Process for Calibrating and Filling Feature Tables

A feature calibration is the process of calibrating lookup tables and their normalizers by comparing a collection of lookup tables to a statistical model. The strategy determines the collection of lookup tables. A feature refers to the object that contains the statistical model and the collection of lookup tables.

This is the process for calibrating, filling, and validating feature lookup tables for a subsystem that estimates engine torque.

Step		Description
1	Set Up Feature and Strategy	Add a feature and construct or import a strategy from a Simulink® model file.
2	Set Up Models for Calibration	Import and select a response model that relates to the feature that you are filling. A Model-Based Calibration Toolbox™ response model is a function of a set of variables. Typically, you construct a model using the Model Browser.
3	Set Up Variables and Constants (optional)	Optionally, align the feature inputs to a response model. You can use variables, constants, and aliases to create function models.
4	Initialize Feature and Calibration Tables (optional)	Optionally, initialize the feature model lookup table parameters.
5	Fill Feature Lookup Tables	Initialize and optimize your feature model lookup table parameters.
6	Validate Feature Fill	Validate the feature fill by comparing the feature model and response model outputs at the same input values.
7	Import and Export Calibrations	Export feature calibration tables to a file.