This project implements a Python-based A path planning algorithm* for autonomous mobile robots navigating a simulated warehouse environment.
The system models a 30 × 25 m warehouse with obstacles such as narrow passages and dead-end structures, and computes collision-free paths for robots of different sizes.
Designed for autonomous mobile robots (AMRs) in warehouse automation systems.
-
A* path planning algorithm for shortest-path navigation
-
Collision avoidance using configuration space (C-space)
-
Multiple robot geometries (0.5m, 1m, 2m)
-
Handles complex edge cases:
- Narrow passage constraints
- U-shaped dead-end avoidance
-
Grid-based environment simulation
-
Visualisation using Matplotlib
The project uses the A* search algorithm:
[ f(x) = g(x) + h(x) ]
g(x)= cost from start nodeh(x)= estimated cost to goal (Euclidean distance)
- Warehouse size: 30 × 25 meters
- Start point:
(2, 1) - Goal point:
(28, 23) - Includes narrow passage and dead-end obstacles
This simulation demonstrates how robot size affects path feasibility and efficiency in constrained warehouse environments.
- Smaller robots take shorter, more direct paths
- Larger robots avoid narrow passages due to collision constraints
- All robots successfully avoid dead-end traps
git clone https://github.com/jessysutherns/warehouse-path-planning.git
cd warehouse-path-planning
pip install -r requirements.txt
python src/path_planning.py- Python
- NumPy
- Matplotlib
- Dynamic obstacle handling
- Multi-robot coordination
- Real-time replanning (e.g., D*)
- Validation in real-world environments
Jessica Sutherns https://github.com/jessysutherns
This project demonstrates the practical application of robotics algorithms, autonomous navigation, and path planning in real-world warehouse automation scenarios.
It highlights how autonomous mobile robots (AMRs) can improve efficiency, scalability, and operational performance in intralogistics systems.