Invariant Filtering for Legged Humanoid Locomotion on a Dynamic Rigid Surface

Article Robotics

Legged Robot State Estimation With Dynamic Contact Event Information

Joon-Ha Kim et al.

Summary: The algorithm applies Maximum A Posteriori estimation for legged robot state, using the Gauss-Newton algorithm and Schur Complement method. Components of cost function and Jacobian are derived utilizing SO(3) manifold structure, with reparameterization for proper linear algebra and vector calculus application. Additionally, a slip rejection method is proposed to reduce the impact of kinematics models fault modeling.

IEEE ROBOTICS AND AUTOMATION LETTERS (2021)

Add to Collection

Article Automation & Control Systems

The Salted Kalman Filter: Kalman filtering on hybrid dynamical systems

Nathan J. Kong et al.

Summary: This paper discusses how probability distributions propagate through hybrid dynamical systems and introduces the Salted Kalman Filter (SKF) method, which outperforms other methods in state estimation after hybrid transition events.

AUTOMATICA (2021)

Add to Collection

Proceedings Paper Automation & Control Systems

DILIGENT-KIO: A Proprioceptive Base Estimator for Humanoid Robots using Extended Kalman Filtering on Matrix Lie Groups 2021

Prashanth Ramadoss et al.

Summary: This paper introduces a contact-aided inertial-kinematic floating base estimation for humanoid robots using a geometrically meaningful estimator that allows fast convergence even with large initialization errors. The method utilizes a matrix Lie group to represent the state and relative positions-and-orientations of contact feet from forward kinematics as observations to estimate the additional six degrees of freedom describing the base position-and-orientation of humanoid robots. An experimental validation on the iCub humanoid platform demonstrates the effectiveness of the proposed estimator.

2021 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA 2021) (2021)

Add to Collection

Proceedings Paper Automation & Control Systems

Legged Robot State Estimation in Slippery Environments Using Invariant Extended Kalman Filter with Velocity Update

Sangli Teng et al.

Summary: This paper presents a state estimator for legged robots operating in slippery environments, utilizing an Invariant Extended Kalman Filter (InEKF) and a method for online noise parameter tuning to adapt to varying camera measurement noise. The filter's consistency with the nonlinear system is demonstrated through nonlinear observability analysis.

2021 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA 2021) (2021)

Add to Collection

Proceedings Paper Automation & Control Systems

Invariant Extended Kalman Filtering for Hybrid Models of Bipedal Robot Walking

Yuan Gao et al.

Summary: This paper introduces a hybrid invariant extended Kalman filtering method for a class of nonlinear hybrid dynamical systems, providing sufficient stability conditions for asymptotic error convergence, complete characterization of nonlinear jump maps, and simplified observer design.

IFAC PAPERSONLINE (2021)

Add to Collection

Article Robotics