Many human social phenomena, such as cooperation1,2,3, the growth of settlements4, traffic dynamics5,6,7 and pedestrian movement7,8,9,10, appear to be accessible to mathematical descriptions that invoke self-organization11,12. Here we develop a model of pedestrian motion to explore the evolution of trails in urban green spaces such as parks. Our aim is to address such questions as what the topological structures of these trail systems are13, and whether optimal path systems can be predicted for urban planning. We use an ‘active walker’ model14,15,16,17,18,19 that takes into account pedestrian motion and orientation and the concomitant feedbacks with the surrounding environment. Such models have previously been applied to the study of complex structure formation in physical14,15,16, chemical17 and biological18,19 systems. We find that our model is able to reproduce many of the observed large-scale spatial features of trail systems.
Previous studies have shown that various observed self-organization phenomena in pedestrian crowds can be simulated very realistically. These phenomena include the emergence of lanes of uniform walking direction and oscillatory changes of the passing direction at bottlenecks7,10. Another interesting collective effect of pedestrian motion, which we have investigated very recently, is the formation of trail systems in green areas. In many cases, the pedestrians' desire to take the shortest way and the specific properties of the terrain are insufficient for an explanation of the trail characteristics. It is essential to include the effect of human orientation. To simulate the typical features of trail systems, we have extended the earlier model of pedestrian motion to an active-walker model by introducing equations for environmental changes and their effect on the chosen walking direction.
First, we represent the ground structure at place r and time t by a function G(r,t) which reflects the comfort of walking. Trails are characterized by particularly large values of G. On the one hand, at their positions r = rα(t), all pedestrians α leave footprints on the ground (for example, by trampling down some vegetation). Their intensity is assumed to be I(r)[1 − G(r,t)/Gmax(r)], as the clarity of a trail is limited to a maximum value Gmax(r). This causes a saturation effect [1 − G(r,t)/Gmax(r)] of the ground's alteration by new footprints. On the other hand, the ground structure changes owing to the regeneration of vegetation. This will lead to a restoration of the natural ground conditions G0(r) with a certain weathering rate 1/T(r) which is related to the durability T(r) of trails. Thus, the equation of environmental changes reads:
where δ(r − rα) denotes Dirac's delta function (which yields only a contribution for r = rα).
The attractiveness of a trail segment at place r from the perspective of place rα decreases with its distance |r − rα(t)| and depends on the visibility σ(rα). Taking this into account by using a factor exp(−|r − rα|/σ(rα)) and taking the spatial average by integration of the weighted ground structure over the green area, we obtain:
The trail potential Vtr(rα,t) reflects the attractiveness of walking at place rα. It describes indirect long-range interactions via environmental changes, which are essential for the characteristics of the evolving patterns18.
On a plain, homogenous ground, the walking direction eα of pedestrian α is determined by the direction of the next destination dα, that is eα(rα) = (dα − rα)/|dα − rα|. Without a destination, a pedestrian is expected to move into the direction of the largest increase of ground attraction, which is given by the (normalized) gradient ∇rαVtr(rα,t) of the trail potential. However, because the choice of the walking direction eα is influenced by the destination and existing trails at the same time, the orientation relation
was taken as the arithmetic average of both effects. Considering cases of rare interactions, the approximate equation of motion of a pedestrian α with desired velocity vα0is:
A comparison of simulation results with photographs shows that the above described model is in good agreement with empirical observations. In particular, the evolution of the unexpected ‘island’ in the middle of the trail system in Fig. 1 can be correctly described (Fig. 2). The goodness of fit of the model is quite surprising, as it contains only two independent parameters κ= IT/σ and λ= V0T/σ, where V0denotes the average of the desired velocities vα0. This can be shown by scaling the model to dimensionless equations. The parameter λ was kept constant.
Our simulations are based on a discretization of the considered area in small quadratic elements of equal size, which converts the integral in equation (2)into a sum. Temporal and spatial derivatives are approximated by difference quotients. The examples we present here begin with plain, homogenous ground. All pedestrians have their own destinations and entry points, from which they start at a randomly chosen point in time. In Fig. 2 (Fig. 3) pedestrians move between all possible pairs of three (four) fixed places; in Fig. 4, the entry points and destinations are distributed over the small ends of the ground.
At the beginning, pedestrians take the direct ways to their respective destinations. But after some time they begin to use already existing trails, as this is more comfortable than clearing new ways. By this, a kind of selection process16,20,21 between trails sets in: frequently used trails are more attractive than others. For this reason they are chosen very often, and the resulting reinforcement makes them even more attractive. However, the weathering effect destroys rarely used trails and limits the maximum length of the way system which can be supported by a certain rate of trail usage. As a consequence, the trails begin to bundle, especially where different trails meet or intersect. This explains why pedestrians with different destinations use and produce common parts of the trail system (Figs 2 and 3).
A direct way system (which provides the shortest connections, but covers a lot of space) only develops if all ways are almost equally comfortable. If the advantage κ of using existing trails is large, the final trail system is a minimal way system (which is the shortest way system that connects all entry points and destinations). For realistic values of κ, the evolution of the trail system stops before this state is reached (Fig. 2). Thus, κ is related to the average relative detour of the walkers. We conjecture (but cannot yet prove) that the resulting way system is the shortest one which is compatible with a certain accepted relative detour. In this sense, it yields an optimal compromise between convenience and shortness.
