Watch this video to see the Rover in action.
This demo session showcases our use of UML for Real-Time (UML-RT) and Papyrus-RT in order to drive the Eclipse PolarSys Rover. UML-RT is a profile of UML specifically designed for real-time embedded (RTE) systems. It has a long, successful track record of application and tool support via, e.g., IBM Rational RoseRT, IBM RSA-RTE, and now Eclipse Papyrus for RealTime (Papyrus-RT). Papyrus-RT is an Eclipse-based, open-source modelling development environment for UML-RT systems. It allows the generation of complete, executable code from models.
The PolarSys Rover is a small robotic system used for promoting collaboration across PolarSys solutions. It consists of a small robot driven by four wheels. It includes a set of sensors and actuators such as a detection sensor, and embeds a single-board computer Raspberry PI 3 running Linux. It can be used to develop and showcase simple robotic applications (e.g., follow the line, obstacle detection and avoidance).
Objective: this demo session intends to accompany a regular talk on modelling and code generation for RTE systems with UML-RT and Papyrus-RT. The goal is to attract people interested in the model-based development of robotic applications. It give attendees an opportunity to gain practical experience with Papyrus-RT in this specific context. It includes the following steps:
- a) Light introduction to the core modelling concepts in UML-RT (capsules, protocols, ports, and state machines), and b) the graphical representation and model modification in Papyrus-RT;
- a) Introduction to the PolarSys Rover, and b) and adaptations required to the models to support the generation and deployment of fully functional code on the target platform;
- a) Presentation and use of more complete models in Papyrus-RT to showcase applications of the Rover.
Audience: the intended audience includes everybody interested in using modelling techniques for the development of robotic systems. Industrial practitioners, researchers from academia and industry, as well as educators are welcome and should be able to benefit. Some prior exposures to class modelling and behavioural modelling with state machines in UML is useful. Some knowledge in single-board computers is not mandatory.