Active neuro-fuzzy integrated vehicle dynamics controller to improve the vehicle handling adn stability at complicated maneuvers

• Autores: Rana Raouf Hasan Farag
• Directores de la Tesis: Vicente Díaz López (dir. tes.), Beatriz López Boada (dir. tes.)
• Lectura: En la Universidad Carlos III de Madrid ( España ) en 2013
• Idioma: inglés
• Tribunal Calificador de la Tesis: José Luis San Román García (presid.), José Luis Muñoz Sanz (secret.), Julio Fuentes Losa (voc.)
• Materias:
  • Ciencias tecnológicas
    • Tecnología de vehículos de motor
      • Automóviles
• Enlaces
  • Tesis en acceso abierto en: e-Archivo
• Resumen

  • With the recent advancements in vehicle’s industry, driving safety in passenger vehicles is considered one of the key issues in designing any vehicle. According to other studies Electronic Stability Control (ESC) is considered to be the greatest road safety innovation since the seatbelt. Yet ESC has its drawbacks, that encouraged the development of other stability systems to correct or compensate these draw backs. But to efficiently make up for the ESC problems the integration of various control systems is needed, which is a pretty complicated task on its own. Lately, solving this stability problem became a hot research topic accompanied by the market demands for improving the available stability systems. Therefore, this thesis aims to add an innovative approach to help improve the vehicle stability. This approach consists of an intelligent algorithm that collects data about the vehicle characteristics and behavior. Then it uses an Artificial Neural Network to construct a fuzzy logic control system through learning from the optimum control values that was generated beforehand by the intelligent algorithm. This way, the proposed controller didn’t depend only on experts’ knowledge like the other controllers presented in the literature. This makes the controller more generic and reliable which is a very important aspect in designing a safety critical controller, like the presented one, where any fault in it can lead to a fatal accident. Also using the technique of using an Artificial Neural Network to construct a fuzzy logic control allows benefiting from the learning and autoautoadaption capability of neural networks and the smooth controlling performance that fuzzy logic controllers offers. Simulations results show the effectiveness of the proposed controller for improving the vehicle stability in different driving maneuvers. Where the controller’s results were compared to an uncontrolled vehicle and another vehicle controlled by a controller from the literature. -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------