PathMover - AGV Path Movement Simulation Library

PathMover is a discrete-event simulation library built on the O2DESNet framework for modeling and simulating Automated Guided Vehicle (AGV) movements in warehouse and logistics networks.

Features

• Discrete-Event Simulation: Built on O2DESNet framework for accurate time-based simulation
• Path-Based Movement: Models AGV movement along predefined paths between control points
• Capacity Management: Tracks path occupancy and prevents collisions
• Routing Table Support: Precomputed routing tables for efficient path finding
• Large-Scale Networks: Supports networks with thousands of control points
• Performance Metrics: Tracks utilization, throughput, and waiting times
• Visualization Tools: Python-based visualization of network and trajectories

Project Structure

PathMover/
├── PathMoverLibrary/          # Core library (NuGet package)
│   ├── PathMover.cs          # Main simulation engine
│   ├── Vehicle.cs            # AGV vehicle interface and implementation
│   ├── ControlPoint.cs       # Network control points
│   ├── PathMoverStatics.cs   # Static network configuration
│   ├── PmPath.cs             # Path between control points
│   └── IdMapper.cs           # ID to name mapping
├── PathMoverDemo/            # Demo applications
│   ├── Program.cs            # Main demo program
│   ├── Simulator.cs          # Simple 6-node demo
│   ├── LargeScaleSimulator.cs # Large-scale simulation
│   ├── NetworkLoader.cs      # JSON network loader
│   └── SimulationLogger.cs   # Simulation logging
├── PathMoverRoutingGenerator/ # Routing table generation
│   ├── RoutingTableGenerator.cs
│   ├── AStarPathfinder.cs
│   └── DijkstraPathfinder.cs
├── RoutingTableBuilder/      # Routing table builder tool
├── NUnitTest_PM/             # Unit tests
├── Visualization/            # Python visualization tools
└── layout_gen/               # Network layout generation

Installation

NuGet Package

dotnet add package PathMover

From Source

git clone https://github.com/chen-tianhao/PathMover.git
cd PathMover
dotnet build

Quick Start

Basic Usage

using PathMover;
using O2DESNet;

// Create network statics
var statics = new PathMoverStatics();

// Add control points
var A = new ControlPoint(0);
var B = new ControlPoint(1);
var C = new ControlPoint(2);

// Add paths
statics.AddPath(A.Id, B.Id, new PmPath(A, B));
statics.AddPath(B.Id, C.Id, new PmPath(B, C));

// Create PathMover instance
var pathMover = new PathMover(statics, smoothFactor: 1.0, coldStartDelay: 0.0);

// Create AGV
var agv = new Vehicle
{
    Name = "AGV-001",
    Speed = 1.0,
    CapacityNeeded = 1
};

// Request entry
pathMover.RequestToEnter(agv, A);

Running the Demo

cd PathMoverDemo
dotnet run

Choose simulation mode:

1. Simple Demo: 6 control points, 5 AGVs
2. Large-Scale Network: 14,944 control points from JSON data

Core Concepts

Control Points

Network nodes where AGVs can enter, exit, or change direction.

Paths

Connections between control points with capacity constraints.

Vehicles (AGVs)

• Move along paths between control points
• Have speed and capacity requirements
• Follow routing tables to reach destinations

Routing Tables

Precomputed shortest paths between all control point pairs.

Advanced Features

Large-Scale Networks

Load network data from JSON files:

var networkLoader = new NetworkLoader();
var statics = networkLoader.LoadFromJson("network.json");

Performance Metrics

• Path occupancy rates
• Vehicle waiting times
• Throughput statistics
• Utilization metrics

Visualization

Use the Python visualization tools:

cd Visualization
pip install -r requirements.txt
python visualize_network.py
python visualize_trajectories.py

API Reference

PathMover Class

• RequestToEnter(IVehicle vehicle, ControlPoint cp): Request AGV entry
• AttemptToDepart(IVehicle vehicle): Attempt to depart from current path
• ReadyToExit(IVehicle vehicle, PmPath path): Signal ready to exit path

IVehicle Interface

• Name: Vehicle identifier
• Speed: Movement speed
• CapacityNeeded: Path capacity required
• CurrentPath: Currently occupied path
• TargetList: Destination control points

Examples

Simple Network Simulation

See PathMoverDemo/Simulator.cs for a complete example with 6 control points and multiple AGVs.

Custom Vehicle Implementation

public class CustomVehicle : IVehicle
{
    public string Name { get; set; }
    public double Speed { get; set; }
    public int CapacityNeeded { get; set; }
    public PmPath CurrentPath { get; set; }
    public PmPath? PengingPath { get; set; }
    public bool IsStoped { get; set; }
    public List<ControlPoint> TargetList { get; set; }
    public PathMoverStatics PathMoverStatics { get; set; }

    // Implement interface methods
    public void RemoveTarget(ushort controlPointId) { ... }
    public PmPath NextPath(ushort currentPointId) { ... }
}

Development

Building from Source

dotnet build PathMover.sln

Running Tests

cd NUnitTest_PM
dotnet test

Creating NuGet Package

cd PathMoverLibrary
dotnet pack --configuration Release

Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Add tests if applicable
5. Submit a pull request

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments

• Built on O2DESNet framework
• Developed at National University of Singapore (NUS)

Support

For issues and questions, please use the GitHub Issues page.

Change Log

Version 2.0.0

• Performance optimization: Improved path finding algorithm for large-scale networks, enhancing simulation speed
• New feature: Added dynamic routing support, allowing vehicles to adjust paths based on real-time network status
• API improvements: Simplified vehicle interface with more flexible configuration options
• Visualization enhancement: Improved Python visualization tools with real-time simulation monitoring support

Version 1.0.4

• Model revise: Since overtake is not allowed, only the 1st vehicle in OutPengList can be added to relevant InPendingList.

Version 1.0.2

• Bug fix: Add PengingPath to interface IVehicle, to manage item inside path.InPendingList in event AttemptToDepart.

Version 1.0.1

• Bug fix: Attempt to depart the specific vehicle in currentPath.InPendingList (rather than the 1st one in previousPath.OutPendingList) when release of capacity propagate backward.

Version 1.0.0

• Initial release