What has our intern, Zhouxin, been up to?

2020-03-09 | Zhouxin Ge | 2 Comments

This is it. The end of my internship. It feels strange to leave this unique office in a place called Malmö. My time spent here was more than just doing an assignment as part of a MSc. degree with the objective that I would gain working experience and contribute to a company.

My last day at the office of Bitcraze, Arnaud was already on parental leave

My time here gave me so much more. I have learned here a healthy way of thinking and problem solving which is part of the unique Bitcraze company culture. Next to that, it felt more like working with friends than just working with colleagues. Going to the office is a delight, as there is always humor, openness and honesty. I got to know everyone and enjoy the French, Swedish and Dutch-American hospitality and culture.

At this point you might think that I only have been drinking coffee and made sure that coffee in the office was not below level. Luckily that was not the case. I had the privilege to be the first user for a new deck. This deck has been in development for quite some time now and has been glossed over in some earlier blog posts. It is the yet to-be-released AI-Deck! At the moment the early-access AI-Decks are a delayed due to the COVID-19 virus. Bitcraze will update you on the blog when they know more.

My task within Bitcraze, in more detail, was to improve user friendliness of the AI-Deck by providing a framework for future users and at the same time to explore user friendliness of the whole ecosystem around the AI-Deck for an engineering student with beginner experience in embedded programming (e.g. me).

At the verge of giving the Crazyflie some AI capabilities, while being micromanaged.

So my mission began. A logical step was to see if the convolutional neural network from the PULP-DroNet project would run on the AI-Deck and fly with the Crazyflie, as the AI-Deck is an evolution of the PULP-Shield developed for this project. More information about this can be found here.

Unfortunately, this was not an easy feat as the PULP-DroNet project is using the pure version of the PULP SDK and an outdated autotiler. While the development partner for the AI-Deck, Greenwaves Technologies, uses the PULP SDK as a base with added functionalities in their SDK, which made it divert from the SDK used in the PULP-DroNet project.

Though, I was able to run the convolutional neural network in a simulated environment and compare this to the original DroNet that was implemented using Python and a Bebop. It was interesting to find out that the convolutional neural network of PULP-DroNet was behaving differently than the original DroNet in Python. There can be many explanations for this, but the main hypothesis is that this is caused by quantizing the network of PULP-DroNet from 32-bit floating point to 16-bit fixed point. In addition, the aforementioned network is trained on a larger dataset which included data created by a Himax camera.

A single Crazyflie obtained self-awareness and spun up a swarm of Crazyflies to gain world domination

While porting PULP-DroNet to the AI-Deck should be possible, the obstacles found along the way made it too troublesome and out of scope for my internship. So I moved on with the main objective, making a framework/example for the AI-Deck using the SDK provided by Greenwaves Technologies, which is called the GAP8 SDK. It contains a set of tools that should make the use of the AI-Deck easier, namely the NNTool and Autotiler tool. These tools make sure that you can automate the conversion of your neural network that is designed and trained in Python (Tensorflow and Keras) to a neural network code that can utilize the GAP8 functionalities.

My internship came to an end before I could overcome the last hurdle for a working example. To still bring this example to you, I have committed the doc/code I wrote and handed over the knowledge that I have accumulated throughout my internship when working with the AI-Deck and its environment to the capable minds of Kimberly and Tobias.

Along the way I have learned a lot about embedded programming and being a first product user. In addition with embedded programming and programming in general comes a different mindset than a conventional planning and deadline fixed mindset you get from university. With these valuable lessons in mind, I will be heading back to the TU Delft to start with my master thesis in either reinforcement learning for aircrafts or dense optical flow nets for quadcopters. Thank you Bitcraze for your time, experience and hospitality!

Drawing with the Crazyflie

2020-03-02 | Kristoffer Richardsson | Leave a comment

There has been some work done earlier to use the Crazyflie for generating images, for instance the dot-drawing by Paul Kry and light painting. I wanted to see if it is possible to put a brush or pen on a Crazyflie and use it to draw lines on a paper. I decided to use a fun Friday to try it out. The idea is simple: mount a pen on the Crazyflie, put a paper on a wall, write a script to draw a figure, fly!

