Drone ready for take off

Before take off on one of the 38 flights

The Brief

We recently received a call asking if it would be possible to create a high-resolution map of a 21 kilometre cable route connecting a wind farm to the National Grid before work started to dig trenches and bury the cable.

Having successfully delivered drone survey data to clients with both small and large sites this would prove to be a new and interesting challenge stretching the capabilities of drone technology and our logistical planning abilities to the limit.

The goal of the survey was to create high resolution orthomosaics, essentially Google Earth on steroids.  The customer would use these orthomosaics to assess the condition of the cable route and to give them an up to the moment view of the condition of the land before the sub-contractors start work on the project.  The idea is then to use this imagery at a later date to defend our client against claims from landowners that they have failed to restore the land to the condition it was in before the work commenced or that they have caused damage to walls or fencing.

The Challenges

There were some big challenges to overcome on this engagement, not least of which is the fact that the law in the UK requires drones to be within visual range of the pilot at all times.  Our drones are more than capable of remaining under control whilst flying for several kilometres from the pilot – this is not permitted and in reality a distance of 500 meters from the pilot is the limit.

Fortunately, Drones on Demand holds an Extended Visual Line Of Sight (EVLOS) exemption on our Civil Aviation Authority (CAA) granted Permit for Commercial Operation, allowing us to fly up to 1.5 kilometres from the pilot if using suitably qualified observers and we are able to comply with strict safety conditions.  This allowed us to reduce the number of flights significantly and speed up the time to complete the job.

The second issue to overcome was the logistics of planning a significant number of flights at a location nearly 300 miles away and making sure that our drones were up to the job.



To plan the flights we used Drone Deploy’s new Corridor Mapping Beta flight planning app, and split the Google Earth KMZ of the route up into legally flyable, safe, easy to access sections.

We use Google satellite imagery and Ordinance Survey mapping to assist with identifying any potential hazards and the ground elevation changes and produce a risk assessment on each flight.  We then complete a further risk assessment when on site to make sure that there are no new hazards (or previously identified hazards have been removed).

The main hazard that we had identified but needed to further risk asses when on site were the high number of electricity pylons and power cables that criss-cross the area – we needed to make sure that we were flying high enough to remove the risk of flying in to them.

Fortunately the flights were all to take place in sparsely populated areas and our client had arranged permission from all of the landowners along the route to be flying from their property.  Before each flight we visited any properties that were on the periphery of the flight path to talk to them about what we were doing and reassure them of the intent of the flight – that really have no interest on what may be on their washing line!  We were always greeted with interest and often gathered a small crowd of spectators.

Data Processing

We completed the mapping flights over three days and then took a further three days to process and QA the data, delivering the route in 21 separate sections, mainly to make any data downloads manageable in size.

We handed the data over to the client in person and also provided training on how to use Drone Deploy’s online tools, demonstrating how to add annotations and measurements to the mapping and also how to get the most out of the Elevation and 3D tools that are available.

In numbers

Here are some statistics from our ‘longest’ mission yet:

Drones used:1 (+1 backup) DJI Phantom 4 Pro for those interested
Batteries available:8 (recharged over night)
Route length:21 kilometres
Number of flights:38
Flight time:4 hours 27 minutes
Distance flown:100.7 kilometres
Amount of data processed:56 GB
Hectares mapped:1122 (2773 acres)
Number of photos captured:6661

 What did we learn?

Sunburned Matthew

Don’t forget to reapply sun cream

  1. The time invested in thorough planning before getting to site was well worth it. Having risk assessments, pre-programmed flight plans, detailed mapping of the route and pre-identified access points saved hours on site.
  2. Take more supplementary oblique photographs of each wall, fence, hedge or road that the route crosses. This will produce more detailed 3D models and provide more reference images to revert to.
  3. Split a flight with significant elevation change into two or more flights or change the drone’s altitude at different waypoints. A 50 metre ground elevation change in a flight means that flying at 60 metres altitude results in images actually captured at 110 metres if you take off from the highest point.  This results in lower resolution deliverables for the areas where the drone is higher above the ground.
  4. Take more battery chargers. Having to wake up in the night to change the batteries on the chargers was not fun.  It would have also halved the charging time to fully charge all eight fully depleted batteries
  5. Don’t forget to reapply sun cream.


Did it deliver value for the customer?

It’s too early to say whether the data is good enough for the purpose it is intended as work hasn’t yet been completed.  The data is already proving useful for identifying some potential sites of archaeological interest along the route that hadn’t been spotted from walking the route or from Google’s satellite imagery.  The data has also highlighted some areas where the ground may potentially be rocky, requiring the route to deviate.  The imagery is also being used for contractor inductions and enhancing communication between the site and office.

Are we are able to rise to your challenge?