Title :
Collision detection for a flexible cantilever-beam subject to random disturbance based on innovation process
Author_Institution :
Dept. of Mech. & Syst. Eng., Kyoto Inst. of Technol., Japan
Abstract :
In this paper, an algorithm for detection of collisions for a flexible cantilever-beam subject to random disturbance based on a model-based fault diagnosis method is presented. The mathematical model of the flexible cantilever-beam is described by an Euler-Bernoulli type partial differential equation. The collision detection filter is constructed based on the Kalman filter for the finite-dimensional system corresponding to the mathematical model of the flexible beam. If collisions of the flexible beam and environmental obstacles occur, the statistics of the residual signal (innovation process) change. To detect such a statistical change of the residual signal due to the collision, a chi-squared like test is introduced. Several numerical results are presented.
Keywords :
Kalman filters; collision avoidance; fault diagnosis; flexible manipulators; manipulator dynamics; partial differential equations; Euler-Bernoulli type partial differential equation; Kalman filter; chi-squared like test; collision detection; collision detection filter; dynamics; environmental obstacles; finite-dimensional system; flexible cantilever-beam; innovation process; mathematical model; model-based fault diagnosis method; numerical results; random disturbance; residual signal statistics; Actuators; Fault detection; Fault diagnosis; Filters; Manipulator dynamics; Mathematical model; Partial differential equations; Statistics; Technological innovation; Testing;
Conference_Titel :
Control Applications, 2002. Proceedings of the 2002 International Conference on
Print_ISBN :
0-7803-7386-3
DOI :
10.1109/CCA.2002.1038771
