Robot Control Mode Design Based On Multi-source Information Fusion

With the development of robotic technology, people hope that robot can understand the environment and perceive its state better, thus reasoning and behavioral control can be more reliable. This requires the use of a variety of sensors in the design of robot in order to obtain a variety of information. How to effectively analyze and deal with the information of multi-sensors, and to make reliable decisions has become one of the key points of the robotic research. The reasoning results need to be realized through the actuators, so the motion control system must be considered while designing the robot. In this paper, the information fusion, reasoning and decision-making, motion control and behavior mode are analyzed and designed based on the wheeled mobile robot.In this paper, a wheeled mobile robot platform is built as an object of analysis and experiments. Firstly, the kinematics of the platform is analyzed, and its motion performance is discussed. Then the design target of the platform system is presented in:the limit of the performance. Combined with the design target, the selection of the sensors is given, and the circuit of the control system is designed, so the integration of the components to the system is completed.In order to deal with the information of multi-sensors more effectively, in this paper the multi-source information fusion technology is studied and applied to the self-made robot system. After the research and analysis of fuzzy logic method, the general process of fuzzy information fusion and the implementation methods of each step are given, then the method is applied to the reasoning decision-making of the wheeled mobile robot in its obstacle avoidance and navigation behavior, and this paper gets a fuzzy implication relation matrix after reasoning and calculation combined with the specific parameters.Motion control system is the basis for the robot to make a variety of actions. The Self-made robot adopts four DC motors to drive itself, so the nonlinear characteristic of this system is obvious and is difficult to establish a precise mathematical model, therefore it is suitable to apply PID control technology. On the basis of the basic principle of PID, this paper combines the characteristics of this system to design a feasible control algorithm. Then a lot of experiments are carried out using this algorithm, and the results show that the system can achieve good control effects. What's more, in order to make the system's reasoning decision-making layer and the motion execution layer combine better, this paper divides the field of the navigation control angle of fuzzy inference output, and analyses motion performance of the self-made robot, on the basis of this, the basic behavior modes of the robot are designed, and the corresponding parameters of each mode are given.At the end of this paper, the software architecture of the self-made robot system is designed. The function flow chart of each subroutine module is analyzed and designed, and the C language codes are compiled to carry out the actual experiments, which prove the feasibility of the method used in this paper.