Crazyflie at the Robotics Developer Day 2024

2024-07-01 | Kimberly McGuire | Leave a comment

Whenever we show the Crazyflie at our booth at various robotics conferences (like the recent ICRA Yokohama), we sometimes get comments like ‘ahh that’s cute’ or ‘that’s a fun toy!’. Those who have been working with it for their research know differently, but it seems that the general robotics crowd needs a little bit more… convincing! Disregarding its size, the Crazyflie is a great tool that enables users to do many awesome things in various areas of robotics, such as swarm robotics and autonomy, for both research and education.

We will be showing that off by giving a live tutorial and demonstration at the Robotics Developer Day 2024, which is organized by The Construct and will take place this Friday, 5th of July. We have a discount code for you to use if you want to get a ticket; scroll down for details. The code can be used until 12 am midnight (CEST) on the 2nd of July.

The Construct and Robotics Developer Day 2024

So a bit of background information: The Construct is an online platform that offers various courses and curriculums to teach robotics and ROS to their users. Along with that, they also organize all kinds of live training sessions and events like the Robotics Developer Day and the ROS Awards. Unfortunately, the deadline for voting in the latter has passed, but hopefully in the future, the Crazyflie might get an award of its own!

What stands out about the platform is its implementation of web-based virtual machines, called ‘ROSJects,’ where ROS and everything needed for it is already set up from the start. Anyone who has worked with ROS(2) before knows that it can be a pain to switch between different versions of ROS and Gazebo, so this feature allows users to keep those projects separate. For the ROS Developer Day, there will be about five live skill-learning sessions where a ROSject is already preconfigured and set up for the attendees, enabling them to try the tutorial simultaneously as the teacher or speaker explains the framework.

Skill learning session with the Crazyflie

One of the earlier mentioned skill learning sessions is, of course, one with the Crazyflie! The title is “ROS 2 with a Tiny Quadcopter,” and it is currently planned to be the first skill learning session of the event, scheduled at 15:15 (3:15 pm) CEST. The talk will emphasize the use of simulation in the development process with aerial robotics and iterating between the real platform and the simulated one. We will demonstrate this with a Crazyflie 2.1 equipped with a Lighthouse deck and a Multi-ranger deck. Moreover, it will also use a Qi-charging deck on a charging platform while it patiently waits for its turn :D

What we will be showing is a simple implementation of a mapping algorithm made specifically for the Crazyflie’s Multiranger deck, which we have demonstrated before at ROSCon Kyoto and in the Crazyswarm2 tutorials. What is especially different this time is that we are using Gazebo for the simulation parts, which required some skill learning on our side as we have been used to Webots over the last couple of years (see our tutorial for that). You can find the files for the simulation part in this repository, but we do advise you to follow the session first.

You can, if you want, follow along with the tutorial using a Crazyflie yourself. If you have a Crazyflie, Crazyradio, and a positioning deck (preferably Lighthouse positioning, but a Flowdeck would work as well), you can try out the real-platform part of this tutorial. You will need to install Crazyswarm2 on a separate Ubuntu machine and add a robot in your ROSject as preparation. However, this is entirely optional, and it might distract you from the cool demos we are planning to show, so perhaps you can try this as a recap after the actual skill learning session ;).

Here is a teaser of what the final stage of the tutorial will look like:

Win a lighthouse explorer bundle and a Hands-On Pass discount

We are also sponsors of the event and have agreed with The Construct to award one of the participants a Crazyflie if they win any contest. Specifically, we will be awarding a Lighthouse Explorer bundle, with a Qi deck and a custom-made charging pad similar to the ones we show at fairs like ICRA this year. So make sure to participate in the contests during the day for a chance to win this or any of the other prizes they have!

It is possible to follow the event for free, but if you’d like to participate with the ROSjects, you’ll need to get a hands-on pass. If you haven’t yet gotten a hands-on ticket for the Robotics Developer Day, please use our 50% off discount code:

19ACC2C9

This code is valid until the 2nd of July, 12 am (midnight) Central European Time! Buy your ticket on the event’s website: https://www.theconstruct.ai/robotics-developers-day/

RSS 2024 aerial swarm workshop

On a side note, we will be at the Robotics: Science and Systems Conference in Delft from July 15th to 19th, 2024—just about two weeks from now. We won’t have a booth as we usually do, but we will be co-organizing a half-day workshop titled Aerial Swarm Tools and Applications (more details on this website).

