dc.rights.license | CC-BY-NC-ND | |
dc.contributor.advisor | Dastani, Dr. Mehdi | |
dc.contributor.advisor | Meyer, Prof. Dr. John-Jules | |
dc.contributor.advisor | van Luipen, Ir. Jelle | |
dc.contributor.author | Tielman, W.L. | |
dc.date.accessioned | 2015-02-17T18:01:31Z | |
dc.date.available | 2015-02-17T18:01:31Z | |
dc.date.issued | 2015 | |
dc.identifier.uri | https://studenttheses.uu.nl/handle/20.500.12932/19402 | |
dc.description.abstract | Railway simulations have been a useful tool within the railway industry. One of the simulators that ProRail, the Dutch infrastructure manager for the railways, uses is the micro-railway simulator FRISO. ProRail wants to find out if the validity of this simulator could be improved through adding agent based train drivers. In this thesis the development of these agents will be described. Different data sources about train driver behaviour were available and could be used to create a train driver model that could be implemented within the designed agents. Using this, an agent DLL was written in C++ to work together with FRISO. Simulations were then done in order to find out if the validity had improved with the added agents. Through comparing the resulting driving times with the previous train driver implementation and realisation data, it was concluded that the agents scored better. When looking at the driving behaviour of the agents, it was concluded that this lied closer to the realisation data then that of the FRISO train drivers. It was also noted that certain aspects of train driver behaviour were not modelled correctly by the agents and/or FRISO which resulted in deviations seen in driving times and driving behaviour. The presence of these aspects indicated that a sufficiently accurate model of train driver behaviour is required if reliable simulation results are desired. | |
dc.description.sponsorship | Utrecht University | |
dc.format.extent | 5305452 | |
dc.format.mimetype | application/pdf | |
dc.language.iso | en | |
dc.title | An Agent-based Approach to Simulating Train Driver Behaviour | |
dc.type.content | Master Thesis | |
dc.rights.accessrights | Open Access | |
dc.subject.keywords | agents, simulation, railway, modeling human operator, artificial Intelligence, train driver behaviour, machine learning | |
dc.subject.courseuu | Technical Artificial Intelligence | |