The International Conference and Competition on MAV was organized by TUDelft mid-September 2022. The ENAC took part to the outdoor competition, a package delivery challenge, with several multi-rotors and a hybrid tail-sitter, all powered by Paparazzi.

They won the competition with a quite large amount of points. Nevertheless, all teams performed well with some innovative platforms.

Several papers have been presented to the conference by the ENAC team and the other participants. The proceedings are freely available on the conference website.

The latest Paparazzi stable release is out today with many changes and addition since 5.16 one year ago. One of the biggest changes is the complete rewrite of the build system for the aircraft code generation (good old Ocaml stuff), plus the addition of a proper dependency resolution for the modules. It has no big impact for users so far, but in order to move forward, we will most likely break a few things…

Hence, 5.18 will be the last of the series 5. The main branch is now labeled with 6.0_unstable and will start by a big cleaning. Most noticeable one is that the boards and tools based on lpc21 and stm32 bare-metal (using libopencm3) will be dropped. So the only supported architectures will be Linux for simulation and Parrot drones and ChibiOS for micro-controllers.

More in-depth changes will also come, including better usage of the new dependency mechanism, rearrangement of the static scheduling of the autopilot, and if possible a new set of messages for pprzlink (lot’s of cleaning here as well…).

The complete changelog and release package are available at https://github.com/paparazzi/paparazzi/releases/tag/v5.18.0_stable

Now, you can have your aircraft tracking a (smooth) 3D parametric trajectory.  This is an extension of the existing GVF in Paparazzi, in fact, now we allow even 2D trajectories with self-intersections.

It is based on our accepted article in IEEE Transactions on Robotics Singularity-free Guiding Vector Field for Robot Navigation

All the instructions to use the algorithm are in the wiki.

The Paparazzi UAV System as a research tool will be presented as an invited talk at the SPLASH2020-REBASE conference (Mon 16 Nov 2020 18:00, Paris time). See the program of the conference for details. The presentation will be available on live and free youtube.

https://2020.splashcon.org/details/splash-2020-rebase/19/Paparazzi-an-open-source-UAV-for-research

UPDATE: the video of the conference is online

Enac UAV lab have participated to an atmospheric research flight campaign at the Barbados Island for three weeks from end of January 2020. 

This was part of a project called NEPHELAE in collaboration with the French National Meteorological Research Center from Météo-France and the robotics lab LAAS-CNRS.

NEPHELAE is atmospheric science driven (with a focus on cloud microphysical processes) via robotic technological development (adaptive, model-guided path planning in a UAV fleet). Essentially, by merging the state-of-the-art technology, the observational tools for conducting research in atmospheric science will take a quantum leap forward. At the same time, we are using this novel approach to address decades-old questions on entrainment and the onset of precipitation that have been limited by traditional observing methods.

The UAVs that have be deployed during a field campaign (Skywalker X10) in Barbados Island are capable of detecting the edges of the clouds and react to adapt their trajectory. Several patterns have been developped in collaboration with the atmospheric scientists. Each of them is dedicated to extract certain parameters at the core of the cloud or at the interface between the free atmosphere and the cloud.

In total, more than 30 operations with one or two UAVs have been done, for a flight time of almost 38 hours.

The popular Open-Source platform Crazylie v2.1 from Bitcraze is now supported by Paparazzi. Only the main frame, with IMU, barometer as well as the NRF communication is working. A bridge between the Crazyradio and the Ivy bus allows the direct and easy connection of the Crazyflie to the ground station. More sensor decks will be supported in the future.

See the (almost) maiden flight inside the flight arena at Enac:

Thank you to the Bitcraze dev team for their support !

Since there are many Windows users who like to try paparazzi on this popular OS and Microsoft introduced the possibility of running Ubuntu programs and tools on windows 10, now Paparazzi developer team is pleased to announce the initial release of the Windows10 support of Paparazzi.

The installation process is not so much different from the Linux, but you need some extra steps to install Ubuntu, and some more for using graphical interface and connecting to the aircraft via telemetry. This first release still retains the look and feel of its Linux sibling as currently it uses X Windows for rendering, so you need to install X server in order to have graphical interface.

Although its still on developing and you need some extra work to communicate with serial port, or stop process manually, … .But maybe it is better you install it yourself and give us the feedback, or if you can help to make it better, we will be happy!

Don’t hesitate to go through the wiki install page for Paparazzi on Windows 10 (thanks to Iman Shidareh).

Enac UAV team and TUDelft MAVLAB team took part to the IMAV2018 conference and competition in Australia with great success!

For the first time, the two academic lead developers of the Paparazzi UAV system join efforts to participate to the Outdoor competition and won this event far ahead.

Several tasks have been completed, including a mapping (performed using OpenDroneMap) of the flight area with automatic detection of hazards hidden in the field (with a light fixed wing aircraft) and an autonomous obstacles avoidance navigation near trees (with Bebop2 + SlamDunk from Parrot).

During the Indoor event, Enac and TUDelft were competing. And if Enac finally won the competition regarding to the score with ultra light modified CX10 quadcopters, TUDelft made an outstanding demonstration of the Delfly Nimble. This small flapping wing UAV has an amazing agility and autonomous vision-based navigation, recognized with the Indoor Innovation prize.

All these success have been possible thanks to the hard work of the Paparazzi community over the past 15 years of course, but also thanks to many other Open-Source projects including:

• OpenDroneMap
• ChibiOS
• OpenCV
• Linux, GCC and others

For the hardware parts:

• 3D printing of the plane fuselage using Markforged printer
• Embedded camera uEye IDS
• Embedded processor NanoPi Neo Air
• Autopilot Apogee and Lisa/MXS
• Parrot Bebop2 and SlamDunk

The conference papers can be found here: http://www.imavs.org/imav2018-proceedings/, including several papers presented by Enac and TUDelft teams.

Thanks to everyone and see you next year in Spain !

This is a first step of a hybrid IDS method based on the analysis of spectral traffic and a robust controller / observer for the estimation of anomalies in UAV networks. This module is currently designed to observe the traffic between the drones and the Paparazzi GCS. It provides a statistic signature of the traffic which can later be used to determine the nature of the traffic. The module is tested in face of a DoS attack and the results are very promising!

Check out our paper for more details: https://hal-enac.archives-ouvertes.fr/hal-01652296

Since the beginning of the year, ENAC (French Civil Aviation University) is equipped with a new facility dedicated to UAVs research and education.

The building includes a flying arena with a size around 10x10x10 meters, several workshops for mechanics, electronic, composite, 3D printing and storage. It also includes a teaching room for automatic control and the student’s robotics club.

The official opening was the 17th of May in the presence of the French minister of transport. Several demonstrations have been performed, all of them showing the formidable capabilities of Paparazzi: distributed formation flight, hybrid vehicles, on-board image processing, efficient adaptive control and autonomous navigation.