STEP: State Estimator for Legged Robots Using a Preintegrated Foot Velocity Factor

IEEE Robotics and Automation Letters, published online, 2022

Yeeun Kim (1)    Byeongho Yu (1)    Eungchang Mason Lee (1)    Joon-Ha Kim (1)    Hae-won Park (1)    Hyun Myung (1)

(1) Korea Advanced Institute of Science and Technology

---

Abstract

We propose a novel state estimator for legged robots, STEP, achieved through a novel preintegrated foot velocity factor. In the preintegrated foot velocity factor, the usual non-slip assumption is not adopted. Instead, the end effector velocity becomes observable by exploiting the body speed obtained from a stereo camera. In other words, the preintegrated end effector’s pose can be estimated. Another advantage of our approach is that it eliminates the necessity for a contact detection step, unlike the typical approaches. The proposed method has also been validated in harsh-environment simulations and real-world experiments containing uneven or slippery terrains.

Paper: [PDF]

Video

Bibtex

@ARTICLE{STEPKim2022,
  author={Kim, Yeeun and Yu, Byeongho and Lee, Eungchang Mason and Kim, Joon-ha and Park, Hae-won and Myung, Hyun},
  journal={IEEE Robotics and Automation Letters},
  title={STEP: State Estimator for Legged Robots Using a Preintegrated Foot Velocity Factor},
  year={2022},
  volume={7},
  number={2},
  pages={4456-4463},
  doi={10.1109/LRA.2022.3150844}
}