M. Karlsson, F. Näsström, and G. Haapalahti (Sweden)
Simulation, Multi-sensor, HLA, radar, IR, Physically based modeling.
This paper describes a modular framework for high quality simulation of multi-sensor systems that is being developed to increase our ability to simulate sensor systems in varied conditions. The incentive for this was the need for production of extensive datasets for algorithm develop ment as well as evaluation of, mainly autonomous, sensor systems. In this work, HLA (High Level Architecture) has been used to integrate simulation software, terrain models, atmospheric models and sensor data analysis algorithms into a distributed and reconfigurable multi-sensor simulation framework. The main focus of the framework is high fidelity sensor data simulation; hence non real time federates are mainly used. This allows us to use calculation intensive simulation tools that are primarily based on detailed physical models. A significant part of the integration work has been to extend the RPR-FOM (Real-time Platform Reference Federation Object Model) to include the capability for distributed sensor simulation on a functionality level. The aim of this paper is to present this framework and to contribute with design solutions chosen for added objects and interactions. At the end some examples are briefly presented, where this frame work has been successfully used in various projects for algorithm development and system evaluation.
Important Links:
Go Back
