Unmanned Systems Technology 001 | UAV Factory Penguin C | Real-time operating systems | Hirth S1218 two-stroke twin | Base stations | ASV C-Enduro | Composites | Datacomms

28 T he heart of any autonomous system is reliable control software that can respond in real time. In addition, given the growing interconnectivity between UAVs and external support systems, the software needs to be secure from hacking or even hijack attempts. These are the aims behind the continuing development of real-time operating systems (RTOSs), which are based on the idea of a scheduling function that ‘slices’ instructions into time slots so that a routine such as checking a sensor can be run within a defined time, called the maximum latency. This is a very different concept from that of a traditional operating system, where ‘interrupt’ instructions determine when things happen. An interrupt from a keyboard or incoming communications can override processing of data or delay the response to that data. This can delay activity in a system, which you see all too clearly with a PC suddenly running slowly for no apparent reason, for example. As a result, RTOSs have carved out a niche for themselves in safety-critical control systems in industries such as manufacturing, power and water supply, telecoms and aerospace, and are now at the heart of the development of autonomous systems. Many such systems are based around a dedicated scheduler that has been built on a particular processor chip hardware called a ‘bare metal’ design. The scheduler itself is not complicated, and can serve a specific application well, but adding a communications capability to connect to a network or the wider internet creates a need to integrate it In your own time Nick Flaherty explains the issues around developing the essential software for safe, secure and reliable autonomous systems The elements of a real-time operating system and software for an unmanned vehicle (Courtesy of Lynx) November 2014 | Unmanned Systems Technology