Getting StartedLink

You have two options available 1) Get a working code base container using Docker, or 2) Download the source code and its dependencies.


For Multi-platform execution (Windows, MAC, Linux, Azure, AWS)

You can quickly get a contained working copy in few minutes of the latest LAPKT version in the repository using Docker. For instructions, please visit:

Download sourcesLink

The command:

    git clone <directory>

will create a clone of the LAPKT master repository in <directory>. The directory is created if it does not yet exist.


LAPKT requires the following libraries & tools:

  • scons
  • boost
  • boost::program_options
  • varjudy

optional FF-parser:

  • makedepend
  • flex
  • bison

optional FD-Parser:

  • python
  • python-dev
  • boost::python-dev

In order to compile LAPKT, we recommend g++ 4.7 or better. However, any compiler able to handle both boost libraries and C++11 standard new features, should also be usable (we have been able to compile it under Visual Studio 2010, llvm, and the new Linux bash shell on windows 10). makedepend comes in xutils-dev package.

Building LAPKTLink

In order to build LAPKT you need to install scons (a GNU Makefile replacement) in your system. Refer to for directions on how to achieve this.

In order to compile some of the examples, you will also need a version >= 1.49 of the Boost C++ libraries available on your system. You can check the version you have either manually by looking at the macro defined in boost/version.hpp or, on debian systems, by running dpkg -s libboost-dev.

Finally, LAPKT requires the Judy library ( to support the bitmap array class 'Varset Judy'. NOTE: This dependency will be optional or entirely deprecated in the future.

The following command installs all the dependencies on Ubuntu version > 12.04LTS:

sudo apt-get update && sudo apt-get install --no-install-recommends -y \
build-essential \
ca-certificates \
xutils-dev \
scons \
gcc-multilib \
flex \
bison \
python \
python-dev \
python3 \
python3-dev \
libboost-python-dev \
libboost-dev \
libjudy-dev \
libboost-program-options-dev \

Build InstructionsLink

Issue the command


at the root of the source directory to obtain the (static) library containing essential data structures and other miscellaneous utilities. If debug symbols are needed, the command

scons debug=1

builds the library with optimizations disabled and debug symbols enabled.

If you want to use FF-parser, compile the ff into a library by running the following commands:

cd external/libff
make clean
make depend

If you are a Mac OS X user, run this command to create the final dynamic library

libtool -o libff.a *.o

As Mac OS X don\'t like static libraries, if you run scons and you get the following error:

ld: library not found for -lcrt0.o

edit the SConstruct and comment (add #) the following line:

common_env.Append( LINKFLAGS = [ '-static' ] )

Compiling Examples and PlannersLink

To compile any example go to a specific folder and type


In FD-parser based examples type


The examples for the 'planner agnostic' interface can be found on


and cover the following topics:

* agnostic-examples/assembling_strips_problems

    Shows how to define a STRIPS planning problem programatically.

* agnostic-examples/successor_generation

    Discusses the different way of generating successors during search.

* agnostic-examples/bfs
* agnostic-examples/bfs-double-queue 
* agnostic-examples/bfs-double-queue-secondary-heuristic

    Shows how can one assemble available components to deliver a
    planner built around a BFS search engine, with multiple queues and
    secondary heuristics, on a parametrized planning task.

* agnostic-examples/das

    Shows how can one assemble available components to deliver a
    planner built around Deadline Aware Search.

Alternatively, To learn how to ensemble different heuristics go through simple planners like


To learn how to use or create a planner using FD-parser, copy and edit a simple planner like


and to learn how to use or create a planner using FF-parser


or copy and edit a simple Breadth first search planner like


or start from the classic heuristic Search Planner


    Shows how to assemble a best-first search using h_max with multiple modes: greedy/delayed/anytime,
    over a task specified in PDDL and parsed by FF-parser