We will be organizing this workshop together with our collaborators at Crazyswarm2, as well as the developers of CrazyChoir and Aerostack2. We’re excited to showcase demos of these frameworks with a bunch of actual Crazyflies during the workshop, if the demo gods are on our side :D. We will also have great speakers, including: SiQi Zhou (TU Munich), Martin Saska (Czech Technical University), Sabine Hauert (University of Bristol), and Gábor Vásárhelyi (Collmot/Eötvös University).

Hope to see you there!

Discover the Impact of Structural Defects on Nano-Quadcopter Performance: CrazyPAD dataset

2024-06-24 | Tagir Muslimov | Leave a comment

This week we have a guest blogpost by Kamil Masalimov (MSc) and Tagir Muslimov (PhD) of the Ufa University of Science and Technology. Enjoy!

As researchers passionate about UAV technology, we are excited to share our recent findings on how structural defects affect the performance of nano-quadcopters. Our study, titled “CrazyPAD: A Dataset for Assessing the Impact of Structural Defects on Nano-Quadcopter Performance,” offers comprehensive insights that could greatly benefit the Crazyflie community and the broader UAV industry.

The Motivation Behind Our Research

Understanding the nuances of how structural defects impact UAV performance is crucial for advancing the design, testing, and maintenance of these devices. Even minor imperfections can lead to significant changes in flight behavior, affecting stability, power consumption, and control responsiveness. Our goal was to create a robust dataset (CrazyPAD) that documents these effects and can be used for further research and development.

Key Findings from Our Study

We conducted a series of experiments by introducing various defects, such as added weights and propeller cuts (Figure 1), to nano-quadcopters. For the experiments, we used the Lighthouse Positioning System with two SteamVR 2.0 virtual reality stations (Figure 2).

**Figure 1. Propeller with two side defects**

**Figure 2. Schematic of the experimental setup with Lighthouse Positioning System**

Here are some of the pivotal findings from our research:

Stability Impact: We observed that both added weights and propeller cuts lead to noticeable changes in the stability of the quadcopter. Larger defects caused greater instability, emphasizing the importance of precise manufacturing and regular maintenance.
Increased Power Consumption: Our experiments showed that structural defects result in higher power consumption. This insight is vital for optimizing battery life and enhancing energy efficiency during flights.
Variable Control Responsiveness: We used the standard deviation of thrust commands as a measure of control responsiveness. The results indicated that defects increased the variability of control inputs, which could affect maneuverability and flight precision.
Changes in Roll and Pitch Rates: The study also highlighted variations in roll and pitch rates due to structural defects, providing a deeper understanding of how these imperfections impact flight dynamics.

We show Figure 3 as an example of a graph obtained from our dataset. In this figure, you can see the altitude and thrust command over time for different flight conditions. The blue line represents the normal flight, while the orange line represents the flight with additional weight near the M3 propeller. In Figure 4, you can see the 3D flight trajectory of the Crazyflie 2.1 quadcopter under the cut_propeller_M3_2mm condition with the corrected ideal path. The blue line represents the actual flight trajectory, while the red dashed line with markers represents the ideal trajectory. Figure 5 shows the Motor PWM values over time for the add_weight_W1_near_M3 condition. The plot shows the PWM values of each motor (M1, M2, M3, and M4) as they respond to the added weight near the M3 propeller.

More examples of graphs obtained from the CrazyPAD dataset can be found in our research paper specifically describing this dataset: https://doi.org/10.3390/data9060079

**Figure 3. Altitude and thrust command over time for different flight conditions**

**Figure 4. 3D flight trajectory of the Crazyflie 2.1**

**Figure 5. Motor PWM values over time**

Leveraging Research for Diagnostic and Predictive Models

One of the most exciting aspects of our research is its potential application in developing diagnostic and predictive models. The CrazyPAD dataset can be utilized to train machine learning algorithms that detect and predict structural defects in real-time. By analyzing flight data, these models can identify early signs of wear and tear, allowing for proactive maintenance and reducing the risk of in-flight failures.

