System and method for determining a quantity in a motor vehicle

A system for determining iteratively at least one quantity of interest used to control at least one component of a motor-vehicle powertrain, said quantity of interest not being measured directly by a sensor,

• the system comprising:
• a supervised-learning-based estimating module, which receives as input a set of first input data, and which delivers as output at least one intermediate output quantity,
• a likelihood-checking module, which receives as input the set of first input data and a set of second input data, and which delivers as output a likelihood index, the likelihood index characterizing a distance metric between the set of first input data and a domain of first reference input data, and/or between the set of second input data and a domain of second reference input data, and the likelihood index becoming bad the higher the value of the distance metric,
• a downstream processing module, which receives as input the likelihood index and the intermediate output quantity, and which delivers as output a final output quantity corresponding to said quantity of interest,
• the downstream processing module being configured so that:
• when the likelihood index is good, the final output quantity is representative of the intermediate output quantity, the final output quantity then being directly equal to the intermediate output quantity, or corresponding to values of the intermediate output quantity after application of a filter,
• when the likelihood index is bad, the final output quantity is obtained from a model of a physical representation.