Therefore, we suggest the above model could be used by urban planners and landscape gardeners, who face the task of constructing the most comfortable and convenient way systems at minimal construction costs. For planning purposes, one needs to know the entry points and destinations within the considered area and the rates of usage of their connections. If necessary, these can be estimated by trip chaining models22, which are also needed in cases of complex lines of access and sight. The effects of the physical terrain and already existing ways can be taken into account by the function G0(r). By varying the model parameter κ, the overall length of the resulting trail system can be influenced (Figs 2 and 3). In the same way, one can check its structural stability. We are at present evaluating typical parameter values of κ and λ by comparison of simulation results with real pedestrian flows which are reconstructed from video films by image processing. These values could be used for designing convenient way systems in residential areas, parks and recreation areas by means of computer simulations (Fig. 3).
It would be interesting to relate our model to work on space syntax23. Repulsive interactions between pedestrians, which are relevant in cases of frequent pedestrian interactions, can be taken into account by generalizing equation (4)in accordance with the social force model of pedestrian motion7,10; in general, they lead to broader trails.
Axelrod, R. & Dion, D. The further evolution of cooperation. Science 242, 1385–1390 (1988).
Clearwater, S. H., Huberman, B. A. & Hogg, T. Cooperative solution of constraint satisfaction problems. Science 254, 1181–1183 (1991).
Helbing, D. Quantitative Sociodynamics Stochastic Methods and Models of Social Interaction Processes(Kluwer Academic, Dordrecht, (1995)).
Makse, H. A., Havlin, S. & Stanley, H. E. Modelling urban growth patterns. Nature 377, 608–612 (1995).
Herman, R., Lam, T. & Prigogine, I. Multilane vehicular traffic and adaptive human behavior. Science 179, 918–920 (1973).
Herman, R. & Prigogine, I. Atwo-fluid approach to town traffic. Science 204, 148–151 (1979).
Helbing, D. Verkehrsdynamik. Neue physikalische Modellierungskonzepte(Springer, Berlin, (1997)).
Henderson, L. F. The statistics of crowd fluids. Nature 229, 381–383 (1971).
Henderson, L. F. Sexual differences in human crowd motion. Nature 240, 353–355 (1972).
Helbing, D. & Molnár, P. Social force model for pedestrian dynamics. Phys. Rev. E 51, 4282–4286 (1995).
Haken, H. Advanced Synergetics 2nd edn(Springer, Berlin, (1987)).
Nicolis, G. & Prigogine, I. Self-Organization in Nonequilibrium Systems. From Dissipative Structures to Order through Fluctuations(Wiley, New York, (1977)).
Schenk, M. Untersuchungen zum Fußgängerverhalten Thesis, Univ. Stuttgart((1995)).
Kayser, D. R., Aberle, L. K., Pochy, R. D. & Lam, L. Active walker models: tracks and landscapes. Physica A 191, 17–24 (1992).
Lam, L. Active walker models for complex systems. Chaos Solitons Fractals 6, 267–285 (1995).
Schweitzer, F. & Schimansky-GGeier, L. Clustering of “active” walkers in a two-component system. Physica A 206, 359–379 (1994).
Schimansky-Geier, L., Mieth, M., Rosé, H. & Malchow, H. Structure formation by active Brownian particles. Phys. Lett. A 207, 140–146 (1995).
Ben-Jacob, E. et al. Generic modelling of cooperative growth patterns in bacterial colonies. Nature 368, 46–49 (1994).
Stevens, A. & Schweitzer, F. in Mechanisms of Cell and Tissue Motion(eds Alt, W., Deutsch, A. & Dunn, G.) 183–192 (Birkhäuser, Basel, (1997)).
Eigen, M. & Schuster, P.The Hypercycle(Springer, Berlin, (1979)).
Feistel, R. & Ebeling, W. Evolution of Complex Systems. Self-Organization, Entropy and Development(Kluwer, Dordrecht, (1989)).
Timmermans, H., van der Hagen, X. & Borgers, A. Transportation systems, retail environments and pedestrian trip chaining behaviour: Modelling issues and applications. Transportation Res. B 26, 45–59 (1992).
Hillier, B. Space is a Machine: A Configurational Theory of Architecture(Cambridge Univ. Press, (1996)).
We thank F. Schweitzer for discussions, and W. Weidlich and M. Treiber for reviews of the manuscript.
About this article
Cite this article
Helbing, D., Keltsch, J. & Molnár, P. Modelling the evolution of human trail systems. Nature 388, 47–50 (1997). https://doi.org/10.1038/40353
Structural and Multidisciplinary Optimization (2021)
Brazilian Journal of Physics (2021)
Large-Scale Dynamics of Self-propelled Particles Moving Through Obstacles: Model Derivation and Pattern Formation
Bulletin of Mathematical Biology (2020)
Automatic quality evaluation of parade by variance of postures of a platoon on single video camera logs
Artificial Life and Robotics (2020)
When and how obstacle size and the number of foragers affect clearing a foraging trail in leaf-cutting ants
Insectes Sociaux (2019)