Diagnostic models can continuously monitor the performance of a UAV, identifying anomalies and pinpointing potential defects. This real-time monitoring can significantly enhance the reliability and safety of UAV operations.

Predictive models can forecast future defects based on historical flight data. By anticipating when and where defects are likely to occur, these models can inform maintenance schedules, ensuring UAVs are serviced before issues become critical.

Why This Matters for the Crazyflie Community

The CrazyPAD dataset and our findings offer valuable resources for the Crazyflie community. By understanding how different defects affect flight performance, developers and enthusiasts can improve design protocols, enhance testing procedures, and ensure higher safety and performance standards for their UAVs.

We believe that sharing our research with the Crazyflie community can lead to significant advancements in UAV technology. The dataset we created is open under the MIT License for further exploration and can serve as a foundation for new innovations and improvements.

Get Involved and Explore Further

We invite community members to explore our full research article and the CrazyPAD dataset. Together, we can drive forward the standards of UAV technology, ensuring that Crazyflie remains at the forefront of innovation and excellence.

Our research paper with a detailed description of this dataset:

Masalimov, K.; Muslimov, T.; Kozlov, E.; Munasypov, R. CrazyPAD: A Dataset for Assessing the Impact of Structural Defects on Nano-Quadcopter Performance. Data 2024, 9, 79. ht tps://doi.org/10.3390/data9060079

Dataset: https://github.com/AerialRoboticsUUST/CrazyPAD

We are eager to collaborate with the Crazyflie community and welcome any feedback or questions regarding our research. Let’s work together to push the boundaries of what’s possible in UAV technology.

Back from ICRA 2024

2024-05-27 | Arnaud | Leave a comment

Two weeks ago, Arnaud, Barbara and Rik were at ICRA 2024 in Yokohama. At our booth we showed our current products as well as the upcoming brushless Crazyflie and the camera deck prototype.

As usual, the conference has been very busy with a lot of visitors and a lot of very interesting discussion. Thanks to everyone that passed by the booth, we have come back to Sweden with a lot of energy and new ideas!

The autonomous lighthouse swarm demo demo has worked pretty well. If you are interested to know more about it you can visit our event page. It is an autonomous demo with 3 brushless Crazyflies and 6 Crazyflie 2.1s flying autonomously. With the extended battery life of the brushless, we could operate the demo pretty much continuously.

If you’ve been at the conference, you may have spotted someone proudly sporting our exclusive ‘Bitcraze took my poster’ button. We’re excited to have received posters covering a wide range of topics, the walls of our office are eagerly awaiting these visual representations of your hard work and dedication. Thank you to everyone who has contributed.

The Crazyflie 2.1 brushless is getting optional propeller guards

2024-05-21 | Tobias | 3 Comments

One of the great features of the stock Crazyflie 2.1 is that it is more or less harmless. The Crazyflie 2.1 brushless weighs roughly the same but has almost twice the amount of thrust force, so a little bit of more care is needed. We therefore decided to provide optional propeller guards. While propeller guards adds safety they also add weight and disrupt the air flow from the propellers. Adding to that, the weight is located far from the center which increases the inertia even further, resulting in a less agile drone. For some applications this is not a problem but for others it is, this is why we are making them optional, meaning they are easy to replace with simple landing legs by utilizing a snap-on fitting.

The design is not fully finalized yet but we are getting close, voilá!

If the design goes according to plan they will also withstand some bumping against walls which will be a very nice feature for many applications.

Further the landing legs and propeller guards are designed in a way so they will detach during high force impacts to prevent the PCB arms from breaking.

Come see us at ICRA 2024 Yokohama!

2024-05-13 | Kimberly McGuire | Leave a comment

“What? You are in Japan? Again!?”. Yup that is right! We loved IROS Kyoto 2022 so much that we just couldn’t wait to come back again. Barbara, Arnaud and Rik are setting up the booth as we speak to show some Bitcraze awesomeness to you! Come and say hi at booth IC085.

