A Dual-Core Real-Time Embedded System for Vision-Based Automated Guided Vehicle

As a cell of flexible manufacturing system (FMS), automated guided vehicle (AGV) is widely used in factory. In this paper, we presented a novel technology used for vision-guided AGV. An embedded dual-core processor system with real-time operating system (RTOS) DSP/BIOS is designed for system manager and image processor. We adopt DSP TMS320DM642 as image processor and ARM LPC2210 as controller. The embedded RTOS DSP/BIOS is transplanted on DSP to construct a software development platform for tasks management, which enhance reliability and real-time response of system. Two CCD cameras are fixed on vehicle to capture the scene of lane. One, on the front of vehicle, is used for prediction, the other, at the center of vehicle, is used for accurate positioning. Radio frequency identification (RFID) reader is loaded on vehicle to collect the information stored in RFID tag which placed by the side of lane. The experimental results demonstrate the advantages of real-time and robust.