Legged Robot State Estimation in Slippery Environments Using Invariant Extended Kalman Filter with Velocity Update
Sangli Teng,Mark Wilfried Mueller,Koushil Sreenath,Sangli Teng,Mark Wilfried Mueller,Koushil Sreenath
This paper proposes a state estimator for legged robots operating in slippery environments. An Invariant Extended Kalman Filter (InEKF) is implemented to fuse inertial and velocity measurements from a tracking camera and leg kinematic constraints. The misalignment between the camera and the robot-frame is also modeled thus enabling auto-calibration of camera pose. The leg kinematics based velocity...
