Development of a distributed simulation environment and model driven engineering framework to support the verification & validation of complex autonomy components

The Harwell Robotics and Autonomy Facility (HRAF) is a facility funded by the European Space Agency (ESA) to support the integration and verification & validation of autonomy systems and associated technologies from unit up to mission level. An essential capability of the HRAF core infrastructure is the provision of software-based simulation and modelling environments, based on a Service Orientated Architecture using the High-Level Architecture (HLA), IEEE 1516-2010 standard for distributed simulation. A key feature of HLA is support of information exchange models, called Federation Object Models (FOM), which can be tailored. A FOM can be standardized (typically within a given domain) or developed from scratch. On behalf of ESA, a number of new functionalities aimed to enhance and extend current simulation and modelling capabilities were added to the HRAF core infrastructure. A distributed simulation architecture between multiple contractor (GMV) and agency (ESA) sites was developed, based on HLA, together with the SISO Space Reference FOM standard (SISO-STD-018). To validate this new simulation environment and contained software, two representative scenarios were selected for implementation and testing at various fidelities, based on reused GNC simulators: 1) Mars Sample Return (MSR) Rendezvous and Capture (RVC) phase, and 2) Precision landing of a spacecraft on a low gravity Near Earth Object (NEO). Additionally, a Model-Driven Engineering Framework for the semi-automated generation of any generic space domain FOM based using the Space Reference FOM was developed. The tool performs a transformation of an abstract input system model describing the scenario to be simulated in annotated SysML form, from which a SpaceFOM compliant FOM is generated. This paper presents new HRAF functionalities and their validation, specifically: a distributed simulation environment based on HLA and the Space Reference FOM, and a Model-Driven Engineering framework for semiautomatic generation of the corresponding FOM. It provides an overview of the federation architecture, main components and methodologies applied in federating simulations using existing simulator elements (software and hardware) and validation the system-level tests that have been performed.