2.6 KiB

Raw Blame History

Robot SLAM Simulation with Sensor Crosstalk Management

This project simulates a robot equipped with Ultrasound and IR sensors navigating a 2D environment, performing SLAM (Simultaneous Localization and Mapping), managing dynamic crosstalk, and building an occupancy grid map in real-time, with optional GUI visualization.

📦 Installation (Linux)

1. Set up a Python virtual environment

python3 -m venv slam_env
source slam_env/bin/activate

2. Install required Python libraries

pip install numpy matplotlib PyQt5 scikit-learn pytmx

3. Install GTSAM manually from GitHub

Follow the official GTSAM installation guide:
https://github.com/borglab/gtsam

Summary of typical steps:

sudo apt-get install cmake libboost-all-dev libtbb-dev libeigen3-dev
git clone https://github.com/borglab/gtsam.git
cd gtsam
mkdir build && cd build
cmake ..
make -j$(nproc)
sudo make install

Then, install Python bindings:

cd ../python
mkdir build && cd build
cmake ..
make -j$(nproc)
sudo make install

If needed, update your PYTHONPATH:

export PYTHONPATH=/usr/local/lib/python3.x/dist-packages:$PYTHONPATH

Replace python3.x with your installed version.

🚀 Usage

Run the simulation in GUI mode

python your_script.py

or explicitly:

python your_script.py --gui

Run the simulation in headless (non-GUI) mode

python your_script.py --no-gui

Load a custom environment from a Tiled map

python your_script.py --tiled_map your_map.tmx

Controls (in GUI mode)

Key	Action
`Z`	Move forward
`S`	Move backward
`Q`	Turn left
`D`	Turn right
`X`	Stop movement
`SPACE`	Pause / Resume
`ESC`	Save sensor layout

When closing the GUI, the following files are automatically saved:

map.pkl → Occupancy grid as a pickle file
map.png → Occupancy grid image

🔧 Project Structure

File/Folder	Description
Sensor, MotionModel, SLAMEngine, etc.	Core simulation classes
SensorLayoutEditor	GUI editor for live sensor placement
Visualizer	2D occupancy grid visualization
Environment	Wall and obstacle management (including Tiled map support)
OccupancyGrid, CoverageAnalyzer	Mapping and blind spot detection
main()	Entry point for launching GUI or headless simulation

2.6 KiB Raw Blame History