Crazyflie Brushless and Camera expansion

Of all the prototypes we are the most excited of showing you the Crazyflie Brushless and the ‘forward facing expansion connector prototype’ aka the Camera deck. Here you can see them both in action at a tryout of our demo. We have also written blogposts about both so make sure to read them as well (Brushless blogpost, Camera expansion blogpost)

The Crazyflie Brushless flying with a Camera deck.

Also we will explain about the contact charging prototype (see the blogpost here) and will be showing all of our decks at the booth as well. And of course our fully autonomous, onboard, decentralized peer-to-peer and avoiding swarm demo will be displayed as always. Make sure to read this blogpost of when we showed this demo at IROS 2022 to understand what is fully going on!

Also take a look at our event page of the ICRA 2024 demo.

Hand in your Crazyflie posters at our booth!

We will be providing a ‘special disposal service’ for your conference poster! We would love to see what you are working on and get your poster, because we have a lot of space in our updated office/flight space but a lot of empty walls.

If you hand in your poster at the booth, you’ll get a special, one-of-a-kind, button badge that you can wear proudly during the conference! So we will see you at booth IC085!

Prototype of forward facing expansion connector

2024-04-29 | Marcus | 2 Comments

A great feature of the Crazyflie is its ability to keep evolving, both by using software but also through hardware expansions. Hardware expansions allow us and our users to keep exploring new problems and doing new experiments, without having to change the flying base. Over the years lots of new decks have been released and we’ve seen lots of users building their own custom electronics and attaching it to their Crazyflies. Although very versatile, the current deck system is limited to up/downwards facing expansions. Adding electronics that face forward, like a camera, has been harder and has required additional mechanics. Over the last couple of months we’ve been experimenting with a new expansion connector aiming at solving this issue. The idea is to be able to add a new class of expansions facing forward. This week’s blog post and next week’s developer meeting are about these experiments.

Goals and design

We’re always trying to find ways to make our platform more versatile, making it easier to expand and to be used in new ways. So we’ve been looking for a new way to be able to expand the platform even more, this time with electronics facing forward instead of up/down. The goal is to easily be able to add things such as cameras, ranging sensors etc. Making a custom deck with custom mechanics for each sensor hasn’t been a good solution, it takes lots of time and it doesn’t enable our users to do their own custom electronics. Finding a generic solution is hard since we’re constrained both in space and in weight. We need a solution which is very small and light, each gram adding cuts into the flight time. The solution also needs a way to handle the data generated from cameras/sensors as well as possibly to stream it over a faster connection than the Crazyradio.

Our current prototype is made of two parts, a new deck with WiFi and more computational power as well as several smaller expansions which can be added to it. The expansions fit straight into a small right angle connector, making it easy to change boards. The current connector we’re testing has 30 positions, of which 6 is used for power and 1 for 1-wire, leaving 24 pins for signaling. The 1-wire works the same way as our current decks, additional added hardware is auto-detected at startup and can be used without recompiling or reconfiguration.

The current prototype uses an ESP32-S3 and weighs in at 3.7 grams. Added to this there are a number of expansions that we’re evaluating:

OV2640 + VL53L5CX: RGB camera and ranging sensor (1.6 grams)
Flir Lepton 3.5: Thermal camera (2.1 grams)
MLX90640: Thermal camera (2.0 grams)

So the current prototype with RGB camera (and ranging sensor) weights in at a total of 5.3 grams (0.9 grams more than the AI deck).

Current status and continuation

We’re currently experimenting with connectors, modules and dimensions. In the coming months we will try to get more flight time to test the solution and we’re hoping to get some feedback from our users. So please post any comments and/or suggestions you might have.

If you’re interested in knowing more and discussing this then join our developer meeting next week on Wednesday. We will also be showing off the prototypes at ICRA, so make sure to swing by the booth if you’re attending.

CrazySim: A Software-in-the-Loop Simulator for the Crazyflie

2024-04-15 | Christian Llanes | Leave a comment

This week we have a guest blogpost by Christian Llanes, a Robotics PhD of from Formal Methods & Autonomous Control of Transportation Systems Lab of the Georgia Institute of Technology. Enjoy!

Why do we need simulators?

