Distances to group centroid

Compute the distance of a given agent to the group centroid. This group centroid is computed as a summary statistic of the predicted x- and y- coordinates of all pedestrians belonging to the same group as the agent. The summary statistic of choice should be one of mean-tendency, but can be specified by the user through the argument fx.

Usage

distance_group_centroid(predictions, centers, number_agents, fx = mean)

Arguments

predictions

Numeric matrix with shape N x 2 containing predicted positions of all pedestrians that belong to the social group of the agent.

centers

Numerical matrix containing the coordinates at each position the object can be moved to. Should have one row for each cell.

number_agents

Integer indicating number of people in the pedestrian's social group.

fx

Function used to find the group centroid. Defaults to mean

Value

Numeric vector containing the distance from each cell in the `center` to the group centroid. If not other agents belong to the same group as the agent, returns NULL.

Details

Note that this function has been defined to be in line with the m4ma utility functions.

See also

gc_utility, utility-agent