Prediction and predictive control for economic optimisation of vehicle operation

Kock, Peter (2013) Prediction and predictive control for economic optimisation of vehicle operation. (PhD thesis), Kingston University.

Full text not available from this archive.

Abstract

Truck manufacturers are currently under pressure to reduce pollution and cost of transportation. The cost efficient way to reduce CO[sub]2 and cost is to reduce fuel consumption by adaptation of the vehicle speed to the driving conditions - by heuristic knowledge or mathematical optimisation. Due to their experience, professional drivers are capable of driving with great efficiency in terms of fuel consumption. The key research question addressed in this work is the comparison of the fuel efficiency for an unassisted drive by an experienced professional driver versus an enhanced drive using driver assistance system. The motivation for this is based on the advantage of such a system in terms of price (lower than driver's training) but potentially it can be challenging to obtain drivers' acceptance of the system. There is a range of fundamental issued that have to be addressed prior to the design and implementation of the driver assistance system. The first issue is related to the evaluation of the correctness of the prediction model under development, due to a range of inaccuracies introduced by slope errors in digital maps, imprecise modelling of combustion engine, vehicle physics etc. The second issue is related to the challenge in selecting a suitable method for optimisation of mixed integer non-linear systems. Dynamic Programming proved to be very suitable for this work and some methods of search space reduction are presented here. Also an analytical solution of the Bernoulli differential equation of the vehicle dynamics is presented and used here in order to reduce computing effort. Extensive simulation and driving tests were performed using different driving approaches to compare well trained human experts with a range of different driving assistance systems based on standard cruise control, heuristic and mathematical optimisation. Finally the acceptance of the systems by drivers been evaluated.

Item Type: Thesis (PhD)
Physical Location: This item is held in stock at Kingston University library.
Research Area: Mechanical, aeronautical and manufacturing engineering
Faculty, School or Research Centre: Faculty of Science, Engineering and Computing
Depositing User: Jennifer May
Date Deposited: 26 Aug 2016 14:25
Last Modified: 26 Aug 2016 14:25
URI: http://eprints.kingston.ac.uk/id/eprint/35861

Actions (Repository Editors)

Item Control Page Item Control Page