A Scalable and Unified Multi-Control Framework for KUKA LBR iiwa Collaborative Robots

Abstract

The trend towards industrialization and digitalization has led more and more companies to deploy robots in their manufacturing facilities. In the field of collaborative robotics, the KUKA LBR iiwa is one of the benchmark robots. To communicate these robots with different components and generate an interoperability infrastructure, the software libraries provided by Robot Operating System are now widely widespread. However, the latency that such communication between devices often generates, diminishes the potential of machine learning control techniques, such as reinforcement learning, when the robot must react swiftly in an unstructured environment. This paper presents a scalable and unified control system that supports both Robot Operating System and direct control and outperforms current control frameworks in terms of exploiting the functionalities of the KUKA LBR iiwa. The framework's documentation can be found at https://libiiwa.readthedocs.io and its source code is available on GitHub at https://github.com/Toni-SM/libiiwa.