The first thing I looked into was to investigate if a Crazyflie can fly with a brush or pen mounted on it. I wanted to keep the weight down and my initial approach was to use a cotton swab (0.6 g) dipped in paint. I found one that was long enough to extend in front of the propellers and I mounted it by squeezing it between the battery and the PCB. Flying was no problem with such low extra weight.

For positioning I decided to use the Lighthouse system. It is very accurate, simple to use and the easiest way to get started. I mounted a piece of cardboard in the YZ-plane of our lighthouse coordinate system, where I could attach a drawing paper. The idea of setting up the drawing surface parallell to the YZ-plane was to make the scripting easier. I (of course) used the Crazyflie and lighthouse system to measure that the cardboard was mounted at the right position.

Finally I wrote a simple python script that utilized the high level commander to move towards the drawing surface and yawing at the right position to draw a stroke on the paper. It sort of worked, but the cotton swab has to be “refilled” before each stroke which took a lot of time, and the results were a bit random.

I decided to try out a pen instead. The upside is that it does not require refill, on the other hand it is much heavier which makes the Crazyflie a bit sluggish when flying. I mounted the pen on the top side of the PCB, squeezed under the Lighthouse deck, and moved the battery to under the Crazyflie to distribute the weight.

Initial tests – both cotton swab and pen

The script was updated to draw the outline of the Bitcraze logo. I had a couple of variations where I tried to draw the full outline in one long stroke, as separate strokes, going up or going down and some other flavours.

So was it successful? Currently the Crazyflie is not a new Picasso, but the painting skills could maybe be improved with some more work. I think the main problems were:

the pen is too heavy and requires too much force on the paper
the controller cannot handle the situation in a good way. In essence I set the set point a few millimeters “into” the paper to push the pen against the surface which seems to be confusing as the controller can not reach the set point.
Flying that close to the drawing surface creates an air flow that disturbs the flight.

Video showing the Crazyflie drawing the logo

The Bitcraze logo (17×17 cm), drawn by the Crazyflie

Out of Control

2020-02-03 | Kimberly McGuire | 2 Comments

Two weeks ago, we had a blogpost about the state estimators that are available within the Crazyflie. So once the Crazyflie knows where it is, it would need to be determined where it wants to go, by means of the high level commander (implemented as part of the crazyswarm project) or set-points given by CFclient or directly from scripts using Crazyflie python lib. But exactly how would the crazyflie get to those desired positions in the first place? The differences between the current state estimates and the desired state, will need to be transformed to inputs given to the motors. Unfortunately, quadrotors like the Crazyflie do not have easy dynamics to maintain, so if you want to learn more, see this blogpost to read more about it!

So in order use the thrust of the motors in an useful way to get the Crazyflie to do what you want to do, there are several controllers to consider, which you can see on this quick overview here underneath. It shows the different control paths that can be taken from the high level commander all the way to the power distribution of the motors. Bear in mind that these are still simple representations and that the actual implementation is of course a bit more complicated, but at least it will give you a rough idea of which paths are possible to pursue.

PID Controller

So the default settings in the Crazyflie firmware is the proportional integral derivative (PID) control for all desired state aspects. So the High Level Commander (HLC) will send desired position set-points to the PID position controller (which used to be done off-board, so outside of the Crazyflie firmware before this blogpost). These result in desired pitch and roll angles, which are sent directly to the attitude PID controller. These determine the desired angle rates which is send to the angle rate controller (which is… you guessed… also a PID controller). This is also called Cascaded PID controller. That results in the desired thrusts for the roll pitch yaw and height that will be handled by the power distribution by the motors. (Note that height is mostly handled by the position controller)

INDI Controller

So the Incremental Nonlinear Dynamic Inversion (INDI) controller is an controller that immediately deal with the angle rates to determine the trust. This is a very new addition to the Crazyflie firmware by one of our community members and is based on the implementation of this paper. Currently, the position control is still handled by the same PID controller mentioned in the last paragraph, Nevertheless for handling the angles, it should be faster than the attitude and rate PID controller combined. We have not yet fully tested this out but if you do, let us know how you like it on the Bitcraze forum!

