NEURO FUZZY CONTROL OF ROBOTIC ARM MOVEMENT Thesis report submitted in the partial fulfillment of requirements for the award of the degree of Master of Engineering in Electronics Instrumentation Control Engineering Submitted by ANSHUL Roll No 80751026 Under Guidance of Ms GAGANDEEP KAUR Lecturer SG EIED DEPARTMENT OF ELECTRICAL AND INSTRUMENTATION ENGINEERING THAPAR UNIVERSITY Patiala 147001 Punjab INDIA June 2009 THAPAR UNIVERSITY PATIALA 1 THAPAR UNIVERSITY PATIALA 2 ACKNOWLEDGEMENT First of all I would like to thank the almighty who has always guided me to work on the right path of life This work would not have been possible without the encouragement and able guidance of my supervisor Ms Gagandeep Kaur Lecturer SG Electrical and Instrumentation Department Most of the novel ideas and solutions found in this thesis are the result of our numerous stimulating discussions Her timely feedback and editorial comments were also valuable for writing this thesis I am thankful to the Head of the Department of Electrical and Instrumentation Engineering Thapar University Patiala Dr Samarjit Ghosh for his encouragement support and for providing the facilities for the completion of this thesis I am also thankful to the entire faculty and staff members of Electrical and Instrumentation Department for their direct indirect help and co operation Their enthusiasm and optimism made their experience both rewarding and enjoyable I am deeply indebted to my parents and brother Vipul and sisters Vanita Manu for their inspiration and ever encouraging moral support which enabled me to pursue my studies I am also thankful to the authors whose works I have consulted and quoted in this work Last but not the least I would like to thank God for not letting me down at the time of crisis and showing me the silver lining in the dark clouds Lastly I would like to thank friends for always supporting me and being there on my side whenever I needed them ANSHUL ROLL NO 80751026 THAPAR UNIVERSITY PATIALA 3 ABSTRACT This thesis first outlines the theory historical background and application of neural networks and fuzzy logic The review of neural networks and fuzzy logic is followed by a discussion of the combination of the two technologies neuro fuzzy techniques The two tools have been successfully combined to maximize their individual strengths and compensate for shortcomings A survey is given of previous work done in applying these technologies to control systems The problem of moving a robotic arm in the presence of an obstacle is discussed In particular trajectory planning of a planar redundant manipulator is studied The primary weakness of previous methods for determining acceptable trajectories is the massive amount of computer time needed to obtain a solution Neuro fuzzy systems offer not only the benefit of the parallel nature of its computations but also the ability to learn the control of an arm by following a human s example Several neuro fuzzy controllers are trained using sample data obtained from a human s control of a robotic arm Their performance is quantified and compared It is shown that the definition of the fuzzy membership functions plays a significant role in the ability of the neuro fuzzy controller to learn and generalize Possible directions for future work are suggested THAPAR UNIVERSITY PATIALA 4 LIST OF CONTENTS Certificate i Acknowledgement ii Abstract iii List of contents iv List of figures vi List of tables viii List of abbreviations ix CHAPTER 1 INTRODUCTION 1 1 1Introduction about robotics 1 1 1 1 Basic components 1 1 1 2 Three laws of robotics 2 1 2 Control of robotic arm 2 1 3 Fuzzy Logic 4 1 4 Neural networks 9 1 5 Neuro Fuzzy systems 17 1 5 1 Adaptive neuro fuzzy inference system 20 CHAPTER 2 REVIEW OF LITERATURE 25 2 1 Fuzzy Logic 25 2 2 Neural Network 27 2 3 Neuro Fuzzy Technology 29 2 4 Trajectory Planning 32 CHAPTER 3 ROBOTIC ARM PROBLEM 34 3 1 INTRODUCTION 34 3 1 Problem formulation 34 THAPAR UNIVERSITY PATIALA 5 CHAPTER 4 NEURO FUZZY SOLUTION 39 4 1 Introduction 39 4 2 State Representation 41 4 3 Robotic Arm Simulation 44 4 4 Neuro Fuzzy Programming 46 4 5 Neuro Fuzzy Controller 48 CHAPTER 5 EXPERIMENTAL RESULTS 52 5 1 INTRODUCTION 52 5 2 SIMULATION RESULTS 61 CHAPTER 6 CONCLUSION FUTURE WORK 69 6 1Conclusion 69 6 2 Future work 70 REFERENCES 71 APPENDIX A 76 Linear Function 76 Sinusoidal function 77 Two dimensional input function 77 The learning rate constant 79 The momentum constant 79 The fuzzy membership functions 80 THAPAR UNIVERSITY PATIALA 6 LIST OF FIGURES Figure1 1 A typical fuzzy system 5 Figure1 2 A sample fuzzy system 6 Figure1 3 An artificial neuron 10 Figure1 4 A feed forward neural network 10 Figure1 5 A forward neural network 11 Figure1 6 the Graph of sample activation function 14 Figure1 7 A fuzzy system whose membership functions are adjusted by a neural network 17 Figure1 8 A fuzzy system defined by a neural network 17 Figure1 9 A neural network of fuzzy neurons 18 Figure1 10 A fuzzy system with neural network rule base 19 Figure 1 11 ANFIS architecture with two fuzzy if then rules 21 Figure3 1 A diagram of robotic arm in presence of an obstacle 33 Figure 3 2 A sample configuration for the problem application 34 Figure3 3 The Robotic arm definition variables 35 Figure3 4 Normalized dimensions for remote arm and reachable region 37 Figure 4 1 A sample neuro fuzzy system 39 Figure4 2 The relative position of the objects is same 41 Figure4 3 The obstacle and goal switch places 41 Figure4 4 The state representation for the obstacle problem 42 Figure4 5 The robotic arm link data structure 43 Figure4 6 A sample training trajectory obtained from the simulator 44 Figure4 7 The neuro Fuzzy data structure 45 Figure4 8 A sample FNN data file 46 Figure4 9 Determination of a collision between the obstacle and the link 48 Figure4 10 Source code for collision detection function 49 Figure5 1 The Fuzzy membership function definition FNN1 50 Figure5 2 The Fuzzy membership function definition FNN2 51 THAPAR UNIVERSITY PATIALA 7 Figure5 3 The Fuzzy membership function definition FNN3 51 Figure5 4 The RMS training error of FNN1 FNN2 and FNN3 52 Figure5 5 Histogram of the neuro fuzzy controller for FNN1 FNN2 FNN3 52 Figure5 6 Samples of collisions while under neuro fuzzy control 53 Figure5 7 Samples of neuro fuzzy control in which constraint limits were exceeded 54 Figure5 8 Samples of successful control by the neuro fuzzy controller 55 Figure5 9 The fuzzy membership function