Zombie Outbreak in a City

This model showcases an ABM running on a map, using OpenStreetMapSpace.

Constructing the end of days

using Agents
using Random

We'll simulate a zombie outbreak in a city. To do so, we start with an agent which satisfies the OSMSpace conditions of having a position of type Tuple{Int,Int,Float64}. For simplicity though we shall build this with the @agent macro.

@agent Zombie OSMAgent begin

To be explicit, this macro builds the following type:

mutable struct Zombie <: AbstractAgent

where a tuple (i, j, x)::Tuple{Int,Int,Float64} means a position on the road between nodes i, j of the map, having progressed x distance along the road.

The model constructor we build consists of a map, and 100 agents scattered randomly around it. They have their own agenda and need to travel to some new destination. Unfortunately one of the population has turned and will begin infecting anyone who comes close.

function initialise(; seed = 1234)
    map_path = OSM.test_map()
    properties = Dict(:dt => 1 / 60)
    model = ABM(
        properties = properties,
        rng = Random.MersenneTwister(seed)

    for id in 1:100
        start = random_position(model) # At an intersection
        speed = rand(model.rng) * 5.0 + 2.0 # Random speed from 2-7kmph
        human = Zombie(id, start, false, speed)
        add_agent_pos!(human, model)
        OSM.plan_random_route!(human, model; limit = 50) # try 50 times to find a random route
    # We'll add patient zero at a specific (longitude, latitude)
    start = OSM.nearest_road((9.9351811, 51.5328328), model)
    finish = OSM.nearest_node((9.945125635913511, 51.530876112711745), model)

    speed = rand(model.rng) * 5.0 + 2.0 # Random speed from 2-7kmph
    zombie = add_agent!(start, model, true, speed)
    plan_route!(zombie, finish, model)
    # This function call creates & adds an agent, see `add_agent!`
    return model
initialise (generic function with 1 method)

In our model, zombies are seemingly oblivious to their state, since they keep going about their business, but start eating people along the way. Perhaps they can finally express their distaste for city commuting.

function agent_step!(agent, model)
    # Each agent will progress along their route
    # Keep track of distance left to move this step, in case the agent reaches its
    # destination early
    distance_left = move_along_route!(agent, model, agent.speed * model.dt)

    if is_stationary(agent, model) && rand(model.rng) < 0.1
        # When stationary, give the agent a 10% chance of going somewhere else
        OSM.plan_random_route!(agent, model; limit = 50)
        # Start on new route, moving the remaining distance
        move_along_route!(agent, model, distance_left)

    if agent.infected
        # Agents will be infected if they get too close (within 10m) to a zombie.
        map(i -> model[i].infected = true, nearby_ids(agent, model, 0.01))
agent_step! (generic function with 1 method)

Visualising the fall of humanity

using InteractiveDynamics
using CairoMakie
ac(agent) = agent.infected ? :green : :black
as(agent) = agent.infected ? 10 : 8
model = initialise()

abmvideo("outbreak.mp4", model, agent_step!;
title = "Zombie outbreak", framerate = 15, frames = 200, as, ac)
 Downloading artifact: osm_map_gottingen
[ Info: Created OSMGraph object with kwargs: `network_type=drive`, `weight_type=distance`, `graph_type=static`, `precompute_dijkstra_states=false`, `largest_connected_component=true`
┌ Warning: Since there are a lot of edges (3639 > 500), they will be drawn as straight lines even though they contain curvy edges. If you really want to plot them as bezier curves pass `edge_plottype=:beziersegments` explicitly. This will have much worse performance!
└ @ GraphMakie ~/.julia/packages/GraphMakie/Smv4A/src/recipes.jl:524