Mellinger Controller

As part of the Crazyswarm project, the controller designed by Daniel Mellinger has been implemented in the Crazyflie firmware as well. Please see this paper about the details of the Mellinger controller. It is a sort of “all in one”: based on the desired position and velocity vectors towards those position, it will calculate right away what the desired thrusts are that need to be distributed to all the motors. This results in a much smoother controlled trajectory of the high level commander and therefore advised to use when the Crazyflie has a precise position estimate (lighthouse and mocap). However, as it is so aggressive, any position estimate of a lesser quality (flowdeck or LPS) will not be sufficient for this controller. See some examples of mellinger controlled flights here and here.

Let us know what you think!

So do you have experience working with these controllers or want to know more about them, please drop us a message on the forum! We are currently working on stabilization and documentation of multiple aspects of the Crazyflie and the controllers is one of them, so we are really interested what your experiences are!

Welcome to our new team member, Barbara Rousselot

2020-01-27 | Barbara | Leave a comment

Hello there !

My name is Barbara, and I’m really proud to say I’m the new member of Bitcraze. As of now, I will be dealing with everything non-technical, so that the great minds here can focus on creating more of their amazing products!

If you want to know a little more about me, I should start by the beginning. I come from France originally, but have moved a lot over the years. From my early childhood in Tunisia to my semester abroad in Canada, I never really settled down… Until I came to Sweden, a little over 5 years ago. After having lived in Paris for some time, Malmö was a much needed breath of fresh air. In France I worked on a lot of different fields: receptionist, assistant, salesperson, teacher’s assistant, even as an actress. Those different experiences broadened my horizon and taught me a lot. Most of all, I learned to follow my passions: helping people, singing, and generally loving what I’m doing. My last assignment included community management and content creation in a video game company, but I’m happy to use every bit of my 5 years experience as an administrative assistant to be the office wizard here at Bitcraze.

When I’m not taking care of my family or working, I keep on following my passions. I could talk hours about movies or books (especially about Terry Pratchett books, or musicals, my favorite type of movies!). I took up knitting a year ago and can’t put my needles down now. And I consider myself a real geek. Video-games, movies, comics, I even learned some programming over the years (even tough I can’t compare with the other team members!). My greatest challenge last year was to build my own PC, and I have to say that I managed it with flying colours. Another way I past the time (while making everyone happy) is by baking, and every time a new batch of cookies gets out of the oven, I feel the weight of the family tradition: my grandfather was a french pastry cook, and I proudly follow his footsteps (and recipes!).

I’m really happy to join the Bitcraze team and excited about the future. I hope I’ll learn a lot from working with Arnaud, Kristoffer, Kimberly, Marcus and Tobias. My girly-geek side can’t stop jumping with excitement at working in such an impressive and stimulating environment.

Fosdem 2020

Arnaud is going to Fosdem this weekend, check out his forum post and say hi if you are there.

What’s up in 2020

2020-01-06 | Kristoffer Richardsson | Leave a comment

2020 has arrived! We are back after a long and nice holiday and are ready for new awesome year. So what’s happening in 2020 at Bitcraze?

Frequently we will go through a hectic period of releasing new hardware and adding new features that’s followed by a period of stabilization and maintenance. Since last fall we released a number of new products and added lots of new functionality to the Crazyflie, this spring it’s time to look back, finish up and make things more stable.

First up this year is to release a few of the products we didn’t manage to get out during the fall, the AI-deck and the Active marker deck. The first batches should be ready during the beginning of 2020 and we are really excited to see what our users will do with the new hardware.

Secondly we will be focusing on getting the Lighthouse deck out of Early Access. After the release last year we’ve added lots of functionality such as full pose using only a single base-station, but we’re still missing Lighthouse V2 support. In order to finish up we need to make the system easier to use and to add the V2 support.

The third thing we will be focusing on is stability and usability. We’re really happy about the ecosystem that exists around the Crazyflie platform. With lots of hard work and help/contributions from our community it’s become rather extensive over the years. But with the ecosystem and feature-set continuing to grow it’s not always easy to keep up.

The last thing we will be looking at is to continue adding more swarm-related functionality to simplify the task of flying multiple Crazyflies at the same time. Some of the features we’ve been discussing are: improved positioning, improved communication, drone health, trajectories and so on.

