Average Speed Prediction Using Artificial Neural Networks
Total Page:16
File Type:pdf, Size:1020Kb
Average speed prediction using Artificial Neural Networks Cynthia Jacobs Delft University of Technology Faculty of Information Technology and Systems August 2003 Graduation Committee: drs. dr. L.J.M.Rothkrantz prof. dr. ir. E.J.H. Kerckhoffs prof. dr. H. Koppelaar Jacobs, Cynthia Master thesis, August 2003 Average speed prediction using Artificial Neural Networks Delft University of Technology, The Netherlands Faculty of Information Technology and Systems Data and Knowledge Engineering Group Acknowledgements First of all, I would like to thank Leon Rothkrantz for his feedback, comments and advice during the project. I have really enjoyed interesting discussions about travel time and average speed prediction and many other non-relating topics. Furthermore I would like to express my gratitude towards Pascal Wiggers for the conversations leading to new ideas, the constructive criticism on my report and his friendship. I also want to thank my friends Frans Flippo, Michiel Konstapel and Otto Visser for making most of my time at the university very enjoyable. And at last I would like to thank my parents, my sister and especially Lars van Leeuwen for their moral support and more. Cynthia Jacobs Delft, August 2003 Abstract Nowadays cars have navigation systems guiding the traveler from departure to destination. The navigation system uses route planners to determine the best route for the traveler. However these route planners do not use dynamic traffic information to decide what route is best. These route planners use the maximum speed allowed on a road to determine the travel time. So it is possible that the traveler is guided over congested routes while another route will lead the traveler faster to its destination. In this thesis at first it is explored what influences the average speed on the road and it is tried to predict this average speed using a feed forward neural network. If this average speed is known the real travel time can be determined for parts of the road and the best route can be discovered using this travel time. Analyzing the data showed that time, day of the week, month, weather, events, holidays and special events like accidents are factors that influence the average speed on the road. These influencing factors are used to predict the average speed on a specific point on a road. Therefore three models are developed. Two of these models are tested using JavaNNS developed at the Wilhelm-Schickard-Institute for Computer Science in Tübingen, Germany. The research on the architecture of neural networks and tests showed that a feed forward neural network with one hidden layer, from which the number of hidden neurons is determined by the formula of Fletcher and Goss (between 2Ni+1 and 2√Ni +No), is capable of predicting the average speed using only the time, month, day of the week and the average speed of 40 minutes before. At least 76% of the predictions have a difference smaller or equal to 10% if the prediction is made 40 minutes ahead. From these results it can be concluded that Artificial Neural Networks are capable of making good approximations for problems like predicting the average speed. Table of contents 1. Introduction.................................................................................................................. 11 1.1 Problem setting.................................................................................................... 11 1.2 Goals ..................................................................................................................... 14 1.2.1 Research on traffic .......................................................................................... 14 1.2.2 Research on Artificial Neural Networks ..................................................... 15 1.2.3 Research on travel time prediction............................................................... 15 1.3 Structure ............................................................................................................... 16 2. Literature survey.......................................................................................................... 17 2.1 Introduction ......................................................................................................... 17 2.2 Literature.............................................................................................................. 17 2.2.1 Literature list ................................................................................................... 18 2.2.2 Categorization ................................................................................................. 18 2.2.3 Summaries ....................................................................................................... 19 2.3 Conclusion............................................................................................................ 28 3. Possible influences and their data description ........................................................ 31 3.1 The date and time................................................................................................ 32 3.2 The day of the week............................................................................................ 32 3.3 The weather.......................................................................................................... 33 3.3.1 The past weather............................................................................................. 33 3.3.2 The weather forecast ...................................................................................... 36 3.3.3 The current weather ....................................................................................... 38 3.4 The events ............................................................................................................ 38 3.5 The holidays......................................................................................................... 43 3.6 Traffic measurement along the road ................................................................ 45 4. Data analysis................................................................................................................. 49 4.1 Time ...................................................................................................................... 51 4.2 Days of the week ................................................................................................. 51 4.3 Date ....................................................................................................................... 57 4.4 Weather................................................................................................................. 67 4.5 Special events....................................................................................................... 71 4.6 Conclusions.......................................................................................................... 74 5. Neural networks .......................................................................................................... 75 5.1 Introduction to neural networks....................................................................... 75 5.2 Training of neural networks.............................................................................. 77 5.2.1 The backpropagation algorithm ................................................................... 78 5.2.2 The parameters of the backpropagation algorithm ................................... 79 5.3 Why using a neural network? ........................................................................... 79 5.4 The Architecture of a Feedforward Neural Network .................................... 81 5.4.1 How many layers are necessary? ................................................................. 81 5.4.2 How many neurons must a layer contain? ................................................. 82 5.4.3 Conclusion ....................................................................................................... 83 6. Stuttgart Neural Network Simulator ........................................................................ 85 6.1 Introduction ......................................................................................................... 85 6.2 The most important functions of JavaNNS ..................................................... 85 6.2.1 View and creating a network ........................................................................ 86 6.2.2 Training a network ......................................................................................... 87 6.2.3 Analyzing a network...................................................................................... 88 7. Data preparation.......................................................................................................... 91 7.1 The parameters.................................................................................................... 91 7.1.1 Normalization ................................................................................................. 92 7.2 Pattern file ............................................................................................................ 92 7.2.1 Constructing the pattern file ......................................................................... 93 7.3 The result file ....................................................................................................... 94 8. Model............................................................................................................................