The SWARM UAV network

[Johnathan Phan]

I have started the following project and was wondering if people were willing to help.

Specifically looking for coders and hardware interface specialists.

http://wiki.london.hackspace.org.uk/view/The_SWARM_UAV_network

I wish to do the following.

1> Find people who want to help
2> setup a planning meeting to discuss options on how to achieve each of the goals outlined in the document
3> Plan out stages for testing and catch up meetings.
4> Assign responsibility to complete each task required to complete each stage.
5> rinse and repeat until I have built the necessary parts for skynet to destroy humanity.
6> sit back as humanity is annihilated and I finally get my Apocalypse

Regards

John

[Anish Mohammed]

Hi John,
 happy to give you a hand...
regards
Anish

--
Anish Mohammed
http://uk.linkedin.com/in/anishmohammed
@anishmohammed

--
 
 

[Tim Storey]

like it, happy to help
\t

> --
>
>

[chrisbob12]

Hi John,

I'm interested too. Probably not the best at coding, but have some hardware skills to bring to the party. What happens next?

Regards, Chris.

[Johnathan Phan]

Hi Chris,

Is there a specific part you wish to help with? Can you let me know a bit about what programing experience you have or what you would like to help with?

Regards

John

--
 
 

--
Johnathan Phan
ox-consulting

T: +44 (0)784 118 7080
jo...@ox-consulting.com

www.ox-consulting.com

OX CONSULTING Ltd is registered in England & Wales, number: 07113039, registered address as above.

The information contained in this email message may be privileged, confidential or exempt from disclosure under applicable law. If you are not the intended recipient, you are hereby notified that any use, dissemination, distribution or copying of this transmission is strictly prohibited. If you have received this communication in error, or if any problems occur with transmission, please notify the sender immediately.

[Johnathan Phan]

Hi Chris,

The next step is to organize a meeting in person or on-line of all the active members. (I prefer the first meeting to be in person). Have everyone share there experience and explain to everyone else what they are interested about in the project. Specifically what part of the project they are particular interested in.

I am currently asking this information provably throughout direct emails so I can gauge if we have enough expertise to start the project. To give you an example of why this is necessary, please find a rough agenda for the first meeting below.

1> Individual Introductions
2> Agree finer detail of the starting technical objectives. Ie. what counts as lifting an object. Having a specially designed box that is lifted? robot lifing the box un aided by humans. Or having the units already attached to the object and having the drones lift it from point a to point b. What kind of data needs to be recorded by the Queen or shared by the Queen to manage in flight refuling to stay continuously air born.
3> I can explain how Agile development works and how it will speed up development and make everything really clear on what need to be done.
4> Discuss existing projects like quaduino, AeroQuad, arducopter and highlight what platform the group will use for the base of the UAV based on the expertise of the group members.
5> Lay out the general tasks that will need to completed to achieve the technical requirements agreed in point 2.
6> Agree some basic tasks to be achieved by a specific date for the first sprint. Split the task amongst ourselves.
7> Give access to the private shared resources that will be used for the opensource project. This resource is to only be used for planning and sharing contact details. As well as outline who is doing what and what still needs to be done. so that any member of the team knows the status of the project with out update meetings and can induct new contributing members. All of the code will be in git and opensource. IE. everyone can access it on the net. Explanations and guides will be posted on the hacker space pages. Such as meeting minutes and so on.
8> Exchange contact details and support one another to learn and push the project forward.

I hope this gives you an idea of the steps forward. There is a lot in this project. To give you an example, lets just use the use case of in flight refueling. How will that agreement be made between the Queen and other Queens to manage a set number of drones. IE. humans should never send instructions to drones to move to point a or point b. these should be dictated by the Queen to decide how to achieve the objective. If thats the case, what information will the Queen need to know about each drone to be able to achieve this objectives. This suddenly creates a type system for drones that allows it o make decisions. Drones them selves work out how to achieve the goal they have been set. for example a Queen will be told that there is a 4 kg object to lift at GPS location a which needs to be moved to location b with the dimensions xyz. Thats 6 metrics it needs to use to work out if 4 or 5 drones with a set lift ability is able to carry. The more adaptive the system, the more simple by diverse you need to make the system.

I am not only really waiting for 1 or 2 more people before setting up the first meeting.

Regards

John

[Chris Brasted]

My coding experience is limited, although, I have a spare Arduino, and can have the tool chain and limited experience with flashing PIC chips from c code.

I have lots of hardware building experience, making up electronics, working with plastic, wood and metal, and as a systems integrator.

