ETAS ASCMO accurately predicts complex system behaviors by using a model based on a minimal number of measurements taken from the actual system, thus significantly reducing the effort required for testing on real-world systems—e.g., at the test bench or in the vehicle. As part of this process, the tool applies newly developed statistical learning procedures that permit a high level of model accuracy with relative ease. Once the accurate model has been completed, the control of an entire powertrain can be optimized automatically. With its easy and intuitive operation, ASCMO interactively guides the user through complex task sequences. Based on measurement data captured from the actual system on the test bench, the tool automatically feeds the mathematical model with parameter values. Statistical DOE (design of experiments) methods are used in measurement planning. The company claims that this approach has delivered a proven reduction of measuring activity in excess of 80% in a number of practical applications.
Booth 301 at SAE Commercial Vehicle Engineering Congress.