Abstract
Multi-agent based traffic simulation models have become increasingly important in simulating, studying and analysing traffic behaviours due to their ability to model more sophisticated behaviours of traffic by codifying simple rules into agents. However, such models require selection of appropriate rules and tuning of parameters for the selected rules. This process demands extensive resources if to be done manually. Further, high complexity models (in terms of number of rules and parameters) require a large computational cost to run, imposing scalability problems. In this work, four simple rules are introduced by reformulating existing concepts in the literature in order to simulate the self-organising behaviour of traffic where two lanes form into one and when two types of vehicles (cars and trucks) are present. The optimal rule and parameter combinations are explored via an evolutionary framework to overcome the resource demanding nature of the process. Two forms of objective functions - 1) a macroscopic objective function which focuses on macroscopic properties of traffic 2) a machine learning system trained based on human judgement concerning microscopic interaction of traffic - are studied in order to evolve low complexity and high fidelity traffic simulations. The differences in the rule sets evolved by the two objective functions are discussed highlighting the importance of selecting an appropriate objective function based on the simulation requirements and available resources. Finally, the change of the rule distribution as a function of generation in the evolutionary process is investigated in order to understand the complexity change in the simulations as a function of rule count as simulations are evolving towards high fidelity. This provides an abstract understanding of the relationship between complexity and fidelity in multi-agent based simulations concerning the particular problem of simulation of lane merge traffic.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
When the lane merging happens, the vehicles change their speed according to the situation based on the rules they are codified with.
The speed is reduced with a parameter proportional to reach 0 velocity, as that is the worst case that could happen. This does not mean that the vehicles always reduce the speed to 0. The changes of speed (& velocity) depend on the interactions between the rules, therefore when the rules are correctly evolved the vehicles do not reduce their speed to 0 unless there is chance of a collision.
Note that different number of rules were not drawn from a uniform distribution. Whether a particular gene is on or off was decided based on a uniform distribution considering each position which in turn result in different number of rules. If all the rules are off, a randomly decided rule was activated.
References
Aghabayk K, Sarvi M, Young W (2012) Understanding the dynamics of heavy vehicle interactions in car-following. J Transp Eng 138(12):1468–1475
Bar-Gera H, Ahn S (2010) Empirical macroscopic evaluation of freeway merge-ratios. Transportation Research Part C: Emerging Technologies 18(4):457–470
Benjamin SC, Johnson NF, Hui P (1996) Cellular automata models of traffic flow along a highway containing a junction. J Phys A Math Gen 29(12):3119
Brackstone M, McDonald M (1999) Car-following: a historical review. Transport Res F: Traffic Psychol Behav 2(4):181–196
Clarridge A, Salomaa K (2010) Analysis of a cellular automaton model for car traffic with a slow-to-stop rule. Theor Comput Sci 411(38):3507–3515
Eiben AE, Smith JE (2003) Introduction to Evolutionary Computing. Springer Verlag
Gazis DC, Herman R, Rothery RW (1961) Nonlinear follow-the-leader models of traffic flow. Oper Res 9(4):545–567
Gipps PG (1981) A behavioural car-following model for computer simulation. Transp Res B Methodol 15(2):105–111
Han YS, Ko SK (2012) Analysis of a cellular automaton model for car traffic with a junction. Theor Comput Sci 450:54–67
Hidas P (2002) Modelling lane changing and merging in microscopic traffic simulation. Transportation Research Part C: Emerging Technologies 10(5):351–371
Hidas P (2005) Modelling vehicle interactions in microscopic simulation of merging and weaving. Transportation Research Part C: Emerging Technologies 13(1):37–62
Kesting A, Treiber M, Helbing D (2007) General lane-changing model MOBIL for car-following models.Transportation Research Record. J Transp Res Board 1999(1):86—94
Lakshika E (2014) An Exploration of the omplexity-fidelity trade-off in multi-agent systems via an evolutionary framework. Phd thesis, Engineering & Information Technology,University of New South Wale,Canberra,Australia
Lakshika E, Barlow M, Easton A (2012) Fidelity and complexity of standing group conversation simulations: A framework for the evolution of multi agent systems through bootstrapping human aesthetic judgments. In: Proceedings of 2012 IEEE Congress on Evolutionary Computation (CEC), pp 1044–1051
Lakshika E, Barlow M (2013) Co-evolving semi-competitive interactions of sheepdog herding behaviors utilizing a simple rule-based multi agent framework. In: Proceedings of 2013 IEEE Symposium on Artificial Life(ALIFE),IEEE, pp 82–89
Lv W, Song Wg, Fang Zm, Ma J (2013) Modelling of lane-changing behaviour integrating with merging effect before a city road bottleneck. Physica A: Statistical Mechanics and its Applications 392(20):5143–5153
Nagel K, Schreckenberg M (1992) A cellular automaton model for freeway traffic. J Phys I 2(12):2221–2229
Ni D, Leonard JD II (2005) A simplified kinematic wave model at a merge bottleneck. Appl Math Model 29(11):1054–1072
Panwai S, Dia H (2007) Neural agent car-following models. IEEE Trans Intell Transp Syst 8(1):60–70
Reina P, Ahn S (2014) Prediction of merge ratio using lane flow distribution. In: Transportation Research Board 93rd Annual Meeting, 14-5459
Rickert M, Nagel K, Schreckenberg M, Latour A (1996) Two lane traffic simulations using cellular automata. Physica A: Statistical Mechanics and its Applications 231(4):534–550
Treiber M, Hennecke A, Helbing D (2000) Congested traffic states in empirical observations and microscopic simulations. Phys Rev E 62(2):1805
Wagner P, Nagel K, Wolf DE (1997) Realistic multi-lane traffic rules for cellular automata. Physica A: Statistical Mechanics and its Applications 234(3):687–698
Witten IH, Frank E (2005) Data Mining: Practical Machine Learning Tools and Techniques. In: Morgan Kaufmann Series in Data Management Systems. 2nd. Morgan Kaufmann Publishers Inc, San Francisco
Wu J, Brackstone M, McDonald M (2000) Fuzzy sets and systems for a motorway microscopic simulation model. Fuzzy Set Syst 116(1):65—76
Acknowledgments
The authors would like to acknowledge the financial support provided by the UNSW Canberra paper writing fellowship in completing the first draft of this publication.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Lakshika, E., Barlow, M. & Easton, A. Evolving lane merge traffic behaviour simulations via a macroscopic objective function and a machine learning system trained through bootstrapped human judgement. Appl Intell 44, 862–877 (2016). https://doi.org/10.1007/s10489-015-0733-3
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10489-015-0733-3