Year: 2019
Type: dissertation
Abstract: This thesis discusses two different problems in motion planning for autonomous driving. \nThe first is the problem of optimizing a lattice planner control set for any particular \nautonomous driving task, with the goal of reducing planning time for that task. The \ndriving task is encoded in the form of a dataset of trajectories executed while performing \nsaid task. In addition to improving planning time, the optimized control set should capture \nthe driving style of the dataset. In this sense, the control set is learned from the data and is \ntailored to a particular task. To determine the value of control actions to add to the control \nset, a modified version of the Fréchet distance is used to score how useful control actions \nare for generating paths similar to those in the dataset. This method is then compared to \nthe state of the art lattice planner control set optimization technique in terms of planning \nruntime for the learned task. \nThe second problem is the task of extending the Responsibility-Sensitive Safety (RSS) \nframework by introducing swerve manoeuvres in addition to the nominal braking manoeu- \nvres present in the framework. This includes comparing the clearance distances required by \na swerve to the braking distances in the original framework. This comparison shows that \nswerve manoeuvres require less distance gap in order to reach safety from a braking agent \nin front of the autonomous vehicle at higher speeds. For more realistic swerve manoeuvres, \nthe kinematic bicycle model is used rather than the 2-D double integrator model consid- \nered in RSS. An upper bound is then computed on the required clearance distance for a \nswerve manoeuvre that satisfies bicycle kinematics. A longitudinal safe following distance \nis then derived that is provably safe, and is shown to be lower than the following distance \nrequired by RSS at higher speeds. The use of the kinematic bicycle model is then validated \nby computing swerve manoeuvres with a dynamic single-track car model and Pacejka tire \nmodel, and comparing the single-track swerves to the bicycle swerves.
Source: UWSpace (University of Waterloo)
Author: Ryan De Iaco
Language: English
Subfield: Automotive Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W3011878662