SPADE Simulator Documentation

Welcome to the official documentation for SPADE (Swarm Planning And Decision-making Evaluation) Simulator. This simulator is designed to facilitate research and development in swarm robotics and decentralized decision-making. Below are the key features of the SPADE simulator:

Key Features

  • Swarm Robotics Focus: Optimized for swarm robotics research, supporting large-scale simulations with a lightweight design.

  • Behavior Trees: Uses behavior trees to define and manage agent behaviors, allowing for flexible and structured decision-making.

  • Groot2 Integration: Compatible with Groot2 for visualizing and editing the agents’ behavior trees, enhancing ease of use and analysis.

  • Flexible Configuration: Easily configurable via a YAML file, enabling customization without modifying the simulator’s source code.

  • Custom Plugins: Integrates custom decision-making algorithms as plugins, allowing for tailored testing and experimentation.

  • Local Communication and Awareness: Agents communicate locally within specified radii and maintain situational awareness based on their situational awareness ranges.

  • Dynamic Task Generation: Supports the creation of tasks dynamically during the simulation, adapting to evolving scenarios.

  • Algorithm Comparison: Facilitates the comparison of different decision-making algorithms within a consistent simulation environment and supports Monte Carlo tests for statistical analysis.

For detailed installation instructions, please refer to the Installation Guide.

_images/GRAPE_20_agents_200_tasks.gif

Figure 1: GRAPE with 20 agents and 200 tasks.
_images/GRAPE_500_agents_10_tasks.gif

Figure 2: GRAPE with 500 agents and 10 tasks.