Aerodynamic force and yaw moment control of a generic light truck using machine learning

• Applicant: Dr. Richard Semaan
• Subject Area: Fluid Mechanics
• Term: from 2017 to 2022
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 390189530

The objectives of this research project are the efficient drag reduction and the yaw stabilization of a generic transport vehicle model using machine-learning flow control. The model encapsulates the main features of a light truck, which are representative of a large class of current road vehicles. The project objectives are served by a specially-constructed rotating platform in the wind tunnel, which simulates frontal winds, side winds as well as transients between them. A particular focus is to allow for a rich set of enabling actuation mechanisms targeting various drag-inducing physical phenomena. This diversity is represented by seven independently operated unsteady actuators and many pressure sensors on the body. In addition, the self-learning model-free open- and closed-loop control allows to identify and to exploit the most effective nonlinear actuation opportunities in an unsupervised manner. Thus, control design becomes a regression problem which can be solved with the powerful tools of machine learning. The project shall distill an effective actuation strategy with large engineering benefits for ground vehicles targeting industrial exploitation. In addition, we aim at a general strategy for closed-loop aerodynamic force and torque control towards green and safe ground-and air-borne traffic.

• DFG Programme: Research Grants
• International Connection: France, USA
• Co-Investigator: Professor Dr. Bernd Rainer Noack
• Cooperation Partners: Professor Steven L. Brunton, Ph.D.; Privatdozent Dr. Laurent Cordier