Skip to content

uzh-rpg/agilicious

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

76 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Agilicious: Open-Source and Open-Hardware Agile Quadrotor for Vision-Based Flight


Agile flight done right!


Read the Paper!
Watch the full Video!

Agilicious is a co-designed hardware and software framework tailored to autonomous, agile quadrotor flight, which has been developed and used since 2016 at the Robotics and Perception Group (RPG) of the University of Zurich. Agilicious is described in this Science Robotics 2022 paper. It is completely open-source and open-hardware and supports both model-based and neural-network-based controllers. Also, it provides high thrust-to-weight and torque-to-inertia ratios for agility, onboard vision sensors, GPU-accelerated compute hardware for real-time perception and neural-network inference, a real-time flight controller, and a versatile software stack. In contrast to existing frameworks, Agilicious offers a unique combination of flexible software and high-performance hardware. Agilicious has been used in over 30 scientific papers at our lab, including trajectory tracking for drone racing scenarios at up to 5g and 70km/h (SciRob21_Foehn, Video), vision-based acrobatic flight (RSS20 Kaufmann, Video), obstacle avoidance in both structured and unstructured environments using solely onboard perception (SciRob21_Loquercio, Video), and hardware-in-the-loop simulation in virtual-reality environments. Thanks to its versatility, we believe that Agilicious supports the next generation of scientific and industrial quadrotor research. The full list of publications using Agilicious can be found here.

Credits

If you use the code in the academic context, please cite:

Additionally, please cite the following papers for the specific extensions you make use of:

What's in it for you?

Agilicious is split in two parts, agilib and agiros.

agilib contains base classes and module implementations providing the base functionality necessary for agile flight. It contains controllers, estimators, and logic, but with minimal dependencies (mainly Eigen). It allows users to integrate Agilicious into their own infrastructure and can easily be extended with new modules. Meanwhile, agiros provides bindings to common ROS interfaces, which make it easy and quick to set up and get flying, both in simulation and real world.

In summary, this library offers the following modules:

  • Pipelines provide a modular way to combine controllers, estimators, reference trajectories into a complete control system. It is possible to create multiple pipelines and switch between them at runtime, allowing rapid prototyping.
  • Controllers
  • Estimators
    • A standard EKF using a pose estimate and propagating with IMU measurements.
    • An EKF that uses a constant acceleration model and the Pose + IMU measurements.
    • A feedthrough module that provides a simple way to pipe in your own estimate sources.
    • A mock estimator that can corrupt an estimate with noise and bias to simulate real-world properties.
  • References
    • Hover
    • Velocity Commands
    • Sampled Trajectories
    • Polynomial Trajectories
  • Samplers
    • Time-based sampling along a given reference.
    • Positional sampling along a given reference which allows to robustly handle large disturbances.
  • Bridges
    • RotorS interface
    • Serial bridge to interface with our own flight control software
    • Serial bridge to interface through LAIRD wireless modules.
    • Serial bridge to interface with BetaFlight controllers.
  • Pilot which handles all logic and interfacing.
  • Guard A safety guard implementation that can use a secondary estimate (e.g. motion capture) and take over if some criteria are invalidated. The guard can use a secondary pipeline configuration without relying on the user-defined modules. This allows for rapid prototyping with a virtual "safety net".
  • Simulator A sophisticated and configurable simulation that runs magnitudes faster than real time and optionally provides aerodynamic BEM models as described in Bauersfeld RSS'21.

License

The official license file can be found here.

To request access for collaboration, please go here