Simulators are one of the most important tools used in robotics research. They usually are designed for different purposes with different levels of complexity. For example, simulators with low computational overhead that are parallelizable are mainly used for either training reinforcement learning algorithms or Monte Carlo sampling for verification of task completion in a nondeterministic environment. Some simulators also use rendering engines for the graphical display of models and the environment or when cameras are intended to be used in the robotics platform. Simulation is also useful for the development and deployment of new robotics firmware features where the firmware is compiled on a test machine and run in the loop with a simulated sensor suite. This simulator configuration is known as software-in-the-loop (SITL) because the vehicle firmware is intended to be run in the loop with the simulated vehicle physics and/or rendering engine. This feature is supported by autopilot suites such as PX4, ArduPilot, CogniPilot, and BetaFlight. This feature is not officially supported yet for Crazyflies because it requires a large overhaul of the firmware to be able to compile on a desktop machine and interact with different simulators such as Gazebo, Webots, PyBullet, CoppeliaSim, Isaac Sim, or Unreal Engine.

CrazySim

Last summer I began working with Crazyflies and noticed this Crazyflie simulator gap. I stumbled on a community-developed project for Crazyflie SITL called sim_cf. This project is exactly what I was looking for. However, the firmware used by the project is from July 2019 and the official firmware has had over 2000 commits made since then. The project also uses ROS 1, Gazebo Classic, and doesn’t support the Crazyflie Python library (CFLib). Using this project as a starting point I set out to develop CrazySim–a Crazyflie SITL project that doesn’t require ROS, uses Gazebo Sim, and supports connectivity through CFLib. Using CFLib we can connect the simulator to external software such as Crazyswarm2 or the Crazyflie ground station client. Users test their control algorithms in the external software using the simulator interface before deploying to real flight hardware.

Using the Crazyflie Client for PID Tuning

We have also provided a modified Crazyflie client for CrazySim support. The Crazyflie client is a cool tool for testing a single drone in hardware. We can perform command based flight control, look at real time plots, save log data, and tune PID values in real time. The PID values are typically tuned for an out of the box Crazyflie. However, when we modify the Crazyflie and add extra weight through batteries, decks, and upgraded thrust motors then the behavior of the Crazyflie will change. If a user wants to tune a custom Crazyflie setup, then they can add additional models in this folder with their own motor and mass properties. Then they just need to add it to the list of supported models in either of the launch scripts. There is already an example model for the thrust upgrade bundle. Documentation for installing the custom client can be found here.

PID tuning a simulated Crazyflie using CrazySim on the Crazyflie PC client.

Crazyswarm2

We can now connect to the simulated Crazyflie firmware using CFLib. Therefore, we can set up a ROS 2 interface through Crazyswarm2 for swarm command and control through ROS 2 topics and services. To do this we first startup the drones using any of the launch scripts.

bash tools/crazyflie-simulation/simulator_files/gazebo/launch/sitl_multiagent_square.sh -n 16 -m crazyflie

Then, we bring up Crazyswarm2 after setting up the configuration file for the number of drones chosen.

ros2 launch crazyflie launch.py backend:=cflib

We demonstrate an example of how we can control a swarm of drones using Crazyswarm2 GoTo service commands.

Crazyswarm2 GoTo service commands using CrazySim.

ICRA 2024

CrazySim is also being presented as a paper at the 2024 IEEE International Conference on Robotics and Automation in Yokohama, Japan. If you are attending this conference and are interested in this work, then I invite you to my presentation and let me know that you are coming from this blog post after. For the paper, I created a multi agent decentralized model predictive controller (MPC) case study on ROS 2 to demonstrate the CrazySim simulation to hardware deployment workflow. Simulating larger swarms with MPC may require a high performance computer. The simulations in this work were performed on an AMD Ryzen 9 5950X desktop processor.

Links

Other Crazyflie SITL projects:

Python bindings for the Crazyflie Firmware

2024-04-01 | Kimberly McGuire | Leave a comment

Today, we’d like to take the opportunity to spotlight a feature that’s been in our code base for some time, yet hasn’t been the subject of a blog post: the Python bindings for our Crazyflie firmware. You may have noticed it mentioned in previous blog posts, and now we’ll delve into more detail about what it is, how we and others are utilizing it, and what its future holds.

