Calibration of parallel robots by evolutionary algorithm
The work describes parallel robot calibration procedure by evolutionary algorithm (EA). The algorithm uses nature-inspired features such as genetic information exchange, mutation, survival of the fittest, leap into the unknown and others. Special test function with two minimums peculiar to the calibration problem is used for the EA parameters assessment. The calibration procedure by EA is powerful instrument for improving of the parallel robot accuracy. They are more perspective and more reliable for the robot parameters identification in comparison with known gradient methods.
Unable to display preview. Download preview PDF.
- 1.Vischer P. Improving the accuracy of parallel robots. PhD-thesis No 1570, Ecole Polytechnique Federale de Lausanne. Switzerland. 1996.Google Scholar
- 2.Kirkpatrick S., Gerlatt C.D.Jr. and Vecchi M.P. Optimization by Simulated Annealing, Science, 220, 671–680, 1983.Google Scholar
- 3.Fogel D.B. Evolutionary Computation: Toward a New Philosophy of Machine Intelligence, IEEE Press, Piscataway, NJ.Google Scholar
- 4.First International Contest on Evolutionary Optimization. http:/iridia.ac.be/langerman/ICEO.htmlGoogle Scholar
- 5.Clavel R. Conception d'un robot parallele rapide a 4 degres de liberte, PhD-thesis No 925, Ecole Polytechnique Federale de Lausanne, Switzerland, 1991.Google Scholar