The development of a hardware- and software-based simulation platform for the training of driver candidates

In this study, a traffic simulation system (TSS) to provide novice driver candidates with opportunities for training and testing has been developed. Autonomous vehicles that follow the flow of traffic in a microsimulation environment were prepared using a hierarchical concurrent state machine. Autonomous vehicles are skilled and have behavioral aspects such as tracking the lanes, following the vehicles, complying with traffic rules, decelerating, accelerating, and intersection crossings. These driver behaviors are parameterized to allow the autonomous agents to act while disregarding the rules (i.e. aggressive behaviors) or behaving normally. The TSS is operated in 3 different modes, namely the orientation, training, and testing modes. During the orientation phase, the candidate driver's target is to get used to the system and the vehicle, along with traffic rules in city conditions. During training, the driver is warned orally (when necessary) and given written handouts about the mistakes that occurred during the training session. The rules to be obeyed in traffic are kept in an XML file in the system and the data are used to follow the driver's behaviors. During the testing phase, an evaluation mechanism is employed. It observes whether the drivers obey or disregard the rules during the testing time interval. At the end of the test drive, a report showing the driver's mistakes is given. The reports and test results can be recorded in the database with the drivers' names for further analysis and evaluation processes. Driver candidates perform the driving using a steering wheel and pedals, along with a screen platform consisting of 3 monitors. With the TSS user interface, parameterized items such as vehicle selection, the crowdedness of the traffic, the timing of the traffic lights, and the determination of air conditions are possible. The system provides a cheap, affordable, and nonrisky platform for the trainees. Test results show that the system improves the drivers' skills and builds trust in them.

The development of a hardware- and software-based simulation platform for the training of driver candidates

In this study, a traffic simulation system (TSS) to provide novice driver candidates with opportunities for training and testing has been developed. Autonomous vehicles that follow the flow of traffic in a microsimulation environment were prepared using a hierarchical concurrent state machine. Autonomous vehicles are skilled and have behavioral aspects such as tracking the lanes, following the vehicles, complying with traffic rules, decelerating, accelerating, and intersection crossings. These driver behaviors are parameterized to allow the autonomous agents to act while disregarding the rules (i.e. aggressive behaviors) or behaving normally. The TSS is operated in 3 different modes, namely the orientation, training, and testing modes. During the orientation phase, the candidate driver's target is to get used to the system and the vehicle, along with traffic rules in city conditions. During training, the driver is warned orally (when necessary) and given written handouts about the mistakes that occurred during the training session. The rules to be obeyed in traffic are kept in an XML file in the system and the data are used to follow the driver's behaviors. During the testing phase, an evaluation mechanism is employed. It observes whether the drivers obey or disregard the rules during the testing time interval. At the end of the test drive, a report showing the driver's mistakes is given. The reports and test results can be recorded in the database with the drivers' names for further analysis and evaluation processes. Driver candidates perform the driving using a steering wheel and pedals, along with a screen platform consisting of 3 monitors. With the TSS user interface, parameterized items such as vehicle selection, the crowdedness of the traffic, the timing of the traffic lights, and the determination of air conditions are possible. The system provides a cheap, affordable, and nonrisky platform for the trainees. Test results show that the system improves the drivers' skills and builds trust in them.