Regards, Chris

--
 
 

[Chris Brasted]

Mostly, I'm interested in fabrication. Coding skills are rusty, and
I'll be surprised if you can't find someone who can do it better and
faster. I do have a systems integration background and some insight
into a-life and artificial intelligence tricks. Hardware, sensors and
modelling (in maths) is more my thing.

To be a bit more specific, here's two I made earlier:
www.boutiquerobots.com
www.vimeo.com/34744317

Regards, Chris.

On 1/16/13, Johnathan Phan <jo...@ox-consulting.com> wrote:
> Hi Chris,
>
> Is there a specific part you wish to help with? Can you let me know a bit
> about what programing experience you have or what you would like to help
> with?
>
> Regards
>
> John
>
>
> On Wed, Jan 16, 2013 at 8:55 AM, chrisbob12 <chris...@yahoo.com> wrote:
>
>> Hi John,
>>
>> I'm interested too. Probably not the best at coding, but have some
>> hardware skills to bring to the party. What happens next?
>>
>> Regards, Chris.
>>
>>
>> On Monday, January 14, 2013 5:31:14 PM UTC, Johnathan Phan wrote:
>>>
>>> I have started the following project and was wondering if people were
>>> willing to help.
>>>
>>> Specifically looking for coders and hardware interface specialists.
>>>

>>> http://wiki.london.hackspace.**org.uk/view/The_SWARM_UAV_**network<http://wiki.london.hackspace.org.uk/view/The_SWARM_UAV_network>

> --
>
>
>

[Chris Brasted]

The meeting looks like a good proposal. Out of normal office hours for
me please. I have evening commitments which can be modified if
necessary, and am not available for the first w/e in Feb.

> --
>
>
>

[Johnathan Phan]

My bad Chris.

I noticed that I sent you the email about what you were particularly interested in on the group message. It was meant to be private.

anyway, just waiting for 1-2 more people and I will message you all individually to arrange a meeting.

Regards

John

[Johnathan Phan]

Anyone else interested, just need 1 or 2 more before we hold a meeting?

[Johnathan Phan]

Hi guys,

Been over a week so I think I will hold a meeting anyway and see where we are. Was hoping for 1 or 2 more people to join the project. However that does'nt seem likly at this stage.

Can you all send me a time and preferable location you all want to meet?

To nattow it down so I play a game of algorithmic reduction. Please make your recommendation fit int he following confines.

1> The time is a Weekend between 11am - 5pm
2> location set at the moment is the London hackerspace. Probably sit in the corner. However if you tell me which lines you take to get there. I could probably work out a mutually agreeable station that we all cross to get to the hacker space anyway. That way we may be able to meet at a costa or something to cover everything we need to in the initial discussion.

On Monday, 14 January 2013 17:31:14 UTC, Johnathan Phan wrote:

[Anish Mohammed]

HI ,
want to create a moodle ?
regards
Anish

--
 
 

[Johnathan Phan]

Maybe,

depends on how much time the project wil allow for that.

First we need to get a consensus on the technical objectives.

The moodle can come when we have the basic hubs and we can use the moodle to push to idea of creating different type of drones for the swarm. So a Moodle will only show people the basics of how the core network operates and the source code for how the command and control queens will work. As well as how to build the basics of a drone, how they command the drone to behave will be a modular section which the queen is aware of. That way they can code different modular behavior for each drone type they create and upload them to shared environments.

Regards

John

--
 
 

[Nigel Worsley]

> Was hoping for 1 or 2 more people to join the project. However that does'nt seem likly at this stage.

It may be unlikely, but it has just happened! I have ben meaning to reply for ages, but been rather busy.

The day job hardware design and embedded programming, I suspect that the project
will find a use for those skills somewhere :)

I was thinking about autonomous UAVs a while back, and the tricky bit is giving them
enough awareness of their surroundings that they don't fly into each other or fixed
objects. Collision avoidance, basically.

I am in West London, so pretty much any location is equally inconvenient for me, and
no particular preference on time if it is a weekend. Weekdays aren't good.

Nigle

[Tim Storey]

heh.
to complicate matters weekends can be problematic for me, well
specifically every other weekend.
if its a free weekend then any time is fine, Sunday evenings is always a
good time though.
Free weekends for the next 4 weeks are:
2-3 Feb
16-17 Feb
2-3 Mar

\t

[Anish Mohammed]

Hi guys,
 i am generally free over weekends, i am away from 3-10feb ( singularity univeristy). But otherwise should be around.
regards
Anish

--

[Oscar Marques]

Hi!

