Interoperability

The enormous growth in the field of robotics has created a large number and variant of robot products, including manipulator arms, mobile bases and grippers. To test algorithms on these robotic products can be difficult due to the lack of interoperable hardware and software. Interoperability is the biggest challenge facing the adoption of automation and robotics in industrial applications in Australia (see Chapter 3) presenting commercial and technical challenges. Most users employ a mix of robotic systems from a variety of vendors, making standards for interoperability an urgent requirement in most industries apart from manufacturing, where the rigour in supply chain management in automotive manufacturing has ensured that interoperability is considered a standard feature. The future progress of robotic research relies on continued development of robust unified standards.

The creation of open-source platforms represents a good starting point and will ideally lead to the development of standards vetted by the open-source community. A good example of an open-source robotics platform is the robot operating system (ROS). First created in 2007 at Stanford and developed by start-up Willow Garage, it has now become ubiquitous in academia and is starting to have impact in industry. The term OS is a misnomer; ROS is component-based middleware that allows for the rapid assembly of existing and new components, running on one or more networked computers, to create a robot. The ROS ecosystem includes data loggers, data replay and visualisation. ROS has a large developer group which is guided by the Open Source Robotics Foundation. ROS-Industrial is a consortium that extends ROS to industrial applications. Non-open-source proprietary modules are now being developed for the ROS ecosystem. The accessibility of such plug and play frameworks is critical for ongoing R&D on robotic systems [AAS18]. It is also an area where Australia could potentially take a leading role for very little investment.