Search

Browse

• Communities
• Authors
• Year
• Titles
• Subjects

My Space

• Login

Analysis of cruise control systems for autonomous vehicles on lane-free roads with sampled-data and saturated control inputs

Tzortzoglou Filippos

Full record

---

Summary

Vehicle automation has made tremendous advances in recent decades and the path to full automation of vehicles seems inevitable. In this dissertation, cruise control systems for autonomous vehicles on lane-free roads were studied using the “nudging” effect that depicts the ability of the vehicles to exert a virtual force to the front ones. The model depicting the motion of such vehicles is described by a large number of nonlinear continuous-time differential equations. The application of such models in real systems requires conversion of the continuous-time system into a discrete-time system with the additional feature that the values of the input signals are sampled. The modeling and conversion of the continuous-time system into a sampled-data control system was studied in this diploma thesis, first. Next, appropriate design of controllers with discrete sampling were considered. The design of controllers for discrete sampling systems is performed using the simulation approach in which we first design a controller in the continuous-time domain and then discretize the controller and apply it digitally. Since the digital controller approach can only be successful if the sampling period is sufficiently short, the selection of the sampling period that guarantees the stability of the system to the proposed controllers by discrete sampling was examined through simulations. At the same time, using appropriate proven Lemmas, it was examined the occasion when the sampling period changes non- periodically. Then a study was made through simulations of the discrete-time model when there are saturation effects on the input of the system. Finally, an algorithm was examined that creates realistic initial positions in order to avoid high initial energies which can lead to a collision course.

Available Files

Services