Session Recap: Past, Present, and Future of Robotics Software

[Travis_M.]

Peter Corke's vast experience in robotics spans many robotics areas. In this session, Corke discussed his experiences and looked at some of the priorities facing the industry in the future.

Robotic arms have been around for some time, and Corke began working with them in the early 1980s. In the 1990s, rapid changes in computer technology resulted in improvements for robots and controllers. Some examples of changes include the proliferation of real-time OSs and field-programmable gate arrays (FPGAs). Along with these hardware changes, standardized software tools such as the Robotics Toolbox, which consists of a collection of standard algorithms, allow rapid application development and provide a point of comparison for those developing new tools. Corke wasn’t alone in his development of standard tools. Starting in the 1990s and continuing today, there has been an explosion of tools (“glueware”) such as ROS, JAUS, and NI LabVIEW system design software. Roboticists assemble their robots using these tools to handle aspects of system design such as UIs, safety, navigation, instrument I/O, and datalogging.

Today, hundreds of tools exist. Determining how to assemble these tools to build a robot is one of the biggest challenges. Corke continues to work on his Robotics Toolbox by working to port the toolchain to a diverse set of environments such as Python, ROS, and LabVIEW. Corke gives some general advice to NI: stick to what you’re good at. I/O and sensor connectivity has always been a strong point, and coupled with an FPGA, NI can make big contributions in robotics.

Corke is excited about the future of robotics education. Advances in tablet computers and electronic textbooks have opened doors for a more interactive education experience for future engineers. Robotics is more accessible than ever through the growing ecosystem of low-cost controllers and components sold through Internet stores.