Schematized visualization of code within the Crazyflie

What are the Python bindings?

Language bindings, in essence, are libraries that encapsulate chunks of code, enabling one programming language to interface with another. For instance, consider the project Zenoh. Its core library is crafted in Rust, but it offers bindings/wrappings for numerous other languages like Python, C/C++, and so on. This allows Zenoh’s API to be utilized in scripts or executables written in those languages. This approach significantly broadens the functionality without necessitating the rewriting of code across multiple programs. A case in point from the realm of robotics is ROS(1), which initially created all of their APIs for different languages from scratch—a maintenance nightmare. To address this, for ROS 2, they developed the primary functionality entirely in C and provided wrappers for all other programming languages. This strategy eliminates the need to ‘reinvent the wheel’ with each iteration.

Rather than redeveloping the firmware in Python, our esteemed collaborators Wolfgang Hönig and James Preiss took a pragmatic approach. They selected parts of the Crazyflie firmware and wrapped them for Python use. You can see the process in this ticket. This was a crucial step for the simulation of the original Crazyswarm (ROS1) project and was continued for its use in the Crazyswarm2 project, which is based on ROS 2. They opted for SWIG, a tool specifically designed to wrap C or C++ programs for use with higher-level target languages. This includes not only Python, but also C#, GO, Javascript, and more, making it the clear choice for implementing those bindings at the time. We also strongly recommend checking out a previous blogpost by Simon D. Levy, who used Haskell to wrap the C-based Crazyflie Firmware for C++.

Where are the Python bindings being used?

As previously mentioned, the Crazyswarm1 & 2 projects heavily utilize Python bindings for testing key components of the firmware (such as the high-level commander, planner, and controller) and for a (hybrid) software-in-the-loop simulation. During the project’s installation, these Python bindings must be compiled so they can be used during simulation. This approach allows users to first test their trajectories in a simulated environment before deploying them on actual Crazyflies. The advantage is that minimal or no modifications are required to achieve the same results. While simulations do not perfectly mirror real-world conditions, they are beneficial because they operate with the same controller as the one used on the Crazyflie itself. In our own Crazyflie simulation in Webots, it’s also possible to use these same bindings in the simulator by following these instructions.

Three controllers (PID, Mellinger, and Brescianini), intra-drone collision avoidance, and the high-level commander planner have all been converted into Python bindings. Recently, we’ve added a new component: the Extended Kalman Filter (EKF). This addition is ideal as it allows us to test the filter with recorded data from a real Crazyflie and experiment with different measurement models. As we discussed in a previous blogpost, estimators are complex due to their dependence on chance and environmental factors. It’s beneficial for developers to have more control over the inputs and expected outputs. However, the EKF is deeply integrated into the interconnected processes within the Crazyflie Firmware. After a significant refactoring effort, these were added to the bindings by creating an EKF emulator (see this PR). This enabled Kristoffer to further enhance the TDOA outlier filter for the Crazyflie by emulating the full process of the EKF, including IMU data.

In addition to SITL simulation and EKF development, Python bindings are also invaluable for continuous integration. They enable comprehensive testing that encompasses not just isolated code snippets, but entire processes. For instance, if there’s a recording of a Crazyflie flight complete with sensor data (such as flow, height, and IMU data), and it’s supplemented with a recorded ground truth (from lighthouse/mocap), this sensor data can be fed into the EKF Python binding. We can then compare the outputted pose with the ground truth to verify accuracy. The same principle applies to the controllers. Consequently, if any changes are made to the firmware that affect these crucial aspects of Crazyflie flight, these tests can readily detect them.

If you like to try the python bindings tests for yourself, clone the Crazyflie-firmware repo and build/install the python bindings via these instructions. Make sure you are in the root of the repository and run: python3 -m pytest test_python/. Mind that you might need to put the bindings in the same path with export PYTHONPATH=<PATH_TO_>/crazyflie-firmware/build:$PYTHONPATH (please see this open ticket)

The next steps of the python bindings

We’ve seen how Python bindings have proven to be extremely useful, and we’re keen to further expand their application. At present, only the Loco positioning system has been incorporated into the EKF part of the Python bindings. Work is now underway to enable this for the Lighthouse system (see this draft PR). Incorporating the Lighthouse system will be somewhat more complex, but fortunately, much of the groundwork has already been laid, so we hope it won’t be too challenging. However, we have encountered issues when using the controller bindings with simulation (see this open ticket). It appears that some hardware-specific timing has been hardcoded throughout the PID controller in particular. Therefore, work needs to be done to separate the hardware abstraction from the code, necessitating additional refactoring work for the controller.

