ETD

Archivio digitale delle tesi discusse presso l'Università di Pisa

Tesi etd-11212012-160933


Tipo di tesi
Tesi di laurea specialistica
Autore
PROFUMO, LUCA
URN
etd-11212012-160933
Titolo
Haptic support systems for curve negotiation in a driving simulator
Dipartimento
INGEGNERIA
Corso di studi
INGEGNERIA DELLA AUTOMAZIONE
Relatori
relatore Prof. Pollini, Lorenzo
correlatore Prof. Abbink, David
Parole chiave
  • SIMULATOR EXPERIMENT
  • DRIVING SIMULATOR
  • HAPTICS
  • SHARED CONTROL
  • CURVE NEGOTIATION
Data inizio appello
14/12/2012
Consultabilità
Completa
Riassunto
A human operator performing a manual control task can be assisted by haptic shared control, a novel approach in literature which makes use of a continuous-time force feedback to guide the operator in a specific control direction.
In a previous research, a haptic controller has been designed and tested in a for curve negotiation support in a driving simulator. This support system provides a force feedback the operator has to give way to in order to correct the vehicle deviations from a reference trajectory. It was proved in an experiment to yield benefits in terms of increased performance and reduced effort from the operator with respect to manual driving.
As a variation from the haptic shared control philosophy, a novel approach has been introduced in literature for supporting a human operator piloting RPVs in a simulated environment. This haptic controller is called Indirect since the force feedback it provides has the only effect of changing the neutral point of the control interface. The operator can exploit this aiding by contrasting the force feedback and keeping the control device close to the central position. The Indirect haptic controller was proved in an experiment to increase performance and was found helpful by the pilots, as well as the Direct controller, which is the classic approach, and compared to manual piloting.
This novel approach in haptic shared control has only been investigated as a support for piloting RPVs. In this project the Indirect controller is being designed in a modeling study and tested for a curve negotiation task in a driving simulator.
The Indirect controller for curve negotiation is designed in a model-based study, where a scheme is employed simulating the behaviour of a human operator and the haptic controller. The outcome of this study are some previsions on different driving conditions, where an operator can drive manually or be assisted by the classic Direct haptic controller or by the novel Indirect controller.
An experimental campaign with a driving simulator is performed, based on the previsions figured in the modeling study. The results of the experiment tells us that both the Direct and Indirect controllers can be helpful for performance and effort of the operator in low visibility conditions, while in normal visibility a contrast in goal can arise between operator and shared controller, due to a different way of the operator to cut curves. In a situation of evasive maneuver with a faulty controller, where the operator has to reject the force feedback, the Indirect controller is proved to be more damaging than the Direct support. Finally, some negative after-effects on performance and effort are encountered after the Indirect controller is switched-off, due to an intense adaptation of the operator to the different dynamics to control.
The Indirect haptic support can be a valid alternative to the typical Direct control scheme for curve negotiation support, although there is room for improving its functioning. Future developments should be focused on making the Indirect controller easier to understand, to cope with more effectively in case of failures and possibly to switch to manual driving without confusion.
File