LISA: Logistics Intelligence & Swarm API

LISA is the core fleet management and dispatching system for automated warehouses. It acts as an intelligent Middleware, orchestrating fleets of AMRs (Autonomous Mobile Robots), such as MONA, and other AGVs.

Architecture & How It Works

LISA is designed to be lightweight, scalable, and independent of any specific robot hardware. It communicates with higher-level systems (WMS/ERP) via REST API and dispatches commands to the robot swarm using industrial IoT standards (MQTT + VDA 5050).

graph TD
    %% Define Styles
    classDef higherLevel fill:#2d3436,stroke:#74b9ff,stroke-width:2px,color:#fff;
    classDef lisaCore fill:#0984e3,stroke:#fff,stroke-width:2px,color:#fff;
    classDef broker fill:#fdcb6e,stroke:#e17055,stroke-width:2px,color:#2d3436;
    classDef robot fill:#00b894,stroke:#fff,stroke-width:2px,color:#fff;

    %% Nodes
    WMS["WMS / ERP System<br>(Warehouse Management)"]:::higherLevel
    LISA["LISA API<br>(FastAPI Backend + Trajectory Planning)"]:::lisaCore
    Broker["MQTT Broker<br>(Message Router)"]:::broker
    
    MONA1["MONA Robot 001"]:::robot
    MONA2["MONA Robot 002"]:::robot
    AGV["Custom AGV Forklift"]:::robot

    %% Connections
    WMS <-->|REST API / WebSockets| LISA
    LISA <-->|MQTT Actions| Broker
    Broker <-->|VDA 5050 Standard| MONA1
    Broker <-->|VDA 5050 Standard| MONA2
    Broker <-->|VDA 5050 Standard| AGV

Loading

Example Workflow

1. Task Creation: The warehouse ERP system sends a REST POST request to LISA: "Move Pallet #123 from Zone A to Zone B".
2. Path Planning: LISA calculates the optimal path, ensuring no robots will collide at warehouse intersections.
3. Dispatch: LISA sends a VDA 5050 standard JSON payload via MQTT to the nearest available robot (e.g., MONA 01).
4. Execution & Telemetry: MONA 01 executes the physical movement and continuously streams its battery level, position, and status back to LISA via MQTT.

Quick Start

To spin up the local development environment (LISA API + MQTT Broker):

docker compose up --build

Once running, interactive API documentation (Swagger UI) is available at: http://localhost:8000/docs.