2.7 Using simulation

Assurance objective: Simulations used as part of the assurance process provide a representation of the real world sufficient for their use.

Contextual description: Simulation may be used in a number of different roles as part of assuring RAS (such as training of ML systems, testing and understanding system behaviour). In all cases it is important for assurance that a sufficient level of correspondence can be demonstrated between the simulation model and the real-world that it models. What is sufficient will depend on what is being modelled and why. For example simulations of the sensing functions will require a detailed correspondence in the simulation to raw sensor data from the real sensors as well as an accurate model of environmental effects, whereas a detailed model of the vehicle itself and its dynamic behaviour is not required. Simulation may include hardware-in-the-loop approaches where simulated inputs are provided to real physical systems.

Practical guidance:

Contact us

Assuring Autonomy International Programme

assuring-autonomy@york.ac.uk
+44 (0)1904 325345
Institute for Safe Autonomy, University of York, Deramore Lane, York YO10 5GH

Contact us

Assuring Autonomy International Programme

assuring-autonomy@york.ac.uk
+44 (0)1904 325345
Institute for Safe Autonomy, University of York, Deramore Lane, York YO10 5GH