Enhancement of a VEX Robot with an Onboard Vision System

摘要

This paper documents the augmentation of Raspberry PI (plus its camera module) to a typical VEX robot, yielding a mobile robot with onboard image processing capability. The VEX robot denotes a robot constructed using the VEX robotic kit, which is controlled and programmed by the Cortex microprocessor that comes with the kit. Though the Cortex microprocessor is powerful enough for motion control and processing sensors such as integrated encoder, sonar range sensor, and touch sensor, the current VEX robotic kit does not provide a true vision sensor that allows real-time image processing. To enhance the functionality of the VEX robot, a Raspberry PI (with its camera module) is introduced to the robotic system, where the PI will be dedicated for image processing. Information extracted from the video sequence are then sent via UART port to the Cortex, which controls the basic motion of the robot. Detailed procedures to construct the vision-enhanced VEX robot is described. The resulting robotic system with the Raspberry PI and Cortex combination greatly expands the usage and application domain of the robot, making it suitable for more complex and advanced tasks such as vision-based control, visual servoing, and visual tracking. This paper also provides a sample lesson plan that consists of basic motion in 2D (controlling the mobile base only), trajectory generation in 3D (controlling a robotic arm on top of the mobile base), and examples of advanced topics (e.g., vision-based control). We believe these projects will significantly improve an existing robotics curriculum, enhance students' hand-on skills, and lead to more challenging undergraduate projects.