___

  • [1] T. Hirata, Development of Driving Simulation System: MOVIC-T4 and its Application to Traffic Safety Analysis in Underground Urban Express Ways, PhD dissertation, Tokyo Institute of Technology, Department of Civil Engineering, Tokyo, pp. 2–3, 20–22, 157–160, 2005.
  • [2] S. Das, B.A. Bowles, C.R. Houghland, S.J. Hunn, Y. Zhang, “Microscopic simulations of freeway traffic flow”, Thirty-Second Annual Simulation Symposium, p. 79, 1999.
  • [3] Y. Miao, U. Hoppe, N. Pinkwart, “Naughty agents can be helpful: training drivers to handle dangerous situations in virtual reality”, Proceedings of the 6th IEEE Computer Society International Conference on Advanced Learning Technologies, pp. 735–739, 2006.
  • [4] N. S¨umer, “Trafik kazalarında sosyal psikolojik etmenler: s¨ur¨uc¨u davranı¸sları, becerileri ve sosyal politik ¸cevre”, T¨urk Psikoloji Yazıları, Vol. 5, pp. 1–36, 2002.
  • [5] Y. Yasak, “Trafik psikolojisi”, TS¸OF-Psikoteknik: S¨ur¨uc¨u De˘gerlendirme E˘gitim ve Ara¸stırma Merkezi, pp. 3–11, 2002.
  • [6] L. Evans, Traffic Safety, Bloomfield Hills, Michigan, Science Serving Society, 2004.
  • [7] M. Demir, Trafik E˘gitimi ˙ I¸cin Zeki Etmenler Geli¸stirme, PhD dissertation, Gazi University Institute of Information Technology, 2008.
  • [8] N. S¨umer, A.B. Unal, “Acemi s¨ ¨ ur¨uc¨ulerde tehlike algısıbecerisi”, U¸ ¨ c¨unc¨u Trafik ve Yol G¨uvenli˘gi Ulusal Kongresi, pp. 177–182, 2005.
  • [9] H. Ece, “Trafik kazaları ¨ozeti”, T.C. Bayındırlık ve ˙ Iskan Bakanlı˘gı Karayolları Genel M¨ud¨url¨u˘g¨u, pp. 3–5, 2003.
  • [10] N. S¨umer, A.B. Unal, A. Birdal, “Assessment of hazard perception latencies using real life and animated traffic ¨ hazards: comparison of novice and experienced drivers”, Proceedings of the 4th International Driving Symposium on Human Factors in Driver Assessment, Training and Vehicle Design, pp. 488–494, 2007.
  • [11] T. Sun, J.F. Wang, “A traffic cellular automata model based on road network grids and its spatial and temporal resolution’s influences on simulation”, Simulation Modelling Practice and Theory, Vol. 15, pp. 864–878, 2007.
  • [12] J. Cremer, J. Kearney, Y.E. Papelis, R.A. Romano, “The software architecture for scenario control in Iowa driving simulator”, Proceedings of the 4th Computer Generated Forces and Behavioral Representation Conference, pp. 523–532, 1993.
  • [13] A. Okutano˘glu, V. ˙ I¸sler, “Sim¨ulasyon sistemlerinde esnek senaryo altyapıları”, Birinci Ulusal Savunma UygulamalarıModelleme ve Sim¨ulasyon Konferansı, pp. 115–126, 2005.
  • [14] J. Archer, I. Kosonen, “The potential of micro-simulation modelling in relation to traffic safety assessment”, Proceedings of the ESS Conference, pp. 427–443, 2000.
  • [15] A. Laschet, “Computer simulation of vibrations in vehicle powertrains considering nonlinear effects in clutches and manual transmissions”, Society of Automotive Engineers Technical Paper 941011, 1994.
  • [16] Delta3D Open Source Gaming & Simulation Engine, September 2010, http://www.delta3d.org/.
  • [17] OpenSceneGraph, September 2010, http://www.openscenegraph.org/
  • [18] Open Dynamics Engine, September 2010, http://www.ode.org/.
  • [19] Character Animation Library, September 2010, http://gna.org/projects/cal3d/.
  • [20] OpenAL, September 2010, http://connect.creativelabs.com/openal/default.aspx.
  • [21] Portable Game Library, February 2011, http://plib.sourceforge.net/.
  • [22] KarayollarıTrafik Y¨onetmeli˘gi, 18.07.1997, http://www.traport.com.tr/doc/trafikyonetmeligi.pdf.
  • [23] A. C¸ avu¸so˘glu, ˙ I. Kurnaz, “Developing a parameterized simulation platform with intelligent synthetic agents for training driver candidates”, Procedia Computer Science, Vol. 3, pp. 727–731, 2011.
  • [24] H.H. G¨okta¸s, A. C¸ avu¸so˘glu, B. S¸en, “AUTOCITY: A system for generating 3D virtual cities for simulation systems on GIS map”, Intelligent Automation and Soft Computing, Vol. 15, pp. 29–39, 2009.
  • [25] G. Chen, S. Xiang, G.Q. Ji, Q. Ding, “Scene simulation platform based on data fusion of multiple format 3D models”, IEEE International Conference on Computer Modeling and Simulation, pp. 342–346, 2009.