Project Everest

Proposal (Operational/Other)

Proposal Timor-Leste: Use of Drones to assess rooftops

In the handover documents, the December Energy Assessment Team mentioned that they used a drone to render a 3D model of Paradises. The plan for this was to see if this was a viable option for assessment of potential customers. As it turned out, this was not the case as it could take three days for the image to finalise and even one of the images needed to be taken back to Australia and was finished on a more powerful computer. Although the 3D rendering turned out to not be a viable option, with all the potential of drone technology and computer software, there had to be another option that could be explored that would achieve similar goals. Especially considering visual potential it offered for when pitching to potential customers.


So the first thing that was investigated was how could we create the same concept the December team came up with but in a significantly more viable way that was easier and faster to do. The idea of using photoshop or a similar program to create an image of the rooftop with the solar panels attached to scale. For a variety of reasons, the GNU Image Manipulation Program (GIMP) was chosen for this. By using a collection of photos from the drone at varying heights, locations and camera angles a scale was able to be determined which allowed the area of the roof to be determined as well as a test to see if the solar panels would fit. This process would be sped up significantly if the schematics of the building are provided but in the end this is not necessary.  Below is an example of this using a section of Paradises Rooftop with the GPV8 (Small system).


While this image is not as high quality as the 3D rendering, it has the possibility of being completed on the site of assessment rather than being required to come back and be processed. This will allow a visual representation of what we are trying to sell available at the time of pitch. The image will be created by photoshopping the solar panels to scale onto the photos of the rooftop, to give a real-time idea of what the roof will look like with the solar panels.


Following on from the visual analysis of the rooftop, a google sheets document was developed to be able to complement and support the visual analysis by automatically calculating the scale of photos, the area of suggested rooftops for solar panels as well as the required solar panels for the energy usage. The calculation will also test if they have enough roof space for the amount of solar panels required. By completing this drone analysis and the calculations, accurate quotes, figures and examples can be given at the time of assessment. This allows a faster sale process without the requirement of multiple meetings, to finalise prices when a customer has confirmed definite interest in our product.



By building upon the December Team’s initial ideas of using the drone, we have been able to come up with a process that can both significantly increase the strength of our pitch to customers as well simply concepts that some may struggle with.

edited on 2nd July 2018, 10:07 by Wade Tink

Matt Allan Jan 13, 2018

Have you considered AWS to overcome your compute power constraint?

Reply 0

David Gailey Jan 18, 2018

Hi Jonathon
Great to see the evolution of the drone idea. I was a member of the December team that originally devised the idea and its interesting to see how others tackle the problems that stumped us.

I agree with your concerns about render time and feasibility of 3D models but thought I’d update you on my experimentation back in Australia. Using a faster computer and a better technique, I have been able to create reasonable models in as little as three hours.

To continue the aviation theme which permeates through this concept, the computer I used in Timor that took three days can be likened to the Wright Flyer. In this sense, that first three-day render like the Wright Flyer, was slow and achieved relatively small results. But it proved the concept possible.

Really, we just need to consider the roof and any obstacles that could impact the lighting. This is easy enough to do with a basic model. Plus, given how long a physical assessment of a building and its electricity usage would take, is a few days really too long?

Certainly, while there are issues regarding feasibility, I think with experimentation and refinement, 3D models may still be a feasible tool for future teams to use particularly if we consider cloud-computing and access to mid-range hardware.
Food for thought and good luck with remainder of project.

Reply 1

Wade Tink Jul 2, 2018

Has this been developed? Is this part of the current Energy operation?

Reply 0

Jonathon Anderson Jul 2, 2018

This was a major idea by the January team for the assessment of rooftops for potential customers. In January it was only used for one customer which was CNEFP (the trade school). I am unsure of its current status and if the idea ia still being explored.

Reply 1

Brett Jenkin Jul 3, 2018

Hi Tink, with the shift in focus towards the Rural Expansion Plan and the distribution of small scale equipment over the large scale business targets, the Timor Group as a whole has not included a drone in the resource requirements for July. Subject to upcoming discussions with Hugh Collett, we may explore alternate assessment options with the Actemium Student Solar Technicians.

Reply 0

View all replies (2)

Wade Tink Jul 2, 2018

Status label added: Proposal (Operational/other)

Reply 0

Alice Horton 3 months ago

Project Everest is the big and the brand name in the hiking and the mountain climbing because they have all the resources to so this. The project Everest venture is full of the hilly graphics by the best resume help use of the droned to access the rooftops and to do it the most.

Reply 0

MariaJGoodrich 3 months ago

Thanks for the nice post.
[url=]a review[/url]

Reply 0

greentech5 3 months ago

good for all readers Top 10 Digital Currencies

Reply 0