# Fuzzy Logic Design of a Nondestructive Robotic Fruit Collector

• S. G. Tzafestas
• F. V. Hatzivasiliou
• S. K. Kaltsounis
Part of the International Series on Microprocessor-Based and Intelligent Systems Engineering book series (ISCA, volume 11)

## Abstract

After many years of research, the "repetitive movement robot" seems to have reached its peak of evolution and has become nowadays a routine in the product line. Recently, research was directed towards the developments of machines that are able to perform more complicated tasks. Examples are robots eqiupped with vision, built-in intelligence, double- arms and so on xc1,2. In this paper an attempt is made to create a higher level robot equipped with vision and artificial intelligence. An interesting application of artificial intelligence robotics can be found in the Hirota - Arai - Hachisu project xc5 where a fuzzy algorithm of twenty four rules is used to estimate the distance that the gripper must traverse according to the speed and the present distance between the gripper and the object on a belt conbeyor. Fuzzy logic theory was first indtroduced by Zadeh xc4 and suggests a mathematical formulation for the linguistic sets and human reasoning. In this way the human style of reasoning can be stimulated by a computer. The fuzzy logic has been successfully applied to industrial and chemical process control xc6,8 and recently attempts have been made to apply its methods to the control of robot arms.

In this paper, by "non destructive collecyion of fruits" we mean the process of collecting a fruir without destroying the actual tree. Two main issues that must be considered in any attempt to create a machine capable of non-destructive collection of fruits are the following:
1. a)

The fruits of most types of trees, and particularly the apple tree, are found in bunches. The fruits of a certain bunch are connected to some point of the tree and thus, as a result of this, each fruit of a certain bunch must be detached towards the side at which it is connected with a bunch. The obot, through the picture it receives, must determine the side on which it muts be detached.

2. b)

It is possible that a fruit of the bunch may not be directly visible to the robot, as its line of sight may be blocked from another fruit of the same bunch. This must be taken under consideration.

## Keywords

Membership Function Fuzzy Logic Fuzzy Subset Fuzzy Algorithm Maximum Membership
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

## References

1. [1]
S.G. Tzafestas.: Expert Systems in Engineering Applications, Springer — Verlag, Berlin, 1993.
2. [2]
S.G. Tiafestas.: Intelligent Robotic Systems, Marcel Dekker, New York, 1991.Google Scholar
3. [3]
R.G Gonzalez and R.E. Woods.: Digital Image Processing, Addison Wesley, pp.574–5, 1992.Google Scholar
4. [4]
L.A. Zadeh: Outline of a new approach to the analysis of complex systems and decision processes. IEEE Trans. Syst., Man, Cybern., vol.SMC-3, no.1,pp28–44,Jan.1973.Google Scholar
5. [5]
K. Hirota, Y. Arai and S. Hachisu: Moving mark recognition and moving object manipulation in fuzzy controlled robot. Control: Theory and Adv. Techn.,vol 2, no.3, pp 399–418, Nov. 1986.Google Scholar
6. [6]
E.H. Mamdani: Application of fuzzy logic to approximate reasoning using linguistic synthesis. IEEE Trans. Computers, vol c-26, no.12,Dec. 1977.Google Scholar
7. [7]
K.S. Fu, R.C. Gonzalez, C.S.G. Lee: Robotics: Control, Sensing, Vision and intelligence. McGraw-Hill 1987.Google Scholar
8. [8]
R.M. Tong: A control engineering review of fuzzy systems. Automatica,vol.13, pp.559–569,1977.