3D virtual modelling and stabilization control of triple links inverted pendulum on two-wheeled system using enhanced interval type-2 fuzzy logic control

• Main Author: Muhammad Firdaus, Masrom
• Format: Thesis
• Language: English
• Published: 2020
• Subjects: TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://umpir.ump.edu.my/id/eprint/30377/1/3D%20virtual%20modelling%20and%20stabilization%20control%20of%20triple%20links%20inverted%20pendulum%20on%20two-wheeled%20system.wm.pdf; http://umpir.ump.edu.my/id/eprint/30377/

This research embarks on the investigations for modelling and control of a triple link inverted pendulum on two-wheeled system. This new model of triple links inverted pendulum on two-wheeled system is tested due to its highly non-linearity, more complex nature and more degree of freedom, as compared to the existing triple link model which is solely limited to ‘on cart’ systems. This model is proposed to represent a real hardware application for example mobile robot, walking robot and an airplane landing system. Moreover, this proposed model enables it to be more flexible so it can perform various task like walking, jumping and stair climbing task. The model is designed in SimWise 4D to overcome limitations of the linearized mathematical model, whereby it does not represent the actual system and is unable to retain the model's complexity. This modelling technique also allow provides visualization for user to evaluate the performance of the system. The completed model system is further incorporated with Matlab/Simulink for analysis, control design and evaluation purposes. The Interval Type-2 Fuzzy Logic Control (IT2FLC) approach is implemented as the control strategy for system stabilization in an upright position and disturbance rejection during static and linear motions (forward and backward). Without a specific method introduced to determine the optimal values of input and output control gains, as well as the IT2FLC control parameters, optimization algorithm has been applied to enhance the system's performance in term of stability and disturbance rejection. Two optimization algorithms are presented in this work which are Spiral Dynamic Algorithm (SDA) and Particle Swarm Optimization (PSO). These two algorithms are chosen due to the advantages and the capability to find the globa optima. The developed control approaches are then tested and evaluated within simulation exercises in SimWise 4D for visualization purposes. With this, the first and second links have shown better angular position error by 9.3% and 28.4% respectively through the Particle Swarm Optimization (PSO), which proven its superiority as compared to the Spiral Dynamic Algorithm (SDA). Moreover, the proposed controller has outperformed SDA in maximum disturbance rejection by 3% and 16% with disturbance being applied on Link 2 and Link 3 separately. Promising results obtained have further demonstrated the system's potential to maintain stability in an upright position under a variety of conditions. Therefore, the system and control strategy proposed within this study has demonstrated their practicality in real-time applications, including the range of mobile robots with extensive functionality.