Included Modules

There are several modules included with VIPRA.

Model:

Calm Model

Goals:

AStar

Map:

QuadTree

Pedestrian Set:

Grid

Input:

JSON

Output:

JSON

Models

Calm Model

This Model module implements the Calm Model

Parameters:

meanMass : Mean mass for population
massStdDev : Standard Deviation for mass
meanReactionTime : Mean reaction time for population
reactionTimeStdDev : Standard Deviation for reaction time
meanMaxSpeed : Mean max speed for population
maxSpeedStdDev : Standard deviation for max speed
meanShoulderLen : Mean shoulder length for population
shoulderLenStdDev : Standard deviation for shoulder length

In module_params.json:

"model": {
  "calm": {
    "meanMass": 1.0,
    "massStdDev": 0.1,
    "meanReactionTime": 0.5,
    "reactionTimeStdDev": 0.05,
    "meanMaxSpeed": 1.0,
    "meanShoulderLen": 0.2,
    "maxSpeedStdDev": 0.1,
    "shoulderLenStdDev": 0.05
  }
},

Goals

AStar Goals

This Goals module uses the A* algorithm to find paths for each pedestrian.

Parameters:

endGoalType : Name of object in map each pedestrian tries to reach (e.g. "exit")
goalRange : Range, in meters, before a goal is counted as "reached" (e.g. 0.05)
gridSize : Length of each side, in meters, of each grid in the pathing graph (e.g. 0.1)
closestObstacle : Closest a grid center can be to an obstacle before being considered non-traversable

In module_params.json:

"goals": {
  "AStar": {
    endGoalType: "exit",
    goalRange: 0.1,
    gridSize: 0.1,
    closestObstacle: 0.35
  }
},

Maps

QuadTree Map

This Map module uses the Quad Tree to hold the map geometry.

Parameters:

minQuadSize : Size at which quads stop being subdivided

In module_params.json:

"map": {
  "QuadTree": {
    "minQuadSize": 0.05
  }
},

Pedestrian Sets

Grid Pedestrian Set

This Pedestrian Set module holds pedestrians in a grid of cells, allowing for more efficient lookup of nearest neighbors

Parameters:

gridSize: Size of each grid cell

In module_params.json:

"pedestrians": {
  "Grid": {
    "gridSize": 0.5
  }
},

Inputs

JSON Input

This Input module loads JSON data, using nlohmann JSON.

For importing maps, it is recommended to use the DXF module. More information on how to use .dxf files in your project can be found here.

Parameters:

NONE

Outputs

TrajectoriesJson Output

This Output module creates a JSON file containing the positions of each pedestrian for each timestep that matches the output frequency in the simulation module parameters.

Parameters:

filename: Name of the output file

In module_params.json:

"output": {
  "coordinator": {  // Output is special in that it has a coordinator for when there are multiple output modules
    "output_dir": "./sim_output"
  },
  "TrajectoriesJson": {
    "filename": "trajectories.json"
  }
},