Péter Gáspár, Head of the Systems and Control Research Laboratory of SZTAKI, researcher of Coprologs, and also the Technical Director of the National Laboratory of Autonomous Systems (ARNL), held his inaugural lecture as an elected Full Member of the HAS on 19 October 2022.
The title of his talk was "Integrated control methods for self-driving road vehicles: from model-based control to data-driven control”. The lecture was held in the Great Hall of the HAS. The summary of his talk is as follows.
The complexity of traffic situations and the high number of quality requirements to be met require a high level of automation, both for individual vehicles and for vehicles cooperating with each other. Using the model-based control design methodology, one can design for highly complex systems robust control based on reduced model complexity. Such control guarantees the fulfilment of the requirements related to safety-critical quality criteria. The application of data-driven machine learning in vehicle control is a recent, but intensifying development. This interest in this methodology is due to the fact that it can be applied to systems of high complexity while also can handle a significant number of quality criteria. Research results and application examples demonstrate that quality requirements can be guaranteed and robust operation can be achieved simultaneously by the combined application of model-based control and data-driven machine learning methodologies.
Photo: HAS/Szigeti Tamás