I am from Rio de Janeiro, Brazil and me and friends are creating a hackerspace here called Kernel 40 Hacker Clube.

https://www.facebook.com/Kernel40HC

I know Anish who was presented to me by a friend of mine, I am doing drones for 2 long years. I am very interested in participating here and I am studying SWARM Intel for UAV.

Lets talk more, maybe some day I can visit London HS and I invite everyone here to check our HS in Rio de Janeiro, just send a message to me.

Carnaval? Someone going to the best party in the world? :)

Many thanks.

2013/1/25 Anish Mohammed <anish.m...@gmail.com>

--
Oscar Marques
osc...@gmail.com
http://www.dunkelheit.com.br
@f117usbr
+55 21 9293-9343

[Big Will]

We might pop over for the Olympics Oscar.

If I can make the first meeting, I would be interested to learn more about the project and the timescales, etc.  I'm a telecomms network management software engineer by trade and could help with some of the programming and control/ status monitoring systems for the drones.

Currently, only Saturdays are good for me.

[Anish Mohammed]

Given my cover is blown, i would admit , i am giving a talk about drones at Hacker news meetup next thursday :)
regards
Anish

[Johnathan Phan]

Hi guys,

This is what I have so far. the following people want to attend the meeting to setup the preliminary standard and technical agreements. We will then divy up the work amongst ourselves and set time lines and date based on what we know.

Tim Story
Anish Mohammed
Chris Bob
johnathan Phan
Nigel Worsley
Big Will

Will update the hacker space page with the details agreed for Oscar and everyone else on the net.

Meeting time and location.

Based on the limited info I have so far. The best time for all of us to meet is on Saturday 16th Feb 2013 at 11 am. I suggest we meet central London as everyone from all corners of London will be able to get in. This is not set in stone. If there is an objection to it let me know and I will propose an new date based on the new information. This is only to set a start point and get things moving.

So a neutral site I recommend is a Costa Coffee in central London. The address is below.

39 Old Compton St, London Chinatown, London W1D 5JX

The main reason I choose this place is because all major lines, weather it's central, northern, Piccadilly, Bakerloo, Victoria all have a station that stops very close this this one coffee shop. Making it probably the most neutral location possible. They have a large set of tables and chairs in the basement.

Meeting Agenda

1> Introduce our selves and give a little info on why we are there and what we are interested in.
2> Agree technical requirements. (IE. how intelligent the first Queen/drones will be. What counts as intelligence)
3> Outline what makes us different from other drone projects and why we are starting this one. (This should be quiet obvious as no one has tried to set standards in drone communication and swarm behavior. all current projects currently focus on controlling one drone for run to perform tricks or programing multiple drones to behave in a set way for show)
4> Agree technology stack to achieve technical requirements. (review how much of other open source projects we will cannibalize)
6> Divy up the tasks and agree a timetable.
7> Set roles and responsibilities between us. (there  maybe a situation where your asked to mentor or introduce new members to the code and explain what is going on.)
8> Exchange contact details where necessary.
9> Give access to all resources. IE. Jira, confluence, git (although git is open to all to download, I will impose strict change controls so no crap gets added or defacement occurs) and so on. As well as setup lines of communication between members of the core team.

Things that will be helpful if you do or read before our first meeting.

1> If you have a static ip address for accessing the INTERNET, bring it along to the meeting. I will need it to setup restricted access to our management system.
2> Read up about the following technology stacks and code languages used in the following existing projects.
a> http://code.google.com/p/arducopter/
b> aeroquad.com/
c> http://code.google.com/p/quaduino-ng/
d> http://www.scoutuav.com/
2> To make this clear, there are a few people out there that have dabbled in SWARM behavior in drones. However due to the nature of control. I feel the techniques they use as either cheating or simple. Please find a few good examples of similar projects. We are aiming for true SWARM behavior in the end. We may be starting off small to get the basic behavior in place, but the end result will be quiet complex.
a> http://singularityhub.com/2011/12/01/robotics-meets-architecture-%E2%80%93-50-quadcopters-will-autonomously-build-twenty-foot-tower-video/
b> https://www.grasp.upenn.edu/research/highlights

To give you an example, we will start off with the ability to refuel and keep them selves in the air continuously as one of the first technical requirements. other stuff like repair, defend the queen or switching out batteries in mid flight come later. One use case at a time. We still have to agree what info they share to make decisions on who does what. How decisions are made and how they will play out in real life and how roles are assigned in co operative behavior.

ps. All hail skynet

I am really glad that we managed to make the minimum requirement of 5 people for the first meeting. :)

thanks for you times guys.

