University of Surrey

Test tubes in the lab Research in the ATI Dance Research

Autonomous high-speed overtaking in structured environments

Dixit, Shilp (2019) Autonomous high-speed overtaking in structured environments Doctoral thesis, University of Surrey.

[img]
Preview
Text
Dixit - 2019 - PhD Thesis - Autonomous High-Speed Overtaking in Structured Environments.pdf - Version of Record
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (7MB) | Preview

Abstract

In this thesis, we design and develop controllers for trajectory planning and trajectory tracking to tackle autonomous high-speed overtaking for the next generation of autonomous vehicles. Both the controllers are developed for a JLR Range Rover Sport that is capable of autonomous driving functionalities. To assist with controller development, a high-fidelity vehicle model previously developed in IPG Carmaker is utilised that contains all the multi-body interactions and non-linear tyre characteristics. Trajectory Planning Autonomous high-speed driving is a safety-critical task and it is imperative that the planned trajectory of the vehicle can ensure safety (collision-avoidance) while computing smooth and feasible trajectories. We propose a trajectory planning framework that utilises information of the traffic vehicles to identify safe driving zones on the road using potential field functions and a robust model predictive controller for generating feasible trajectories that ensure the vehicle remains within the safe zones while performing the overtaking manoeuvre. The closed-loop performance of this controller is validated in a high-fidelity co-simulation environment. Trajectory Tracking The trajectory tracking controller is designed to ensure that the vehicle tracks the trajectory as closely as possible and preserves the lateral-yaw stability at all times. In this thesis, an Enhanced Model Reference Adaptive Control algorithm is used to design a generic lateral tracking controller for an autonomous vehicle. The control algorithm is applied to a vehicle path tracking problem and its tracking performance is investigated when subjected to external disturbances such as crosswind, road surface changes, modelling errors, and parameter miss-matches in a high-fidelity co-simulation environment. Combined Planning & Control Finally, the design of a combined motion planning & control scheme is carried out. The lateral tracking controller is augmented to include the dynamics of the steering actuator system and the updated tracking controller is combined with the RMPC based sophisticated path-planning framework to present a hierarchical closed-loop control architecture for autonomous overtaking. This architecture is implemented on the IPG CarMaker/Simulink environment and validated with different overtaking manoeuvring scenarios.

Item Type: Thesis (Doctoral)
Divisions : Theses
Authors :
NameEmailORCID
Dixit, Shilp0000-0001-9378-442X
Date : 3 October 2019
Funders : Jaguar Land Rover, Engineering and Physical Sciences Research Council (EPSRC)
DOI : 10.15126/thesis.00853312
Grant Title : Towards Autonomy: Smart and Connected Control (TASCC)
Contributors :
ContributionNameEmailORCID
http://www.loc.gov/loc.terms/relators/THSFallah, Sabers.fallah@surrey.ac.uk
http://www.loc.gov/loc.terms/relators/THSMontanaro, Umbertou.montanaro@surrey.ac.uk
Related URLs :
Depositing User : Shilp Dixit
Date Deposited : 07 Feb 2020 14:49
Last Modified : 07 Feb 2020 14:52
URI: http://epubs.surrey.ac.uk/id/eprint/853312

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year


Information about this web site

© The University of Surrey, Guildford, Surrey, GU2 7XH, United Kingdom.
+44 (0)1483 300800