International Journal of Engineering Trends and Technology (IJETT) – Volume 4 Issue 6- June 2013 Intelligent Controling System for Path Finder Tanvi Patel#1 # Department of Biomedical Engineering, Govt. Engineering College, Gandhinagar, India Abstract— Nowadays in the world of globalisation and drastically increasing technologies and changing environment, problems of the visually impaired person are increasing day by day. This paper shows how the intelligent controlling stimulation system and algorithm developed and verified in MATLAB (.fis file) for the path finder. The scheme proposed below will allow the path finder to navigate in an unknown environment with auto guidance facility based on fuzzy logic (IF-THEN rules) approach. The inputs considered here signals from the sensors after obstacle detection and output we get desired angle rotation for the wheels. The proposed scheme has been presented below for the path finder for navigating through an unknown environment and known environment their merits and demerits. Keywords— Auto guidance, Fuzzy logic, Motor rotation, Obstacle avoidance. I. INTRODUCTION Looking towards the scientific development and rapid increasing demands of the society. The problems of the visually impaired persons have been a major consideration regarding their navigation in their day to day life. The international health society stated that 80% of these persons (600 Millions around the world) are living in underdeveloped countries and suffering many social and psychological problems due to poverty and insufficient education. These countries succeeded to overcome this problem by increasing the health care and applying special training programs [1]. Path finder with auto guidance facility is perfect solution for the problems of the visually impaired persons. It based on the intelligent controlling systems which have the ability to navigate through an known and unknown environment based on Fuzzy logic (IF-THEN rules). Rule-based fuzzy logic provides a scientific formalism for reasoning and decision making with uncertain and imprecise information. The main advantages of a fuzzy navigation strategy lie in the ability to extract heuristic rules from human experience, and to obviate the need for an analytical model of the process [2]. In this approach a framework is developed to be implemented in the path finder for obstacle avoidance and controlling. It based on fuzzy logic (IF-THEN rules) based on linguistic terms instead of mathematical equations. For example, considering a temperature control system one can write“IF room temperature is WARM, THEN set fan speed to MEDIUM” [3] Here is a algorithm for the obstacle avoidance and desired steering angle prepared fuzzy inference systems with Fuzzy Logic Toolbox software in MATLAB. Stimulation can also prepared by other software’s like LABVIEW. ISSN: 2231-5381 II. FUZZY LOGIC APPROACH Fuzzy logic uses the entire interval between one and zero, and can therefore be used to closely mimic human reasoning. The design process of a fuzzy logic system can generally be separated into three stages [4]: Fuzzification Rule Base Defuzzification The fuzzification module transforms numerical variables in fuzzy sets, which can be manipulated by the controller [5].The input are fuzzified by the membership function. This membership function (triangular, trapezoidal, bell shaped etc.) maps the crisp inputs to a degree of values of membership for each function. Then the rules are developed after the membership functions are developed. After the rules evaluation, in the defuzzification stage the output fuzzy values are again mapped to provide the crisp output. This can be done by various defuzzification method like centre of gravity, centre of sums etc. III. FUZZY CONTROLLER The fuzzy logic controllers (FLC) make a non-linear mapping between the input and the output using membership functions and linguistic rules (normally in the form if-then). In this paper we have developed a system in which we have taken inputs from 3pairs of sensors (left, front and right) and output is the desired steering angle of rotation of motor, two wheels(left wheel and right wheel) are turned. Specific membership function is chosen for mapping of crisp input to the fuzzy output in range [0, 1]. The ultrasound sensors, IR sensors etc can be used for obstacle detection. According to proper reasoning provided intelligent controlling system is used for navigating the path finder in unknown and known environment. On the bases of input and output parameters membership function are defined and rules are set A fuzzy controller is choosen on the information from the sensors. In this paper we used Mamdani fuzzy controller. http://www.ijettjournal.org Page 2275 International Journal of Engineering Trends and Technology (IJETT) – Volume 4 Issue 6- June 2013 A. Structure This controller consists of three main parts: fuzzification, rules evaluation and defuzzification. The inputs have to be crisp values in order to allow the fuzzification using membership functions, and the