Recent projects like Sim_CF2 (see this blogpost) and Crazysim (see this discussion thread) have successfully compiled the Crazyflie firmware to run as a standalone node on a computer. This allows users to connect it to the Crazyflie Python library as if it were an actual Crazyflie. This full Software-In-The-Loop (SITL) functionality, already possible with autopilot suites like PX4 and Ardupilot, is something we at Bitcraze are eager to implement as well. However, considering the extensive work required by the aforementioned SITL projects to truly separate the hardware abstraction layer from the codebase, we anticipate that refactoring the entire firmware will be a substantial task. We’re excited to see what we can achieve in this area.

Indeed, even with a more comprehensive Software-In-The-Loop (SITL) solution, there’s no reason to completely abandon Python bindings. For developments requiring more input/output control—such as the creation of a new controller or an addition to the Extended Kalman Filter (EKF)—it’s beneficial to start with just that portion of the firmware code. Python bindings and a SITL build can coexist, each offering its own advantages and disadvantages for different stages of the development process. By leveraging the tools at our disposal, we can minimize the risk of damaging Crazyflies during development. Let’s continue to make the most of these valuable resources!

Parameters and System-id code

2024-03-25 | Tobias | Leave a comment

Dumping and loading persistent parameters to and from a file

We have a small quality-of-life update that will allow users to dump and load persistent parameters to and from a file that has recently been merged #PR443 and #PR706. A new persistent parameter management area is introduced to the parameters tab of the client, with buttons for dumping and loading persistent parameters, as well as clearing all stored persistent parameters from the Crazyflie. The persistent parameters are stored in .yml format, allowing for manual editing if desired. If you have any improvement suggestions please drop us a comment!

Crazyflie client screenshot highlighting new persistent parameter management area — *A new persistent parameter management area is introduced to the parameters tab of the client, with buttons for dumping, loading and clearing stored persistent parameters.*

Info dialog showing dumped persistent parameter names and their values — *An information dialog notifies users of the dumped persistent parameters and their values. Loading parameters will result in a similar pop-up*.

Clear stored parameters confirmation dialog — *A confirmation dialog prevents accidental clearing of persistent parameters.*

System-id code merged to master

Back in 2021, we created the system-id deck which we talked about in this blog post. It has not been officially released but a few users have gotten some PCBs and built it themselves. The functionality for the system-id deck has previously been in a branch, but as code in branches tends to become outdated, we have now moved this into the master branch utilizing the kbuild system instead. Building for the system-id deck is now as easy as doing “make sysid_defconfig” and then compiling. While talking about the system-id deck, let’s check the interest of releasing it as a product. It can help with system identification, tuning of controllers, improving efficiency etc. With enough interest there might be an economy in manufacturing it.

Out of stock

Unfortunately, we’re out of stock of Crazyflies at the moment. We expect some at the end of next week, so hopefully, you should be able to find them back in the store quickly.

Developer meeting

The next developer meeting will be on the 6th of March 2024, Arnaud will talk about the Crazyflie Client past, present and future, based on its last blog post. The client is still very useful but starts to show its age so we are looking at what should be kept, what should be improved and what should be removed. We will present what we have in mind, please come and discuss with us so we can shape the next 10 years of Crazyflie client!

Crazyflie client: past, present and future

2024-03-18 | Arnaud | Leave a comment

The Crazyflie client has a quite long history, like a lot of things in the Crazyflie ecosystem it has been started when the Crazyflie was used alone using mainly manual control. It has evolved to follow new use-cases of the Crazyflie but it still has traces of its origins and some limitation are still there with us. Moreover, the Crazyflie client and lib are written in python, one the main goal was to make it easily cross-platform. Unfortunately making a cross platform graphical program that accesses hardware in Python has proven to be quite challenging and we feel Python is not the way to go anymore. In this blog post we would like to discuss a bit the current state of the Client and what we are looking at for the future.

