Fault Tolerance Analysis and Optimization of Centralized Control Platform Based on Artificial Intelligence and Optimization Algorithm

Abstract

To enhance the reliability and self-healing of the system, the research on fault tolerance of the reconfigurable modular centralized control center is its development trend. Most of the previous research has focused on hardware redundancy. Improving fault tolerance performance is an essential topic in the research of centralized control platforms. Firstly, the problem of centralized fault tolerance in the working configuration of a reconfigurable manipulator is studied. The effect of each hinge on fault tolerance in the existing configuration is studied with the criterion of manoeuvrability and tolerable space. The fault module was first modelled to represent the system architecture information. A modular motion rule based on autonomous recombination technology is proposed. A self-organizing deformation algorithm with fault tolerance is studied. The fault tolerance of the motion pairs is compensated by adding a small number of motion pairs to ensure the configuration characteristics. With the addition of a failure compensation device, the joint’s range of motion was reduced, and the fault tolerance rate was enhanced. After the failure of the robot arm, the fault tolerant control method can still ensure that the robot arm can perform work in its tolerable working space. The test results show that the fault tolerance analysis method is practical and feasible. It lays a theoretical foundation for the application in aerospace, industry and other fields.