Intelligent control of flexible manipulator link with friction

The programmable multifunctional rigid manipulators (industrial robots) need continuous improvements to fulfill all conflicting requirements such as improving the robot efficiency and increasing its ability to perform specific and complex tasks in different environments with high precision performance. Consequently, the constructions of the physical robot model and the control strategies need to be improved. The improved robot systems, which are called flexible robot systems (FRSs), are constructed from light mass manipulator links with specific geometry shapes, which are characterized by the high aspect ratio (the ratio of the arm length to its width). The cost of these improvements makes the dynamics of the FRS highly nonlinear and leads to a complicated coupled mechanical systems due to the flexible nature of the light weight robot manipulators and nonlinear effects of friction of the hub rotor joints. This in turn affects the positional accuracy of the end effector of the flexible manipulator link (FML). Such complicated systems face great challenges which are represented by obtaining sensible accurate mathematical model of the FRSs with friction and designing active control strategies to perform the desired tasks with good accuracy at different operating configurations (different positions and directions of the hub rotor, and different loading conditions at the end point of the FML). To study these problems in detail, this work was proposed. The main objectives of this work are: - Adapting a systematic procedure to identify and optimize a mathematical dynamic model of the proposed FRS with friction. - After obtaining a confident mathematical model, the proposed direct and learning control strategies, which are designed and constructed based on the knowledge-based fuzzy logic system, are investigated and improved based on the simulation technique before using it in real implementation for validation purpose.