Wheeled Mobile Robot Control System Design

With computer technology, control theory, artificial intelligence theory, sensor technology developing, mobile robots(MR) have been developed to a new stage. Indoor MR has been extensive attention in academic circles at home and aboard, because that the indoor environment has a high degree of structure and less affected by uncertainties, which leads to reducing the complexity of the research. An indoor wheeled mobile robot(WMR) based on differential drive control is studied, and a WMR control system based on DSP and FPGA is designed in this paper.Firstly, the paper carries out the entire program design and selection of specific devices, based on the analysis of the function requirements of the control system. Then, hardware circuits of the control system is designed. Hardware circuits includes DSP minimum system, FPGA minimum system, parallel communication between DSP and FPGA, signal acquisition of FPGA, driver control, human-computer interaction(HCI), power module, etc.Then, the positioning and motion control algorithms is studied in the paper. Positioning and motion control are two primary issues of MR, and also two key factors of achieving the MR functions. The paper focuses on the navigation control algorithm based on dual-encoder positioning, and completes Matlab simulation and verification. At the same time, the realization ideas based on fiber optic sensor positioning&tracking is given.Finally, software design is completed, based on the hardware circuit and control algorithms. The software part is expanded due to the system functions, including system test, navigation based on dual-encoder positioning, tracking based on fiber optic sensor positioning, and objects grabbing. During software design&implemention process, F28335CPU Timer interrupt is used to complete tasks of updating given control value&sending commands of each sub-task module, scanning handle key, reading sensor information, and SCI wireless communication; eCAN module is used to communicate with servo drivers by CANopen agreement, achieving motors’synchronous control; the idea of finite state machine is used to deal with encoder data. Meanwhile, navigation based on dual-encoder positioning is verified. The experiments show that the designed WMR control system meets the desired function requirements. During the experiment at the maximum speed of lm/s, the system has a good navigation control effect, and the final positioning accuracy can be achieved in less than1cm.