The Use Of Matlab For Robotic Control In An Undergraduate Robotics Laboratory

Abstract

An effective undergraduate robotics course will have strongly coupled laboratory and classroom components. It is important that the students experience the application of classroom theory. Often, this application is transparent when using the vendor supplied programming environments. For example, Cartesian move commands will move the robot to a desired point in the workspace without a need for explicit solution of the inverse kinematics problem by the student. Programming environments such as MATLAB, Maple, or C\C++ have long provided an ideal simulation environment for studying kinematic or dynamic robotic models. Environments such as MATLAB are especially ideal for engineering students with limited programming expertise. By taking advantage of the serial port capabilities in MATLAB's Release 12 and later versions, along with the ability to compile existing C\C++ code under the MATLAB shell, the instructor can devise assignments that allow the student to easily model and control robotic systems in the MATLAB environment. This paper discusses two approaches and representative laboratory assignments.