Recently, I completed one of my biggest mapping projects. We were tasked with mapping 4,000 acres in an urban environment in Nairobi, and the data was to be used for 3D modeling and to improve machine learning methods for accurate building reconstruction and height estimation.
At first, it looks easy. All you need to do is get to the field, launch the drone, and start mapping. But that was not the case. The actual flying took about four days, but the planning and approval process took at least a month.
As I took on this project, I learned a lot, from planning and executing a large-scale mapping project to what to expect when doing one. In this blog, I go into how I planned the project, handled the approval process, and what made the project a success.
Planning
Planning a project of this scale starts from the office before you get to the field. The client had provided a KML file of the area they wanted mapped. Once I had this, I proceeded through these steps;
- Confirming the type of airspace the area of interest falls under. This helps determine the types of approvals you will need and prepares you for the conditions of flying in different airspace types.
- Our area was close to an airport, so we knew there would be several low-flying aircraft.
- The next step is to check the KML for areas that we can fly from. These could be open fields, along the road, etc.
- Once we found the takeoff areas, we divided the KML into smaller sections of about 300 hectares, which were more manageable.
- Next, we checked which sub-county the area falls into to determine where to obtain local authority approval.
The Approval Process
Now, this process took longer than I had planned, delaying the project by almost a month. Since we were mapping a public area, we needed to engage the local authorities and explain why we needed to map such a large area. The client had selected this area because it had the highest concentration of buildings, which they needed for their research. Some local administrators were more receptive than others, which is understandable because drones flying in can raise several questions among the people who live there. Nonetheless, we obtained local authority approval, which enabled us to seek drone flight approval from KCAA.
Preparing for the Field
To accomplish the project, I prepared the following;
- I used one day to visit the takeoff areas selected to make sure they were open and could give enough visibility for Visual Line of Sight. I also changed some areas I had initially selected.
- A DJI Mavic 3 Enterprise drone. This was preferred because it features a mechanical shutter and a 20 MP sensor, which are ideal for high-quality, rapid mapping.
- The project required RTK for higher-accuracy data collection, so I used DJI’s D-RTK base station.
- To ensure continuous mapping, I secured at least 8 batteries. With each battery averaging 25 minutes, I was able to fly continuously and also recharge without halting operations.
- I also planned on the logistics of moving from one section to another. Having a fellow pilot, Kelvin Mbachu of who knew the area well, and local police officers for security and navigation also helped.
- I also monitored the weather using Windy and Dronecast to make sure I was prepared for any changes.
The Actual Mapping
After planning and gaining the approvals, It was time to start. The weather changed drastically on the first day; there was rain and fog, so we didn’t do any drone mapping. Luckily, I had requested approval for a few extra days. The second day went well, and we covered 1300 acres. We averaged the same number of acres and finished the whole 4000 acres in 3.5 days. The weather was perfect for the rest of the days, which helped. The major challenge we had was dealing with manned aircraft. We had to stop and give way several times, which slowed us down. We had also informed the nearest air traffic tower of the operations to keep them aware in case of any issues.
Challenges of Urban Mapping
These are some of the challenges of urban drone mapping I had to experience first-hand;
Buildings and Visibility
Since the area was close to an aerodrome, I was limited to flying at 300 feet (90 Meters). The issue is that in an urban environment, there are a lot of tall buildings, and you could crash into them if you are not careful. Tall buildings also block the drone’s signal, so you have to map small sections at a time and end up wasting more time when moving around.
Manned Aircraft Presence
There were many low-flying aircraft. Some had ADS-B, while others didn’t, so you always had to keep a listening ear out and give way. This halted our operations several times, but it was interesting to see it in action.
Finding Launching Areas
As a built-up area, it was difficult to find launch areas. We had to rely on public roads and any few open spaces we could find.
Weather
We had to halt our operations for a full day due to weather. But over the next days, the weather was favorable.
Battery Charging
As the day got hotter, batteries drained faster than we could get areas to recharge. For such a project, it would be ideal to have a portable backup charging station.
Data Validation
The client was responsible for processing, so my job was to ensure accurate data collection. Every day after collecting the data, I would transfer it to a backup hard disk and load it into processing software to make sure the data was okay and there were no gaps, and the whole area of interest was covered adequately. I made sure there were no blurred images, dark images, or images that weren’t part of the mapping area.
What I Learned From Mapping 4,000 Acres
Mapping 4,000 acres taught me a lot about what it actually takes to execute a large-scale drone mapping project.
1. Large mapping projects are logistics projects
The drone is only one part of the operation.
You also have to think about batteries, charging, transport, launch locations, approvals, weather, security, and moving between different sections.
2. Planning becomes more important as the area gets larger
A small planning mistake on a 10-acre project may cost you a few minutes.
On a 4,000-acre project, it can cost you hours.
Breaking the area into smaller sections and planning the launch locations beforehand made the project much easier to manage.
3. You have to be flexible
The plan you make in the office will not always work in the field.
We had to change some of our launch locations, stop operations because of weather, and pause several times to give way to manned aircraft.
You have to be prepared to adjust without losing sight of the bigger picture.
4. Local knowledge helps a lot
I was fortunate to have Kelvin Mbachu of Birds.co.ke, who knew the area much better than I did.
His familiarity with the area made moving between sections and finding suitable launch locations much easier.
Satellite imagery can tell you a lot, but it doesn’t tell you everything.
5. Data quality matters more than acreage
Covering 4,000 acres means very little if the data isn’t usable.
Every day, I backed up the imagery and checked it in processing software to make sure there were no gaps, blurred images, dark images, or areas we had missed.
The job isn’t finished when the drone lands. It’s finished when you know you have captured the data you need.
6. Pricing isn’t just about the acreage
This was probably one of the biggest business lessons for me.
It’s easy to look at 4,000 acres and calculate a price per acre. But the actual work involves much more than flying.
Approvals, site visits, transport, charging, extra days because of weather, ground support, and unexpected interruptions all affect the cost.
Going forward, I’ll be looking at the whole operation, rather than just the number of acres, when pricing large mapping projects.
A large project isn’t necessarily a profitable project if you don’t account for everything it takes to complete it.
Conclusion
Mapping 4,000 acres taught me that large-scale drone mapping isn’t simply about putting a drone in the air and covering as much ground as possible.
The difficult part often happens before the first flight: understanding the airspace, securing approvals, finding suitable launch locations, planning battery and ground logistics, and preparing for conditions that can change once you’re in the field.
The project also reinforced something I’ve learned from other mapping jobs: successful drone operations depend as much on planning and adaptation as they do on the aircraft itself.
Wow wonderful
Nice blog,