[Anish Mohammed]

Hi John,
 Happy to give high level intro to arducopter, as i am on the dev list :(. If that saves any time...
regards
Anish

[Chris Brasted]

Hi John,

Thanks for organising this. I'll see you there.

Regards, Chris.

On Fri, Jan 25, 2013 at 5:21 PM, Johnathan Phan <jo...@ox-consulting.com> wrote:

--
 
 

[Johnathan Phan]

so at the moment, the first meeting of the group will be on the Saturday16th Feb 2013 at 11pm at 39 Old Compton St, London Chinatown, London W1D 5JX.

If you want to join the meeting let me know.

On Monday, 14 January 2013 17:31:14 UTC, Johnathan Phan wrote:

[Nigel Worsley]

> Saturday16th Feb 2013 at 11pm

Sounds rather a late start :)

Nigle

[Johnathan Phan]

ops meant 11am.

Are you able to come? If you are, can you pass me your mob number? So we can contact each other on the day if we have problems.

--
You received this message because you are subscribed to the Google Groups "London Hackspace" group.
To unsubscribe from this group and stop receiving emails from it, send an email to london-hack-space+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[Chris Brasted]

Civilised :)

[Necati]

Hi Johnathan and all 

I would like to help and involve in this project.  I'm planning to come to the first meeting. I will send the mobile number later.

Necati

[Anish Mohammed]

Hi John,
 i am assuming you still have my phone :). If not i have yours :)
regards
Anish

--
You received this message because you are subscribed to the Google Groups "London Hackspace" group.

To unsubscribe from this group and stop receiving emails from it, send an email to london-hack-sp...@googlegroups.com.

For more options, visit https://groups.google.com/groups/opt_out.
 
 

[Johnathan Phan]

Hi everyone,

Little update/reminder about the meeting.

People Attending

Tim Story - Attending
Anish Mohammed - Attending for now (However may not be able to attend. Dependent on work)
Chris Bob - Attending
Johnathan Phan - Attending
Nigel Worsley - Attending for now (However may not be able to attend. Dependent on work)
Will Gilchrist - Unable to attend. Work issues
Necati - Attending

So 6 people most likely

Date/location -

Saturday 16th Feb 2013 at 11 am.

Costa Coffee in central London.

39 Old Compton St, London Chinatown, London W1D 5JX

Agenda -

1> Introduce our selves and give a little info on why we are there and what we are interested in.
2> Agree technical requirements. (IE. how intelligent the first Queen/drones will be. What counts as intelligence)
3> Outline what makes us different from other drone projects and why we are starting this one. (This should be quiet obvious as no one has tried to set standards in drone communication and swarm behavior. all current projects currently focus on controlling one drone for run to perform tricks or programing multiple drones to behave in a set way for show)
4> Agree technology stack to achieve technical requirements. (review how much of other open source projects we will cannibalize)
6> Divy up the tasks and agree a timetable.
7> Set roles and responsibilities between us. (there  maybe a situation where your asked to mentor or introduce new members to the code and explain what is going on.)
8> Exchange contact details where necessary.
9> Give access to all resources. IE. Jira, confluence, git (although git is open to all to download, I will impose strict change controls so no crap gets added or defacement occurs) and so on. As well as setup lines of communication between members of the core team.

If anyone is not able to attend, please let me know. If you now know if you can attend, please confirm, so I can update in information. If anyone wants to add anything to the agenda, please let me know.

Regards

john

[mentar]

I'm interested in the project as I'm working on something similar using trirotor UAVs with ROS for reconnaissance, however I've got my hands full at the moment. I would like to attend the meeting with the understanding that I can't commit dedicated resources as of yet. But may be able to help in the future.

Let me know if you want me to come.

Cheers
Mentar

[Anish Mohammed]

HI John,
 i shall try my level best to join :)
regards
Anish

[Chris Brasted]

Dear UAV swarm team,

Regrettably, I found that I'm double booked on Saturday and I have to be somewhere else. I called John to talk it over, and have held off emailing till I had a little offering for discussion: I've attached some lightly structured notes from the recent NESTA breakfast meeting, 'Garden Drones' - not sure how much in it will be new, or even if you have a device to read it in a timely fashion, but hope it amuses.

However, I'm very keen to be part of this project and hope to catch up in the near future!

