Model-based hardware in the loop control of collaborative robots: Simulink and Python based interfaces

摘要

Collaborative robots are growing exponentially, with applications in the most diverse domains and operating side by side with humans in unstructured and changing environments. Such a scenario requires robot programmers to develop complex applications, integrating Artificial Intelligence (AI), sensing and control, both rapidly and efficiently. Python and model-based design software packages such as Simulink are widely used in engineering, and when it comes to robotics those tools are very important for robot design, simulation and the development of control algorithms before the implementation on the real robot. Python is widely used in AI while Simulink supports several hardware components, facilitating rapid deployment of the developed programs on the target hardware. In this study, a novel Python library (iiwaPy) and SimulinkIIWA are presented to interface KUKA iiwa robots. The iiwaPy is based on the Transmission Control Protocol/Internet Protocol (TCP/IP) and the SimulinkIIWA on the User Datagram Protocol (UDP). Both support robot motion control (point-to-point and real-time) at the joint level and the end-effector (EEF) level, as well as the monitoring of the robot state (position, orientation, torques feedback). Open-source code and example applications are provided showing the flexibility and the ease of use of the proposed interfaces.