In order to handle the growing product portfolio and Crazyflie platform we’ll also continue growing the team during 2020. Hopefully there will more news on this already during the coming weeks :-)

With a long list of new customers, a growing ecosystem and lots of interesting things to come, we’re really looking forward to an exciting and fun year at Bitcraze! Oh, and as always there’s a few new hardware prototypes cooking… Happy New Year!

Looking back at 2019

2019-12-30 | Kristoffer Richardsson | Leave a comment

2019 is coming to an end and we are soon flipping the calendar to a new year. This is the last blog post of 2019 – time to look back and recap what has been going on during the year.

Community

We have had quite a few blog posts by community members this year. It is exciting for us to hear about the cool things our users are doing with our products, and we are happy to share them with all of you. If you have not read them yet and you might have some spare time during the holidays, it is well worth the time spent!

During 2019, we have also attended a number of conferences and events, where we have met a lot of interesting people, heard about amazing projects and got feedback from users. We attended FOSDEM (Belgium), ICRA (Canada), IMAV (Spain), ISRF (South Korea) and finally IROS (China).

Software

There have been quite a lot of improvements to the software in the Crazyflie ecosystem. Apart from bug fixes there has been some restructuring to simplify modifications and increase the utilization of system resources. The Crazyflie firmware has also been generalized to support multiple hardware platforms. We have added an app layer, Peer 2 Peer communication and support for new decks (see below).

The community has been buzzy contributing new and improved functionality as well as bug fixes to the software stack. Just to mention a few: support for new sensors, improved positioning support, better logging to SD-card, improved communication, new controllers and compressed trajectories. We can not express how grateful we are, thank you all!

Hardware

There have been quite some work on new hardware as well during 2019. We kicked the year off by releasing the two new platforms: the Crazyflie 2.1 and Roadrunner, and finished it by the recent release of the Crazyflie Bolt.

On the deck side there has been a focus on positioning support in the Passive– and Active marker decks that we have released in collaboration with our friends at Qualisys. The Lighthouse deck also falls into this category, we are excited about the performance and have high hopes of future awesomeness when it leaves Early access!

We have put a lot of work into the AI-deck during the year. Unfortunately we did not manage to finalize it 2019, but hopefully it should not be too long into 2020 until it is available in the store.

Documentation

Documentation is always hard, especially when the system is spread over many repositories. We have moved a fair amount of our documentation from the wiki to the code repositories to keep it closer to the code, and hopefully make it easier to keep it up to date. The documentation is now also published on the web to make it easy to access.

Logistics

We have tried out various 3d party shipping solutions earlier, but have settled on shipping our selfs, from our own warehouse in Sweden. This give us better control of the process and we have made a number of improvements and automated as much as possible to keep it lean and smooth.

Bitcraze

It has been an intense year for the Bitcraze family. We have moved to a new office with much more space and opportunities. It has required quite some work to set up labs, flight arena and other areas to our liking, but finally we have settled in and are very happy with the result!

Björn decided to leave the company in the beginning of the year, but on the other hand Kimberly joined in May! We have also had the great pleasure of hosing interns Victor and Zhouxin during the year.

On the system side, we have revamped our server platform for web, forum, wiki and internal services and are now using kubernetis. We also had a rapid increase in spam on the forum during the summer, but managed to counter it with better anti-spam tools.

Conclusions

It is a good exercise to look back and remember what we have done during the year. We are equally surprised each time we do this and realize all the things we have managed to squeeze in, only being 5 persons in the company! It has been yet another hectic year, but full of happiness and excitement.

Thanks for an awesome year!

Production Tests

2019-12-16 | Arnaud | Leave a comment

We are currently finishing production test design for a couple of expansion decks and we figured we never wrote about it and about the more general board production process. In this blog post we wanted to talk a bit about how we test boards in the productions phase, taking as an example the forthcoming active marker deck.

When finalizing an electronic board, we send to the manufacturer documentation that allows to manufacture & assemble a, hopefully, functional board. Although we assume that the individual components are in working order, the problem is that the assembling is not always perfect, so we need to check that everything we do is actually working,. This is what the production test is solving.

