Modelling driver behaviour with Artificial Neural Networks
(2019) In Bachelor's theses in Mathematical Sciences MATK11 20191Mathematics (Faculty of Engineering)
- Abstract
- The goal of this thesis is to create a model that can simulate traffic on a highway by primarily using a supervised artificial neural network. The reader is first introduced to the theory of artificial neural networks, step by step from the fundamental principles to the network that is used in the final model. The network is trained to predict velocities of individual vehicles when being fed information about the vehicles lane and the relative position of nearby vehicles. The model is trained on a dataset recorded at U.S Highway 101 (Hollywood Freeway). It is validated by comparing properties of the simulated highway against properties of the real highway. It ends up being successful in simulating some properties of the highway such as lane... (More)
- The goal of this thesis is to create a model that can simulate traffic on a highway by primarily using a supervised artificial neural network. The reader is first introduced to the theory of artificial neural networks, step by step from the fundamental principles to the network that is used in the final model. The network is trained to predict velocities of individual vehicles when being fed information about the vehicles lane and the relative position of nearby vehicles. The model is trained on a dataset recorded at U.S Highway 101 (Hollywood Freeway). It is validated by comparing properties of the simulated highway against properties of the real highway. It ends up being successful in simulating some properties of the highway such as lane changes, flow and making sure that vehicles are not colliding. However, the model fails at capturing variations in the compared group speed properties Time Mean Speed (TMS) and Space Mean Speed (SMS). The focus of the discussion is to the summarize the findings and use them to clarify strengths and weaknesses of using local information to move cars when simulating a highway. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/8995686
- author
- Svenningsson, Christoffer LU
- supervisor
- organization
- alternative title
- Modellering av trafikbeteende med artificiella neuronnätverk
- course
- MATK11 20191
- year
- 2019
- type
- M2 - Bachelor Degree
- subject
- keywords
- Artificial neural networks, Traffic, Traffic modelling, Mathematical modeling, Traffic Simulation, Simulation, SMS, TMS, Traffic flow
- publication/series
- Bachelor's theses in Mathematical Sciences
- report number
- LUNFMA-4087-2019
- ISSN
- 1654-6229
- other publication id
- 2019:K9
- language
- English
- id
- 8995686
- date added to LUP
- 2024-09-30 14:33:23
- date last changed
- 2024-09-30 14:33:23
@misc{8995686, abstract = {{The goal of this thesis is to create a model that can simulate traffic on a highway by primarily using a supervised artificial neural network. The reader is first introduced to the theory of artificial neural networks, step by step from the fundamental principles to the network that is used in the final model. The network is trained to predict velocities of individual vehicles when being fed information about the vehicles lane and the relative position of nearby vehicles. The model is trained on a dataset recorded at U.S Highway 101 (Hollywood Freeway). It is validated by comparing properties of the simulated highway against properties of the real highway. It ends up being successful in simulating some properties of the highway such as lane changes, flow and making sure that vehicles are not colliding. However, the model fails at capturing variations in the compared group speed properties Time Mean Speed (TMS) and Space Mean Speed (SMS). The focus of the discussion is to the summarize the findings and use them to clarify strengths and weaknesses of using local information to move cars when simulating a highway.}}, author = {{Svenningsson, Christoffer}}, issn = {{1654-6229}}, language = {{eng}}, note = {{Student Paper}}, series = {{Bachelor's theses in Mathematical Sciences}}, title = {{Modelling driver behaviour with Artificial Neural Networks}}, year = {{2019}}, }