top of page

I-PAR-5: DIGIHALL platform for trustable-by-design AI-based system

Experiment Description

In this experiment DIGIHALL supports Sherpa Engineering in the development of an intelligent control system of a simulated fleet of autonomous shuttles. The fleet must ensure the transport of passengers along a closed circuit including several stations near activity zones, living areas and in connection with extra-site transports. Areas are open to pedestrian traffic, which leads to important safety constraints. The focus of the experiment is on the development of a safe, reliable and quality-assured transportation service by means of both, appropriate supervision of fleet behavior, and autonomous operation of single shuttles.


To achieve this goal and with the support of DIGIHALL, Sherpa Engineering uses the DIGIHALL Platform for trustworthy AI-based systems and bridges it with its own design tool suite. The resulting platform builds on top the Papyrus modeling technology and of simulators such as Simulink, Dymola and possibly Gazebo, and it enables (i) the specification of vehicles' software architecture and behavior, (ii) the safety analysis of designed behavior against road scenarios and traffic rules and (iii) testing to ensure safety, reliability and appropriate quality of service of the composition of perception, supervision and control components, that integrate efficient but possibly unreliable machine learning algorithms, with manually designed model-based algorithms, that have less flexibility but provide well-defined stability and performance guarantees.

List Title

Expected Impact

For SMEs, the expected impact is in accelerating design and development and improving the quality and the consistency of the design through the use of an integrated environment. To this extent, the experiment will achieve such results by:

  • Linking each requirement specification (safety and quality/performance) with the simulation model;

  • Link, system definition / use-case and scenario definition;

  • Link, scenario definition/scenario execution / visualisation;

  • Visualizing directly the operational scenarios which allows to verify them and to validate the coherence of the objectives

This will enable a fast transition between system definition and simulation models, which facilitates design-refinement cycles and, in the end, leads to implemented solutions which are trusthworthy by design.
For DIGIHALL, the expected impact is the consolidation and further development of the Papyrus for Robotics platform, which favors its re-use to support other SMEs and companies in the domain of intelligent transportation systems, such as collaborating drones and ground vehicles for the autonomous transport of goods.

bottom of page