The first thing is to find out what to test, for that we need a strategy. The strategy we have been using is to test every step where we have modified or work on: for example we will test all the connections we have soldered in the manufacturing process. We will normally not test all the functionalities of ready-made module. For example, following this strategy, we will usually test all communication interface we have cabled, but we will not test all functionalities of a microcontroller we solder on the board, these are deemed to be already tested and working by the microcontroller manufacturer. This step usually end up with an annotated schematic:

Once we know what to test and roughly how to test it, we document a test rig that will be able to run the tests automatically. Some tests are generic and applicable to all our boards, for example we do test voltages with a multi-meter on every board that has a regulator. Some tests are very board specific. For example, for the active marker deck we want to test IR LEDs and an IR detector, we define a test rig that has reflector to reflect the LED to the detector and we will use the onboard detector to test the LEDs:

Simple block diagram of the test rig for the ActiveMarker Deck

We are normally using a Crazyflie on all our test rig, since it is usually possible to test all functionality from the deck port. We also try as much as possible to integrate the test software into the real software. For the active marker deck it meant adding 38KHz modulated output mode to the LEDs in order to emit a signal detectable by the detector, which will make it to the final firmware. Finally, we have a test software, running on the test computer, that uses the Crazyflie python lib to talk to the Crazyflie and run the test. The last step of all the test is to write the deck One Wire identification memory so that it can be detected by a Crazyflie.

Screenshot of the test program for the test engineer

From these specification, the manufacturer can then build a test rig and start testing boards, non-passing board will be re-worked until they pass or discarded.

Test rig for the Multi-ranger expansion deck

What we have learned in our years at Bitcraze is that testing phase is the most important part of the development process of PCB. Therefore, the earliest we already start thinking about the production tests in the board design, the more smooth the final phase of production of our new products will be.

Designing DroneChi

2019-12-02 | Joseph La Delfa | 1 Comment

This week we have a guest blog post from Joseph La Delfa.

DroneChi is a Human Drone interaction experience that uses the Qualisys motion capture system that enables the Crazyflie to react to movements of your body. At the Exertion Games Lab in Melbourne Australia, we like to design new experiences with technology where the whole body can be the controller and is involved in the experience.

When we first put these two technologies together we realised two things.

It was super easy to keep your attention on a the drone as it flew around the room reacting to your movements.
As a result it was also really easy to reflect on and refine ones own movements.

We thought this was like meditation meditated by a drone, and wanted to investigate how to further enhance this experience through design. We thought the smooth movements were especially mesmerising and so I decided to take beginner Tai Chi lessons; to get an appreciation of what it felt like to move like a Tai Chi student.

We undertook an 8 month design program where we simultaneously designed the form and the interaction of the Crazyflie. The initial design brief was pretty simple, make it look and feel light, graceful and from nature. In Tai Chi you are asked all the time to imagine a flower, the sea or a bird as you embody its movements, we wanted to emulate these experiences but without verbal instruction. Could a drone facilitate these sorts of experiences through it’s design?

We will present a summarised version of how the form and the interaction came about. Starting with a mood board, we collated radially symmetrical forms from nature to match a drone’s natural weight distribution.

We initially went with a jelly fish, hoping to emulate their “push gliiide” movement by articulating laser cut silhouettes (see fig c). This proved incredibly difficult, after searching high and low for a foam that was light enough for the Crazyflie to lift, we just could not get it to fly stable.

However, we serendipitously fell into the flower shape by trying to improve how we joined the carbon rods together in a loop (fig b below). By joining them to the main hull we realised it looked like a petal! This set us down the path of the flower, we even flipped the chassis so that the LED ring faced upwards (cheers to Tobias for that firmware hack).

Whilst this was going on we were experimenting with how to actually interact with the drone. Considering the experience was to be demonstrated at a major conference we decided to keep the tracking only to the hands, this allowed quick change overs. We started with cardboard pads, experimented with gloves but settled on some floral inspired 3D printed pads. We were so tempted to include the articulation of the fingers but decided against it to avoid scope creep! Further to this, we curved the final hand pads (fig d) to promote the idea of holding the drone, inspired by a move in Tai Chi called “holding the ball”.

