An Automatic Race Track Generating System
In this paper, we propose an automatic race track generating system based on difficulty evaluation and feature turns detection for providing users skill-matched contents. Given a start point, a goal point, and a difficulty expectation chart, our system ranks all candidate race tracks according to the similarity with respect to the given difficulty curve. Then, user can choose a satisfied track and export it into a racing car simulator to play.
The system automatically creates the racing line for the input race track. Then, the line is used to segment turns in the race track, and the corresponding ideal maximum speed variation is exploited to evaluate the difficulty by our proposed Turnscore formula. Also, the corresponding curvature chart of the racing line is encoded as a string and the characterized regular expression for feature turns is being matched in the string for identifying feature turns.
As the experimental results show, the feature turns detection is of high accuracy and the difficulty evaluation is reliable so that our system is effective to provide skill-matched race tracks for users.
KeywordsDifficulty Evaluation Race Track Generation Racing Line String Searching Procedural Content Generation Feature Detection
Unable to display preview. Download preview PDF.
- 1.Open Street Map (June 2001), http://www.openstreetmap.org/
- 2.Allianz Sponsoring Media Center (June 2005), http://sponsoring.allianz.com/en/
- 3.Wymann, B.: TORCS Robot Tutorial (February 24, 2005), http://www.berniw.org/
- 4.F1 Tracks for TORCS (June 2011), http://apr-free.info/joomla/
- 5.Google Map (June 2011), http://maps.google.com.tw/
- 6.Hannan, J.: Interview to jeff hannan (2001), http://www.generation5.org/content/2001/hannan.asp
- 7.Lecchi, S.: Artificial intelligence in racing games. In: 2009 IEEE Symposium on Computational Intelligence and Games, CIG 2009, p.1 (2009)Google Scholar
- 8.Riedmiller, M., Braun, H.: A direct adaptive method for faster backpropagation learning: the RPROP algorithm. In: 1993 IEEE International Conference on Neural Networks, vol. 1, pp. 586–591 (1993)Google Scholar
- 10.Schiffmann, W., Joost, M., Werner, R.: Optimization of the Backpropagation Algorithm for Training Multilayer Perceptrons (1994)Google Scholar
- 11.Stern, D., Candela, J.Q., Herbrich, R., Graepel, T.: Playing machines: Machine learning applications in computer games. Microsoft Research Cambridge (2008)Google Scholar
- 12.TORCS Team. TORCS (The Open Racing Car Simulator) Official Site. Latest Version: TORCS 1.3.1 (May 2, 2010), http://torcs.sourceforge.net/
- 13.Coulom, R.: Reinforcement Learning Using Neural Networks, with Applications to Motor Control. PhD thesis, Institut National Polytechnique de Grenoble (2002)Google Scholar
- 15.Cardamone, L., Loiacono, D., Lanzi, P.L., Bardelli, A.P.: Searching for the Optimal Racing Line Using Genetic Algorithms. In: 2010 IEEE Symposium on Computational Intelligence and Games (CIG), pp. 388–394 (August 2010)Google Scholar
- 17.Gran Turismo 5 (2010), http://en.wikipedia.org/wiki/Gran_Turismo_5
- 19.Galin, E., Peytavie, A., Marechal, N., Guerin, E.: Procedural Generation of Roads. Proceedings of Eurographics 29(2), 429–438 (2010)Google Scholar
- 20.The Official Formula One Website, http://www.formula1.com/
- 21.The Official APIGA Website, http://www.apiga.com.tw/
- 22.Soper, H.E., Young, A.W., Cave, B.M., Lee, A., Pearson, K.: On the distribution of the correlation coefficient in small samples. Appendix II to the papers of “Student” and R. A. Fisher. Biometrika 11(4), 328–413 (1917)Google Scholar