This demonstration shows a programmable neural network platform running on a FPGA board, which captures an image from the camera, processes it to detect objects, and then displays the detected objects on the LCD display. The object detection is processed in the programmable neural network accelerator we have designed for automotive applications.

 This demonstration shows a simple neural network platform that recognizes hand-written characters. A hand-written character is captured by the camera, and then the recognized character is displayed on the 7-segment display.

 This demonstration shows that a computational platform we have designed controls a toy car. The platform was designed by including an embedded processor and several peripheral devices. The processor has been designed to have instructions compatible to ARM9. The user on the remote host sends commands to the platform through a Bluetooth wireless channel, and the platform interprets the command to derive detailed operations necessary to control the toy car. The remote host is in fact a cell phone, running an application to accept the direction the user wants to go.

 This demonstration shows that a computational platform we have designed controls a big toy car. The platform is designed by including an embedded processor and several peripheral devices. The processor called Core-A has proprietary instructions designed to take the merits of the previous state-of-the-art instruction sets. The user on the remote host sends commands to the platform through a Bluetooth wireless channel, and the platform interprets the command to derive detailed operations necessary to control the toy car.