Firmament-Autopilot/FMT-Firmware

Firmament Autopilot Embedded System

What it solves

Firmament (FMT) provides a high-performance embedded system for building control systems for autonomous vehicles, including UAVs, cars, ships, and robots. It aims to reduce development efficiency and simplify debugging by using a Model-based design (MBD) approach.

How it works

The system is split into two primary components: the firmware (FMT-Firmware), written in C/C++, and a simulation framework (FMT-Model) built with MATLAB/Simulink. This allows for auto-code generation from Simulink models to be deployed onto various hardware platforms such as ARM, AMD, and Intel.

Who it’s for

Developers and engineers building autonomous vehicle control systems who need a stable, real-time embedded system that supports a wide range of hardware, including Pixhawk controllers.

Highlights

  • Model-Based Design: Enables faster development and easier debugging through simulation and auto-code generation.
  • Real-time Performance: Built on the RT-Thread RTOS for high stability and performance.
  • Hardware Compatibility: Supports a wide array of hardware including Pixhawk (FMUv2 and FMUv5), SIEON S1, and various CUAV boards.
  • Standardized Communication: Compatible with Mavlink and QGroundControl.

Related

  • Project
    commaai/openpilot

    An open-source operating system for robotics that upgrades driver assistance systems in over 300 supported cars.

  • Project
    ArduPilot/ardupilot

    ArduPilot is a comprehensive open-source autopilot software suite that provides advanced control for a variety of unmanned vehicles, including drones, planes, and submarines.

  • Dispatch
    FLUX 3 and FLUX-mimic: Integrating Video Generation with Robot Action Models

    Black Forest Labs and mimic robotics have developed FLUX-mimic, a video-action model based on the FLUX 3 multimodal foundation model that enables robots to perform complex manipulation tasks by decoding world knowledge from video prediction.

  • Project
    InternLM/lmdeploy

    A toolkit for compressing, deploying, and serving LLMs and VLMs, offering high-performance inference engines and quantization to maximize throughput.

  • Project
    TongjiSuperPower/sp_vision_25

    An open-source auto-aiming and fire-control system for combat robots that uses neural network detection and trajectory planning to maximize hit rates and minimize kill times.