The code implemented as part of the thesis consists of two parts:

1. Implementation of the PTO planner in OMPL: https://github.com/janisfreund/thesis-ompl The code is a fork of https://github.com/ompl/omplapp.

2. Implementation of the simulation environment in PyBullet: https://github.com/janisfreund/thesis-pybullet The code builds on the OMPL/PyBullet Interface https://github.com/lyfkyle/pybullet_ompl.

First, both repositories have to be cloned to the same directory. OMPL can be installed by performing the following steps:

1. Run the installation script provided by ompl: https://ompl.kavrakilab.org/install-ompl-ubuntu.sh using the command:./install-ompl-ubuntu.sh --python
2. Create build directory in ./thesis-ompl with mkdir build and go to the folder with cd build
3. Generate Python bindings with make -j 4 update_bindings
4. Compile with make -j 4
5. Install required Python packages with pip install pybullet numpy opencv-python scipy matplotlib progressbar

The most relevant files for the thesis are:

• ./thesis-ompl/ompl/src/ompl/geometric/planners/partial/Partial.h and ./thesis-ompl/ompl/src/ompl/geometric/planners/partial/src/Partial.cpp implement the PTO planner
• ./thesis-ompl/ompl/src/ompl/base/World.h and ./thesis-ompl/ompl/src/ompl/base/src/World.cpp implement the world class

The most relavant files are:

• ./thesis-pybullet/pb_ompl.py: Implements the interface to OMPL
• ./thesis-pybullet/scripts/benchmarks.py: Generates benchmark plots
• ./thesis-pybullet/scripts/camera_state_sampler.py: Implements the camera-based state sampler
• ./thesis-pybullet/scripts/create_env_imgs.py: Can be used to create images of environments
• ./thesis-pybullet/scripts/demos.py: Demos planned path trees
• ./thesis-pybullet/scripts/environments.py: Specifies the environments
• ./thesis-pybullet/scripts/robots.py: Defines the robots
• ./thesis-pybullet/scripts/test_environments.py: Tests specific robot configurations in specified environmnets

Models of the robots and environments are stored in the directory ./thesis-pybullet/models. This includes models from the following sources:

• https://github.com/AutonomyLab/create_robot
• https://www.turbosquid.com/de/3d-models/3d-model-golden-retriever-1850419
• https://www.turbosquid.com/de/3d-models/3d-model-lowpoy-suv-car-1937266
• https://sketchfab.com/3d-models/isometric-office-d31464eed8044190911b221648aca432
• https://sketchfab.com/3d-models/floor-plan-with-furniture-813f854c770b4678a616a9ebf7534fe6
• https://www.turbosquid.com/de/3d-models/bookshelf-for-low-poly-games-3d-model-1860043
• https://www.turbosquid.com/de/3d-models/3d-industrial-shed-warehouse-building-1603744