Abstraction layer for crossing factor graphs over various technologies
Author JuliaRobotics
13 Stars
Updated Last
1 Year Ago
Started In
May 2018


Click on badges to follow links:

Release v0.13 Release v0.14 Dev Coverage Documentation
Build Status Build Status Build Status
Average time to resolve an issue
Percentage of issues still open

DistributedFactorGraphs.jl provides a flexible factor graph API for use in the Caesar.jl ecosystem. The package supplies:

  • A standardized API for interacting with factor graphs
  • Implementations of the API for in-memory and database-driven operation
  • Visualization extensions to validate the underlying graph

Note this package is still under initial development, and will adopt parts of the functionality currently contained in IncrementalInference.jl.


Please see the documentation and the unit tests for examples on using DistributedFactorGraphs.jl.


DistributedFactorGraphs can be installed from Julia packages using:

add DistributedFactorGraphs


The in-memory implementation is the default, using LightGraphs.jl.

It is recommended to use IncrementalInference to create factor graphs as they will be solvable.

using DistributedFactorGraphs
using IncrementalInference

Both drivers support the same functions, so choose which you want to use when creating your initial DFG. For example:

# In-memory DFG
# Initialize the default in-memory factor graph with default solver parameters.
dfg = initfg()
# add 2 ContinuousScalar variable types to the new factor graph
addVariable!(dfg, :a, ContinuousScalar)
addVariable!(dfg, :b, ContinuousScalar)
# add a LinearConditional factor
addFactor!(dfg, [:a, :b], LinearConditional(Normal(10.0,1.0)))