Photo of a Crazyflie quadcopter in front of a laptop running the Crazyflie client — A Crazyflie connected to the CFclient

The Crazyflie client was originally design to be able to fly, inspect and work with one Crazyflie. It still serves this purpose quite well: with the client you can connect your Crazyflie, observe graphs of internal log values in real time, setup different decks and sets and store parameters. It is a very good tool to explore and work with the Crazyflie.

However, it is only working with one Crazyflie at a time will take over the radio, so another script cannot talk to the Crazyflie at the same time using the radio. Unless the script uses the ZMQ API that allows an external program to control client and so to control the Crazyflie while the client is connected and active. This functionality can be very useful but, since it is disabled by default, has not seen a lot of use.

The worst, for us, in the current client state is Python and PyQT distribution situation: we used to have an easy-to-use installer on windows, a Linux Snap package and plan for a Mac App. All these have been pretty much abandoned because they kept breaking down over time and the support weight for us was too big. So the only way to install the client is via Pip, the python package manager. This means that you already need a well setup Python environment to run the client. That is nice unless you have a Mac or a Raspberry pi: the latest MacOS broke part of the client that prevents it to run and there is no PyQT build for raspberry-pi available on PyPi. This is the kind of paper-cut that keeps happening at regular interval with distributing a Python application and that we keep having to look at.

So, we have been looking at ways to improve the situation. The Crazyflie client is more than 10 years old now, so a rewrite would not be such a crazy thing to consider. There are at least 2 angle of attack to make a new client better suited for the next 10 years of Crazyflie development:

Multi-platform and distribution

Making a multi-platform program is and will always be challenging. However, we have discovered that doing it in a dynamic interpreted language like python is even more so. The main challenge come from the fact that things tend to break on the user side depending of the user configuration: we all run a slightly different version of python, python evolve and package management evolves as well, so when things break it breaks at random depending of how up-to-date a particular system.

One solution would then be to switch to a compiled programming language. This increases the probability of finding problem at compile time and not at runtime, which means that we will hopefully be the first to know then Apple decided to change the location of one fundamental piece of library and so allow us to hopefully fix the problem before any user is impacted (assuming we can run CI on the latest version of MacOS early enough…).

So, as you might have guessed, our current idea is to write the client in Rust. We are currently looking at Tauri for the UI which is Web based. We still have ideas of also making a web-client so having a web-based IU on PC would simplify development of that.

We are not letting Python go away though, the idea is still to support Python, but to use it for what it is good at: a great language for developer and experimentation. Rust has great bindings to python so in this plan, the python lib is backed by the Rust lib.

Modularity

The other side of the current client limitation is the fact that it connects one Crazyflie taking over the radio. We actually love using the client to observe and poke the state of a Crazyflie so it would be really great to use it when writing a script or controlling a swarm with Crazyswarm. The current ZMQ implementation was designed to solve this issue, but it goes at it the wrong way around: the client becomes the gateway to the Crazyflie and must always be ON. It would be much nicer to be able to launch the client to connect and inspect a Crazyflie currently control by a script.

One design we are currently looking at would be to use use a protocol like Zenoh between the client and the lib. Basically, when connecting a Crazyflie, be it from a script or from the client, a server would be launched in the background that would handle the connection. All communication would pass through this server and so multiple programs would be able to communicate simultaneously with the Crazyflie.

This would allow us to build easily bridges to ROS to get the client to communicate with a Crazyflie currently connected in ROS. And since ROS2 is working on supporting Zenoh officially a bridge might not even be required.

As an added bonus, the Crazyflie server idea would greatly improve the situation when it comes to supporting Virtual Machines and WSL on Windows: it would move the USB connection handling to a Windows program and only require fast network connections which is something that works really well on WSL or VMs.

Conclusion

In this blog post I have tried to describe our current challenges and some way forward we see. The main message though is that we want to change things when it comes to the client, if you have wishes or ideas now is a good time to get in touch. Let’s make the next 10 years of Crazyflie client problem-free.