outputs of this controller are also crisp values. The membership function of the Left sensor is shown in Fig. 1[4]. The input to the controller from the sensor is considered as Low, Medium and High signal. Fig3.Membership function for right wheel Fig 1: Mamdani Fuzzy Controller Block Diagram B. Fuzzification Process For mapping the crisp values to fuzzy ones, you have to evaluate their membership degree using membership functions. With this you get one fuzzy value for each crisp input. The input from the sensors(left, right and front) are taken with signals obtained after detection of obstacles according to inputs membership function is selected and mapping is done accordingly in range [0 1].The membership function considered over here is Triangular function Fig 2. D. Defuzzification Process To defuzzify the outputs we use the center of sums method. In this method we take the output from each contributing rule, and then we add them. The center of sums is one of the most popular methods for defuzzification because it is very easy to implement and gives good results. The equation (1) is shown below [6]: ∗= ∑ ∑ .∑ ∑ ( ( ) ) ........ (1) C. Rule Evaluation The controllers define rules for controlling the motor and obtain a desired steering angle for wheels. If the signal is S1-low and S2- low and S3- low then the angle for Left wheel is NR and Right wheel will be NR. The following 27 rules are defined as follows: Fig 2: Membership function of Left Sensor ISSN: 2231-5381 IV. RESULT AND CONCLUSION In this paper we developed a algorithm in Matlab(.fis file) based on Fuzzy Logic(If –Then rules) for path finder to navigate safely from an unknown and known environment by providing auto guidance facility. We get the desired steering angle of motor , it has two wheels left and right wheels. Intelligent controlling system based on Fuzzy logic is considered for decision making and controlling the motor rotation. It can be further carried in future with more inputs and implemented in many systems by developing more rules. The major problem regarding this algorithm creation is during its implementation and complexity increases as a person develop many rules as per his/her knowledge. With help of past experiences and knowledge gathered we can develop human-reasoning based operating system by defining certain rules based on their gained knowledge and get the desired output. The rule viewer of the membership function is shown below in Fig 4. When at distances obstacle is considered left sensor1 = 5, S2-front = 2.5 and S3 = 1 then Left-wheel = -40 degree and Right-wheel = - 20 degree. http://www.ijettjournal.org Page 2276 International Journal of Engineering Trends and Technology (IJETT) – Volume 4 Issue 6- June 2013 TABLE I Where, SML –SMALL, NS –NEGATIVELY SMALL, MS –MEDIUM SMALL, VS –VERY SMALL, NR –NO ROTATION OR SLIGHT ROTATION, BG –BIG, PB –POSITIVELY BIG, MB –MEDIUM BIG, TB – TOO BIG Input From The Sensors S1_ S2_ S3_ Left Front Right Output The Steering Angle Left _ Right _ Wheel Wheel RULE - 1 HIGH HIGH HIGH NR NR RULE -2 HIGH HIGH MEDIUM BG BG RULE -3 HIGH HIGH LOW PB PB RULE -4 HIGH MEDIUM HIGH NR NR RULE -5 HIGH MEDIUM MEDIUM BG BG RULE -6 HIGH MEDIUM LOW MB MB RULE -7 HIGH LOW HIGH NR NR RULE -8 HIGH LOW MEDIUM TB TB RULE -9 HIGH LOW LOW BG BG RULE -10 MEDIUM HIGH HIGH NS NS RULE -11 MEDIUM HIGH MEDIUM VS NS RULE -12 MEDIUM HIGH LOW TB PB RULE -13 MEDIUM MEDIUM HIGH SML SML RULE -14 MEDIUM MEDIUM MEDIUM NR NR RULE -15 MEDIUM MEDIUM LOW MB MB RULE -16 MEDIUM LOW HIGH NR NR RULE -17 MEDIUM LOW MEDIUM NR NR RULE -18 MEDIUM LOW LOW BG BG RULE -19 LOW HIGH HIGH MS MS RULE -20 LOW HIGH MEDIUM SML SML RULE -21 LOW HIGH LOW MS MS RULE -22 LOW MEDIUM HIGH MS VS RULE -23 LOW MEDIUM MEDIUM MS VS RULE -24 LOW MEDIUM LOW BG BG RULE -25 LOW LOW HIGH NR NR RULE -26 LOW LOW MEDIUM VS VS RULE -27 LOW LOW LOW NR NR Fig 5: Rule viewer for the membership functions REFERENCE [1.] W. Gharieb*, G. Nagib** “Smart Cane for Blinds” Ain Shams University and Faculty of Engineering Cairo University [2.] M.K. Singh, D.R.Parhi, S.Bhowmik, S.K.Kashyap “Intelligent Controller for Mobile Robot: Fuzzy Logic Approach” The 12th International Conference of International Association for Computer Methods and Advances in Geomechanics (IACMAG) 16 October, 2008, Goa, India. [3.] DOGAN IBRAHIM, TAYSEER ALSHANABLEH “An Undergraduate Fuzzy Logic Control Lab Using a Line Following Robot” Near East University, Lefkosa, Mersin 10, Turkey 18 March 2009. [4.] Hung-Jin Chen, Yu-Hsin Kao “Design of Fuzzy Control Applied to the Path Following of Lego-NXT System” Proceedings of 2012 International Conference on Fuzzy Theory and Its Applications National Chung Hsing University, Taichung, Taiwan, Nov.16-18, 2012. [5.] GABRIEL PIRES and URBANO NUNES “A Wheelchair Steered through Voice Commands and Assisted by a Reactive FuzzyLogic Controller” Journal of Intelligent and Robotic Systems 34: 301–314, 2002. [6.] ISSN: 2231-5381 Pedro Ponce-Cruz • Fernando D. Ramirez-Figueroa “Intelligent Control Systems with Lab VIEW™” http://www.ijettjournal.org Page 2277