Utilities –- Parts and Usage Overview

A module for handling data in and out.

Main public types and functions:

• imgToMap
• maps_dir
• normalize
• output_dir
• output_ext
• produceMap
• produceMaps
• save
• saveBeliefMapToPng
• spatialAve

The default directory containing the maps that can be used as data during simulated sampling.

The default output directory where files can be saved.

The default extension for saved output files.

Struct/function for generating ground truth values from a linear combination of gaussian peaks.

Examples

GaussGroundTruth(peaks) # pass in a list of Peaks

Produces ground-truth value(s) for a point or list of points. Accepts a single vector, a vector of vectors, or a matrix of column vectors.

Each probability distribution component is divided by its own peak height and the highest of all the peaks before being added into the total. This causes the entire ground truth map to have a max value of (about) 1.

Used within a GaussGroundTruth. Holds a 2D normal distribution and the desired height of the peak.

Inputs:

• μ: the peak location (distribution mean)
• Σ: the peak width (distribution covariance)
• h: the peak height

Creates a string containing the date and time separated by dashes. Can pass in a DateTime object, defaults to current time.

Takes a matrix in the format created from an image, re-formats it, and returns a GridMap. Images view a matrix with its indexing top-down and left-right. GridMaps view a matrix with its indexing left-right and bottom-up.

Examples

using DelimitedFiles: readdlm

image = readdlm(file_name, ',')
bounds = (lower = [0.0, 0.0], upper = [1.0, 1.0])
map = imgToMap(image, bounds)
map = imgToMap(image) # or auto bounds

A helper method to normalize an array so its values are within the range [0, 1].

produceMap(
    func,
    map::GridMaps.GridMap
) -> Union{GridMaps.GridMap{_A, _B, _C, Float64} where {_B, _C<:AbstractArray{_A, _B}}, GridMaps.GridMap{_A, N, _C, Float64} where {N, _C<:AbstractArray{_A, N}}} where _A

Generates a GridMap from a function for chosen bounds and dimensions. Can also pass in another GridMap in place of bounds and dims.

produceMaps(
    beliefModel::MultiQuantityGPs.MQGP,
    map::GridMaps.GridMap;
    quantity
) -> Tuple{Union{GridMaps.GridMap{_A, _B, _C, Float64} where {_B, _C<:AbstractArray{_A, _B}}, GridMaps.GridMap{_A, N, _C, Float64} where {N, _C<:AbstractArray{_A, N}}} where _A, Union{GridMaps.GridMap{_A, _B, _C, Float64} where {_B, _C<:AbstractArray{_A, _B}}, GridMaps.GridMap{_A, N, _C, Float64} where {N, _C<:AbstractArray{_A, N}}} where _A}

Generates belief and uncertainty GridMaps from a belief model for chosen bounds and dimensions. Can also pass in another GridMap in place of bounds and dims.

save(
    mission,
    samples,
    beliefs,
    times;
    animation,
    sub_dir_name,
    file_name
) -> Union{Nothing, Plots.AnimatedGif}

A collection of methods for saving data from missions, metrics, and general julia objects.

saveBeliefMapToPng(beliefModel, occupancy) -> Any
saveBeliefMapToPng(beliefModel, occupancy, file_name) -> Any

This is really just to give something out to munch, so it needs to be an rgba png with the last channel as the amount.

A helper method to perform a spatial average on a matrix. The extent of the average can be chosen with its default being 1.

A module with methods to visualize mission data. The form displayed is determined by what data types are passed in.

Main public types and functions:

• vis
• visualize

getAxes(map) -> Tuple{Any, Any}

Method to get the x and y plotting axes. This (re)generates them only if needed and saves them into global module variables for future use.

Simple convenience function that also displays the output of any visualize function. See those.

visualize(map::GridMaps.GridMap; ...) -> Any
visualize(map::GridMaps.GridMap, title; points, clim) -> Any

Method to show any GridMap data.

visualize(
    beliefModel::MultiQuantityGPs.MQGP,
    samples,
    occupancy;
    ...
) -> Any
visualize(
    beliefModel::MultiQuantityGPs.MQGP,
    samples,
    occupancy,
    new_loc;
    quantity
) -> Any

Method to show belief model values of mean and standard deviation and the sample locations that they were generated from. Shows two plots side-by-side.

visualize(sampleCost, samples, occupancy) -> Any
visualize(sampleCost, samples, occupancy, new_loc) -> Any

Method to show sample cost values.

visualize(
    md,
    samples,
    beliefModel::MultiQuantityGPs.MQGP,
    sampleCost,
    new_loc;
    quantity
) -> Any

Main method to visualize the current state of a search. Generates all the other visuals through their respective methods and lays them out in a grid. Currently shows the belief model, the ground truth, and the obstacles.

Arguments pass through to the sub-methods that need them. res is the grid resolution when plotting continuous-valued functions and defaults to [0.01, 0.01].

If no ground truth is available, it is not plotted.

visualize(
    md,
    samples,
    beliefModel::MultiQuantityGPs.MQGP,
    new_loc;
    quantity
) -> Any

visualize(sampler, map) -> Any

Method to show ground truth data from a function.

visualize(maps::GridMaps.GridMap...; titles, points) -> Any

Method to show a ground truth map and up to three other prior data maps. Pass each map in as its own argument.

A module to calculate the metrics from a mission and belief model.

Main public types and functions:

• calcMetrics

A collection of methods that calculates metrics from sampling missions.