For wheeled robot motion control, brushless DC motor speed control system of wheeled mobile robot parameters time-varying, nonlinear, uncertainties and other factors, with the traditional PID control algorithm is difficult to meet the control requirements this article combination of traditional PID control technology and fuzzy control technology, is designed based on the parameters of fuzzy self-tuning PID DC motor control. This article describes the principles and design steps of fuzzy self-tuning PID Matlab simulation of the control program, the results show that the controller has achieved a good speed and stability.

The wheeled robot with non-integrity constraints is a typical nonlinear system, in order to achieve the ideal path tracing, presented a theory based on fuzzy neural network control. Centralized compensation system based on neural network uncertainty can be arbitrary-precision approximation of continuous nonlinear functions as well as the complex uncertainties with adaptive and learning ability. By MATLAB simulation showed that the control method to ensure fast convergence and error robustness of parameter uncertainties and external disturbance.

Wheeled robot walk because of its simple structure,continuous advantages of smooth, fast and easy to control, is often used as special robot, but wheeled robot at run time, facing one of the major issues is how to avoid obstacles, increase system reliability. Introduction of fuzzy control algorithm in this article, by rational design of fuzzy rules, plan routes, to achieve effective avoidance, collision of the simulation results show that the fuzzy control can be effective, improves the reliability of robots running.

Indoor temperature model is one of the large time-delay controlled object, and the response performance of system get worse. In consideration of the Fuzzy PID controllers strong robustness, this article adopts a Fuzzy PID-Smith control method ,and the control effect proves to be good.

For continuous transportation of granular bulk materials measurement and weighing is important link in industrial and agricultural production, warehousing and product flows. Improve said heavy of precision and speed is the key of dynamic measurement, and traditional for the continuous delivery of materials weighing system can not meet the requirements of quickly weighed and measured accurately, while sometimes variable, and nonlinear and random, factors in actual process in the of interference, in order to overcome these factors on continuous conveying material measurement system of effect, this article based on RBF Neural network model for foundation made a dynamic clustering algorithm optimization strategy.By means algorithm calculated the center of the base function, further extended RBF Neural network constants from hidden layer to output layer , then using least-squares method calculates the weight matrix to determine the final network structure and main parameters. This article on measurement of dynamic process has simulation test research, and BP neural network for effect comparison, simulation results indicates that, based on RBF of neural network on continuous conveying bulk material measurement control system is more effective, and weighing accuracy and speed is improved, in order to achieve the continuous delivery of materials weighing process optimization control, and at the same time provides an effective way to solve the existing problems of this type of system.