Loomio
Sat 26 Nov 2016 2:34AM

About the Project - Why & Requirements

OL Oriol López Public Seen by 331

This is a Q&A about why this project is needed and the requirements

OL

Oriol López Sat 26 Nov 2016 2:45AM

  1. What does the missingmaps.org bring on the table compared to google maps and why did you choose it over google maps?

Missing maps is trying to map areas where there are no google maps. The setup hackathons, make the vectorization of the satellite imagery and then this information is placed in openstreet map. Later on google maps use this information to update their maps as well.

as an example, now I´m in Papua New Guinea for a project of drones for transportation, and the are where we want to operate, has super low resolution imagery. https://www.google.es/maps/place/Ihu,+Papua+New+Guinea/@-7.8986863,145.3873264,9453m/data=!3m1!1e3!4m5!3m4!1s0x68558c9e89cf4e03:0x7a8a8bd4b658fb63!8m2!3d-7.8960081!4d145.3945655

Luckily in this case I have decent satellite imagery from Ovi Maps - Digital Globe that allow me to find the good spots for takeoff and landing

loomio logo

As an example missing maps has mapped the full city of Lubumbashi https://www.google.es/maps/place/Lubumbashi,+Democratic+Republic+of+the+Congo/@-11.7123581,27.3942906,11z/data=!4m5!3m4!1s0x19723ea34874cbd9:0xa1c6f5a74f805b2f!8m2!3d-11.6876026!4d27.5026174

But there are still several areas without good mapping in need, were natural disasters or outbreaks happen regularly.

There are cases where there is no good satellite imagery and in those cases the use of drones for mapping offer great help. In addition mapping drones can be used in post desaster response, for emergency preparadness (flooding, patterns....), for seasonal studies... and then there is also the precision agriculture for what you add multispectral camera to for example get images in near infrared to easily evaluate the status of the cops, where you are missing water, fertilizer....

Missing Maps was started by MSF after Haiti earthquake when we had a huge epidemic of cholera, the people come to the hospital and you do not know where the patients come from. If you have maps, and they can tell you where do they come from, you estimate the population in certain areas and decide how better deploy your limited resources to avoid the expansion of the epidemics.

General Questions:
1. Would MSF benefit from extra characteristics added to the drone to create some distinction? As designing the drone for a disaster event, etc.

Having good maps avoid the situation explained above. Now mapping is not crazily expensive, but let´s say that you want to involve the geograpic university of Kinsasha to make mapping of certain remote villages to support missing maps. In this the cost is important, but also should be easy to use, easy to repair mapping drone that would make the difference.
Today you can make your own DIY mapping drone, the idea is to make an "iphone" that is easy to use and fabricate. My approach is to use an existing platform and modify it adding 3D printed parts and a specific and tested setup that we know it works. The system can be sold but you can also DIY and repair it with off the shelve tested and common components.

  1. How did you do the mapping so far? Please explain the process in details if possible. (Very important)

If no satellite images, most of the people is using the ebee drone, which was described in detail in the email sent to Shahab. Basially you select the area to map and resolution you need on a tablet and launch the plane which will make the route automatically. All this can be done in an open source way withouot much problem using mission planner and Pixhawk-Ardupilot controllers. The idea is to modify a plane platform to make an "open source ebee" equivalent

  1. Who is going to pilot and use the drone?

The person whom is going to map, is so simple to use and so small, that you don´t need a "drone pilot license" besides a short training with the basics. Off course when getting aerial imagery, permissions and community work is essential.

  1. Have you heard of QuestUAV? Why didn’t you go for QuestUAV in the first place?

We went with the ebee, but Quest UAV looks like a valid option, there is also the tuff wing, the trimble, delair tech.... in my previous email I put some examples. All those solutions are expensive. I don´t know Quest UAV but only the pelican box for transport looks very expensive. In MSF if we need to do a job we can afford to get something like this and do the job. The idea is to create a tool for missing maps open street maps - open drone maps that allow the capability to make maps to more people.

  1. What are their goals for this project, what do they want to achieve while working with us? What do you expect from us in terms of project completion?

The objective would be to provide a kit similar to the Quest UAV or the eBee, with some extra features. It would base it in one of the several commercial platforms ( I put some names in the previous email): axn clouds fly and windrider eleBee2 are the 2 platforms I´m working on. the colleague whom started Missing maps is working base don the Skywalker X-5. From the Skywalker series there are sevarl interesting platforms.
- Improve the landing with distance sensor (the ebee is self destroyed after few of its "hard landings".

What I´m doing is replicating the design or parts of it in Solid Works and creating parts to specifically it components or eg: make the wings easily detachable.

BAD

Bogdan Alexandru Demeter Wed 7 Dec 2016 10:15AM

Hey Oriol,

Thanks for replying and creating the threads and topics for our conversation. Based on the information so far and by knowing that the drone "will be used for only mapping purposes", we decided to use Jeroen's proposal of a one wing drone.

This doesn't mean that the quadcopter is out of discussion, because there are indeed a few more questions that can still change our way of approaching, acting and choosing the ways and perspectives of this project.

We know we ask a lot, but it is not a simple project that can be covered just by desktop research. We ask you to understand us that this is also a school project, so somehow we need to introduce the design thinking approach and teachings so we can be graded on some sort of criteria.

