework combining inertial, proprioceptive, and acoustic sensing, along with an improved stance phase detection algorithm. This work highlights a practical pathway toward robust, self-recovering humanoid platforms suitable for long-horizon, real-world reinforcement learning experiments.This work was presented at SII 2026, and the paper will be available at ieeexplore soon.
- Kenta Hirashima, Daniel Campos Zamora, Kevin Gim, Joohyung Kim, “Self-Rising Bipedal Robot for Embracing Fall Impact and Fall Detection with Multimodal Sensing,” IEEE/SICE International Symposium on System Integration (SII 2026), 2026
If you are interested in humanoid locomotion and unique leg mechanisms, we hope you enjoy this video and our HybridLeg approach, along with the authors’ previous paper below.
- Implementation of Untethered Biped Robots Utilizing Serial-Parallel Hybrid Leg Mechanisms
ieeexplore.ieee.org/document/10769967
