Skip to contents

Create initial condition

Source: R/simulate.R
create_initial_condition.Rd

Creates a list of agents that can be used as an initial condition to the simulation. Agents are placed at random locations in the room. Main advantage of this function is that you don't have to wait for a room to fill up with the required number of agents, but rather impose that the room is already filled with this number of agents.

Usage

create_initial_condition(
  agent_number,
  model,
  goal_number = function(n) rnorm(n, 10, 2),
  group_size = matrix(1, nrow = 1, ncol = 2),
  space_between = 1.25,
  precomputed_edges = NULL,
  cpp = TRUE,
  ...
)

Arguments

agent_number

Numeric denoting the number of agents that should be included in the initial condition.

model

Object of the predped-class.

goal_number

Numeric, vector, or function that defines the number of goals the agents should accomplish. It's exact value is handled by determine_values. Defaults to \(n) rnorm(n, 10, 2).

group_size

Matrix of size n x 2 containing the group sizes you would like to simulate (first column) and the probability of observing groups with this size (second column). Defaults to a 100 agents.

space_between

Numeric denoting the space that should be left between an object and the created path points for the agents (in radii). Defaults to 2.5, meaning a space of 2.5 * radius(agent) is left between an object and the path points agents use in their strategy. Ignored if precomputed_edges is provided.

precomputed_edges

Output of compute_edges containing the nodes and edges the agent can use to plan its path. Defauls to NULL, triggering the creation of these edges whenever they are needed.

cpp

Logical denoting whether to use the Rcpp (TRUE) or R (FALSE) alternatives for the lower-level functions. Defaults to TRUE.

...

Additional arguments provided to add_agent.

Value

List of instances of the agent-class.

Details

In this function, we use the following approach to simulate an initial condition. First, we create an agent using the add_agent function. This agent is given a specific group id based on the group_size argument.

Once created, we then compute all possible paths that the agent may be walking on through the compute_edges function. One of these paths is then chosen, and subsequently a random point on that path is chosen. We then check whether an agent would be able to take the chosen position without intersecting any of the objects in the environment. If not, then we choose another point until we can place the agent there. If we do not find a fitting position for the agent after 10 iterations, we break out of the loop and stop trying to fit in agent_number agents in the space.

Finally, group members are identified and given the same set of goals.

See also

agent-class, predped-class, add_agent, simulate, simulate