Distributed IMU Pose Estimation of Hyper-Redundant Manipulator Based on ESKF

Aiming at the pose estimation problem of hyper-redundant manipulators, the paper proposes a method based on distributed IMU modules installed within the structure of the manipulator. This proposed method computes joint angles of the manipulator by the IMU data to obtain real-time pose feedback. Kinematic mapping from the joint space to the workspace is established using the Denavit-Hartenberg (D-H) method. On this basis, error state Kalman filter (ESKF) algorithm is adopted to fuse IMU data. Finally, an experimental platform using a motion capture system is utilized for validating the method. Experimental results reveal that the ESKF algorithm achieves high accuracy in solving the IMU attitude, and the method can efficiently acquire the pose estimation of the hyper-redundant manipulator.