Skip to Main Content
The subject of this paper is a state-estimation algorithm which uses the twist data, velocity and angular velocity, of the base of a micro-manipulator, placed on the end-effector of its macro counterpart, to estimate the flexural states of the flexible links of the macro-manipulator. The twist data are inferred from the acceleration signals delivered by an accelerometer array $a kinematically redundant array of tri-axial accelerometers. The array signals can also be utilized to calculate the translational and angular accelerations of the micro-manipulator base, which are in turn used to obtain a set of dynamics equations for the macro-manipulator, thus reducing the order of the dynamics model. Next, the dynamics equations of the macro-manipulator and the state-output relations are linearized, the latter in closed form so as to lower the computational cost in a control loop. The relations thus obtained are then used in an extended Kalman filter to estimate the flexural states of the system.
Robotics and Automation, 2003. Proceedings. ICRA '03. IEEE International Conference on (Volume:3 )
Date of Conference: 14-19 Sept. 2003