Without further ado, these questions are the following:

 1. What is the average time for a mapping mission?
 2. What is the maximum surface the drone needs to cover during a mapping mission?
 2.1. What is the distance has to be covered to reach the area of mapping ?
 2.2 How does it get to this area ( is it brought there or do you always start in the same location ) ?
 3. In case of an emergency is there a need for implementing a procedure for emergency landing ?
  3.1 If so how often do such emergencies occur ?
  3.2. Do you have a requirement implemented by the government or you have to play by the laws and legislations in terms of sizing and power?
 4. What is the usual/most common location that the drone will be operating in? Specific countries/location/positions will already help us, since there are different forms of relief that we need to take into account. Is it an urban settlement/village that you also need to cover?
 5. Can we get in contact with a person or the group of people who will actually do the mapping, that will be interacting with the drone in the field?
 6. Do you want to have the drone controlled by a pilot or by a GPS system/app, or by both at the same time?
 7. In terms of flying capabilities, the drone can either have a big wingspan which will lead to a more gliding type of drone with not so powerful engine or a smaller wingspan, which means more speed, will get the job done faster, but probably in a lower quality. Which section do you tend to lean towards the most? 
 8. What software do you normally use for implementing the coordinates in the drone, and what software do you use to generate the vectors of the maps?
 9. In regards to funding, is MSF able to provide us with budget and connections with companies or businesses that can help us in this manner? If so, what is expected from us to fit in this proposal?

We will cordially wait for your reply as we understand you have other projects to take care of too.

With kindest regards,
Lucia, Jeroen, Thomas and Bogdan

OL

Oriol López Thu 8 Dec 2016 1:51AM

Sorry I´m not aware of Jeroen proposal is it Quest UAV?. If it´s a flying wing, it might be a good proposal, at least big part of the industry is going there.

  1. What is the average time for a mapping mission?

Is not about time but about covered surface and resolution (you fly higher, you get bigger area with more resolution) it all depend on the needs. Not the same to do mapping (low res is good) than do earthquake assessment (I´ve seen 3D models created with pictures where you can see the cracks, check for Skycatch Nepal).

  1. What is the maximum surface the drone needs to cover during a mapping mission?

There is no specific answer, depens on what you can map. I would say that you use the eBee as a reference, you have public info of it´s capacities. in Port Prince (Haiti) after the earthquake, the displaced population created new neighborhoods. An NGo used the ebee to map the area and they basically were limited by their batteries. The ebee, will tell you how many charges and flights you will need for a specific job, will automatically distribute the area in X flights.

2.1. What is the distance has to be covered to reach the area of mapping ?

Usually you go to the area and launch the plane. Look on youtube for eBee, you have tone of examples and you will see the full way of working

2.2 How does it get to this area ( is it brought there or do you always start in the same location ) ?

This is like the painter that starts very fast but end up slow because has to go back to where the paint pot is. The drone is small and transportable. Usually you launch from the area you want to map. Unless something is not allowing you to do this. In this case probably is neither a good idea to map.

  1. In case of an emergency is there a need for implementing a procedure for emergency landing ? The eBee weights 700-900gr and is foamy, they do reduce the risk this way. Other options use parachute, but it adds a lot of complexity. yes there are way. You should have a go back home and abort function to be able to activate from ground. This is all integrated in new control boards

3.1 If so how often do such emergencies occur ?
Depends on how well build and maintained is the drone.

3.2. Do you have a requirement implemented by the government or you have to play by the laws and legislations in terms of sizing and power?

You have regulations in each country, to fly drones, to get aerial pictures (spy). The less the weight the easier to operate. In UAViators you have a wiki with regulations for each country.

  1. What is the usual/most common location that the drone will be operating in? Specific countries/location/positions will already help us, since there are different forms of relief that we need to take into account. Is it an urban settlement/village that you also need to cover?

I don´t know, you want a polyvalent tool that you can use in different settings. Maybe you will not be allowed to fly in urban areas but then you forgot about drones. But if you are, what would be different?

  1. Can we get in contact with a person or the group of people who will actually do the mapping, that will be interacting with the drone in the field?

There is no specific person in the field who has to do this. This is a tool. But I can share you some examples of work done.

  1. Do you want to have the drone controlled by a pilot or by a GPS system/app, or by both at the same time?

In general today you want to design a full automatic tool that does not need a pilot.

  1. In terms of flying capabilities, the drone can either have a big wingspan which will lead to a more gliding type of drone with not so powerful engine or a smaller wingspan, which means more speed, will get the job done faster, but probably in a lower quality. Which section do you tend to lean towards the most?

Polyvalent drone. Look on the thread about flying platforms and see the commercial examples. Most NGO´s are using eBee today.

  1. What software do you normally use for implementing the coordinates in the drone, and what software do you use to generate the vectors of the maps?
    I made a thread of control where you have an introduction to all this. For the pos processing of the images you use open drone map

  2. In regards to funding, is MSF able to provide us with budget and connections with companies or businesses that can help us in this manner? If so, what is expected from us to fit in this proposal?

Commercial mapping drones are available, I find it difficult for MSF to fund this kind of innitiative. Open drone map would got funding external funding (not form MSF) but Steve V Mather have a working tool. I myself I´m doing this kind of stuff from my side. You would need to demonstrate good capabilities before requesting external funding, there is a ton of people who could do that and have a lot of experience with RC planes.