As a beginner practicing Tai Chi I was sometimes overwhelmed by the number of aspects of my movement that constantly needed monitoring, palms out, heel out, elbow slightly bent, step forward etc. However in brief moments it all came together and I was able to appreciate the feelings of these movements as opposed to consciously monitoring them. We wanted this kind of experience when learning DroneChi so we devised a way of mapping the drone to the body to emulate this. After a few iterations we settled on the “mid point” method as seen below.

The drone only followed the midpoint (blue dot above) if it was within .2m of it. If it was outside of this range it would float away slowly from the participant. This may seem like a lot, but with little in the way of visual guidance (eg a laser pointer or an augmented display) a person can only rely on the proprioceptive feedback from their own body. We used the on board LED ring on the drone to let the person know at least when they are close, but that is all the help they got. As a result this takes a lot of concentration to get right!

In the end we were super happy with the final experience, in the study participants reported tuning into their bodies when using the drone, as well as experiencing a unique sort of relationship to the drone; not entirely like a pet and also like an extension of the body. We will be investigating both findings from the study through the design and testing of a new system on the Crazyflie. We see this work contributing to more intimate designs for human drone interactions as well as a being applicable to health contexts such as rehabilitation.

Community work on LPS

2019-11-25 | Marcus | Leave a comment

We have been traveling a lot this Autumn and have been talking to many Crazyflie users. It is always great to talk to our users and to get feedback about how what we make is being used.

One particular subject that stood out was the Loco positioning system (LPS): the LSP seems to be used by a lot of people and we have gotten quite some feedback about it, some are about things that work but also some things that could be improved. This is interesting because we normally do not get that much feedback from people using the LPS.

We released the LPS about 2.5 years ago with Two-Way-Ranging single Crazyflie support, and it has been improved regularly since then among other things by adding 2 TDoA modes that supports multiple Crazyflies as well as releasing the Roadrunner board, a standalone LPS tag.

If you are using the LPS, it would be great to have some feedback about what you are using it for, what works, what does not work and (even better ;) if you have any improvement that can be pushed to the community. Do not hesitate to post in this blog post, on the forum or by posting issues or pull request in the LPS-node or Crazyflie-firmware github projects.

Hello World, this is our new intern: Zhouxin

2019-11-18 | Zhouxin Ge | Leave a comment

Hey there, my name is Zhouxin. I was born in the Netherlands and I still live there, but not for the upcoming four months since I am going to be an intern here, at Bitcraze. I am really looking forward to contribute and to be part of this team! In this blog post I will share with you my motivation of interning here and something about myself.

I am doing this internship as part of my studies at the TU Delft, an university in the Netherlands, but the main reason is that I like the technical challenges related to their product and the dynamic work environment. I am convinced that I can learn a lot here about the practical things like working in a tech company and also about the technical challenges when developing code for practical applications such as the Crazyflie 2.1.

As mentioned before this is part of my studies, at the moment I am studying for my Masters degree in Aerospace Engineering. In this degree there are profiles. I choose the profile Control and Simulation which is mainly focused on the control and navigation systems in aviation. This still sounds quite general, so I will give a few examples where you can think of. A graduate from this profile might work on the automatic pilot of an aircraft, on the simulators for pilot training, on air traffic management systems, or on autonomous micro air vehicles. The latter is something I am interested in and that’s one of the reasons I am doing my internship at Bitcraze.

As a child I was always intrigued by how birds can fly, this led to my desire to fly. I had tried flying several times by wearing a cape and by jumping of the couch while trying to optimize the airtime with flapping my cape. This gave me some adrenaline boost but I never was able to fly. Later I discovered that the only way for humans to fly is to become a pilot. This became my new dream. When I was about 8 years old I started to practice flying by controlling RC airplanes. This made me interested in electronics and technology which later translated into pursuing my degree in Aerospace Engineering.

Besides my academic interests I also occupy myself with other activities. About two years ago I took a gap year and went traveling in East Asia. There I have discovered my enjoyment of nature and exploring cultures. Also I love to snowboard in the French Alps during the winter holidays and to share these wintersport adventures with friends. When I am not traveling or abroad, I enthusiastically play field hockey or tennis.