Application of Fuzzy Optimization Decision Method on Robot Tactile Suit Data Processing

Aimed at the puzzle of robot tactile suit data processing that not only deals with the computer software, but also it is related to the hardware such as tactile sensor, signal conditioning circuit and etc., in this paper we proposed a method of fuzzy optimization decision for solving pattern recognition and classification problems related to robot tactile sensing suit. A robot tactile sensing suit system is a robotic device which obtains the shape, hardness, surface details of contacting objects. Fuzzy decision trees play important roles in many fields such as pattern recognition and classification. We have tried to use tactile images for inputting the data. It provided a framework that generates fuzzy decision trees, as well as fuzzy sets for input data in the paper. The algorithm based fuzzy decision method firstly collects enough training data for generating a practical decision tree. It then uses fuzzy statistics method to calculate fuzzy sets for representing the training data in order to increase generation speed. The algorithm has been applied to a general purpose robot tactile sensing system. Based on the fuzzy optimization decision method, the contact objects can be online recognized precisely. The experiment results in laboratory show that the method can be successfully used to the intelligent robot tactile suit system data processing.

[1]  Han-Pang Huang,et al.  A learning fuzzy decision tree and its application to tactile image , 1998, Proceedings. 1998 IEEE/RSJ International Conference on Intelligent Robots and Systems. Innovations in Theory, Practice and Applications (Cat. No.98CH36190).

[2]  Christian Balkenius,et al.  Neural network models of haptic shape perception , 2007, Robotics Auton. Syst..

[3]  Ren C. Luo,et al.  Tactile array sensor for object identification using complex moments , 1987, Proceedings. 1987 IEEE International Conference on Robotics and Automation.

[4]  Domenico Famularo,et al.  A Fuzzy Classifier for Tactile Sensing , 1997, J. Intell. Robotic Syst..

[5]  J. Rives.  FID3: fuzzy induction decision tree , 1990, [1990] Proceedings. First International Symposium on Uncertainty Modeling and Analysis.

[6]  J. Bezdek,et al.  Generalized k -nearest neighbor rules , 1986 .

[7]  Hannu Koivisto,et al.  Fuzzy classifier identification using decision tree and multiobjective evolutionary algorithms , 2008, Int. J. Approx. Reason..

[8]  Hannu Koivisto,et al.  Identification of interpretable and accurate fuzzy classifiers and function estimators with hybrid methods , 2007, Appl. Soft Comput..

[9]  Kenneth A. Loparo,et al.  Mathematical transforms and correlation techniques for object recognition using tactile data , 1989, IEEE Trans. Robotics Autom..

[10]  Mark Lee,et al.  Review Article Tactile sensing for mechatronics—a state of the art survey , 1999 .

[11]  Thomas Parisini,et al.  Towards the realization of an artificial tactile system: fine-form discrimination by a tensorial tactile sensor array and neural inversion algorithms , 1995, IEEE Trans. Syst. Man Cybern..