1> Introduce our selves
Hi I'm Chris, I have a decade and a half experience working in engineering and manufacturing, and now work in healthcare management. I like making stuff, and am reasonably competent fabricating in wood, plastics and light metal. I've done hobby electronics since 1978, and soldered my first SMDs a couple of years ago, as well as flashing my first PICs. I have some coding background but I wouldn't be my first choice. I did a degree in aeronautical engineering a long time ago, and graduated in physics in the 90s; crossed over to the dark side with an MBA more recently.

I'd like to have a personal mini drone swarm for pure bling purposes.

2> Agree technical requirements
My assumption from the parts list was that the Queen (c'mon, mothership!) would be a tethered blimp. In terms of energy capture and expenditure, I don't think we could rely on it being in a permanent powered hover. If it's urban, I wonder if it would be something that perched on tall buildings and did something with waste heat at night. I'd go for the drones being as small as is consistent with a useful payload, such as a pollution monitor. Need to do the maths to optimise design for endurance, I imagine.

How intelligent will the first Queen/drones be. I think Arducopter is a good start: simply reacting to the environment and not getting stuck in trees would be good.

3> Outline what makes us different from other drone projects and why we are starting this one

I leave that with you; even if we copy something, I think it'll still be pretty cutting edge.

4> Agree technology stack to achieve technical requirements

The more technical challenges we set ourselves, the more I think we should rely on standard and proven modules. Define the mission spec, then optimise the agents to achieve it.

6> Divy up the tasks and agree a timetable

From the original proposal, I think there are two projects:
a) unaided recharging
b) credible swarming to achieve a goal

John's put a technology stack together, but that needs to move on to a Bill of Materials, and hardware design. I think the mission spec needs to be nailed down pretty firmly at an early stage, as this will drive the cost envelope of the whole thing. After that's agreed, a procurement phase, followed by build and code, followed by assembly and testing. Then we see if it does what we wanted. Outside of code management, how far do you want to go in documenting it?

In terms of sharing out work, how about the drone subsystem breakdown proposed in the attachment matrixed across as the number of agents agreed in the mission spec?

• Position orientation
• Comms
• Airframe
• Motors & rotors
• Energy source
• Control & intelligence

7> Set roles and responsibilities between us

Happy to discuss.

8> Exchange contact details where necessary

John has my mobile

9> Give access to all resources

No comment at the moment

Hope it's a productive meeting and hope to meet soon.

Regards, Chris.

[Antaeus]

Hi John,

Just had to email you again to say how insanely excited I am about this project. You're gonna need mechancial engineering big time, and I'd love to be on-board. I'm sure you've been busy since January and I'm wondering where you guys are at so far and when you are meeting again?

Regards,

Antaeus

[chrisbob12]

Hi Antaeus,

The project kind of went off-line after the meeting, but we put up a wiki page:
http://wiki.london.hackspace.org.uk/view/The_SWARM_UAV_network

I was keen to build a set of identical wheeled robot chassis, so I'll hold my hand up to the slow progress. To date motors have been specified and bought, low level control board design identified and control chips procured. Next tasks to check over PCB design and order a batch, source suitable wheels, design and print physical chassis.

The design should be consistent with ant weight fighting robots and one of the robot football classes to give them some future use.

For my own part, I had a longer than anticipated hiatus building a keyboard.

[Antaeus Wheatley]

Hi there,

In a way im glad its gone slightly offline so that I dont miss out on the fun! I have checked out the wiki and I like the overall plan. When are you next around hackspace? It would be good to meet.

I would like to prototype a phisycal chasis if that sounds like a plan? Have you thought of seteering mechanism control? Servo and drive motor or separate parallel drives to allow cornering maybe?

Im also keen to understand what you mean by " The design should be consistent with ant weight fighting robots and one of the robot football classes to give them some future use."
Cheers :)
Antaeus

--
You received this message because you are subscribed to a topic in the Google Groups "London Hackspace" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/london-hack-space/AET7G0iYNF0/unsubscribe.
To unsubscribe from this group and all of its topics, send an email to london-hack-sp...@googlegroups.com.

[invent_or]

Definitely some irony in a project to create a killer network of, er, killer flying drones going off-line...

N

[Chris Brasted]

Hi Antaeus,

I'm planning to be in the space on Sunday 1st September; probably afternoon, but I'll clarify later.

Regards, Chris.

[Antaeus Wheatley]

Hi Chris,

I'm outta london two weeks on holiday. Will try catchup when back.

Thanks,
Ant