Scanning in large, feature-poor outdoor areas is one of the most difficult scenarios for Hovermap. The key to a successful scan is effective mission path planning. Start the mission in the centre of the target area or the area with the most features, then follow a path that avoids large open spaces. Examples of suitable flight paths include an outward spiral or a grid-type mission. Close the loop where possible, and ensure plenty of overlap between scans.
The recommended process is as follows.
What you will need
A drone with Hovermap mounted.
Emesent Commander installed on the tablet, with a working connection to Hovermap.
Aura installed for processing the scan data after capture.
Ground Control Points (GCPs) for georeferencing the scan.
Familiarity with the available flight modes (Mapping, Pilot Assist, Autonomous).
Procedure
Step 1: Plan the mission path
Start in the centre of the target area, or the area with the most features, to begin the process with a good quality scan.
If scanning a large area, break it into smaller scans by unique assets. Ensure there is at least 20% overlap between each scan, which helps Hovermap keep track of its position.
For each smaller scan, close the loop as often as possible. For more information on closing the loop and scan path strategies.
Give careful consideration to where the GCPs are placed. They are extremely helpful when aligning smaller scans during merging. Always place at least two GCPs in the overlap area between adjacent scans.
If particular areas are of high importance, consider performing additional walking scans so that the area is sufficiently targeted. These walking scans can then be merged with the flight data.
The GCPs referenced here are manually placed survey-controlled ground markers, not Emesent's automated GCP system.
Step 2: Decide on mission parameters
Choose the mission mode to use, based on the environment, the level of automation required, and the platform compatibility.
Mode | Description |
|---|---|
Mapping | Hovermap operates as a passive payload while a third-party mission planning application controls the platform. Hovermap has no control over the platform. |
Pilot Assist | Hovermap provides collision avoidance protection from obstacles such as trees. Flying a regular pattern can be more challenging, but this mode is highly adaptable to unforeseen conditions. |
Autonomous | The easiest way to scan a target area. Hovermap always faces forward, so taking over control occasionally to focus on an area of interest may be necessary. |
Step 3: Prepare the environment
Place the GCPs carefully across the target area in the positions decided during mission planning.
Step 4: Complete pre-mission checks
Complete the required pre-mission checks.
For background on how Hovermap automatically determines the most suitable navigation method based on the environment, see How does Hovermap navigate?.
Step 5: Perform the mission
Perform the scan in accordance with the planned mission path.
Fly as low and as slow as possible. Keep below 40 m flight height, and keep speed below 5 metres per second. Stay 5 m to 10 m above the tallest obstacle for that particular flight.
Focus specifically on the GCPs during the scan to ensure they are captured cleanly.
While scanning, check the live point cloud in Emesent Commander to ensure the target area is covered sufficiently.
The flight height varies by environment. In feature-poor environments, the maximum flight height is significantly lower. Experiment with flight heights in similar environments before committing to a specific height.
Step 6: Process the scan data
Process the scan data in Aura. Select the Standard profile for the first job.
If the results are not as desired, change the processing parameters. For large outdoor areas, the recommended profiles are Low Features or Forest.
To merge the data, start with the most accurate individual scan, which is ideally the scan from the centre of the target area. Align all other scans to this central scan. Starting with the most accurate scan avoids accumulated errors.
These parameters work more often than not, but results still depend on the specific environment. If the result is not as expected, adjust the remaining settings. Be aware that tuning beyond the recommended parameters significantly increases processing times.
Outcome
The large outdoor area has been scanned, the individual scans have been merged into a single point cloud, and the data has been processed in Aura.
Related documentation and support
For assistance, contact the regional Emesent partner or the Emesent Client Support team.