EVLOS Observer Placement with Terrain Intelligence
Place observers where the terrain allows — not where you hope it does.
Understanding EVLOS
What is Extended Visual Line of Sight?
EVLOS operations extend the pilot's situational awareness through a network of trained observers — but the terrain has the final say on whether that network actually works.
In standard Visual Line of Sight (VLOS) operations, the remote pilot maintains direct visual contact with the drone at all times. The pilot can see the aircraft, judge its distance, altitude, and heading, and take immediate action if something goes wrong. The applicable authority, operating category, and current official guidance determine what is required for each operation.
Extended Visual Line of Sight (EVLOS) pushes beyond that envelope. The pilot no longer needs to see the aircraft directly for the entire flight. Instead, one or more observers are positioned along the route, each maintaining visual contact with the drone within their segment and relaying information back to the pilot. The observer network effectively extends the pilot's eyes across distances and terrain that would otherwise be impossible to cover from a single position.
This sounds straightforward on paper. In practice, it introduces a dependency that many operators underestimate: every observer must actually be able to see the drone from their assigned position. If a ridge, tree line, or terrain depression blocks the view, the observer cannot fulfil their role — and the safety case collapses.
Regulatory context
- UK CAA: Check current official guidance for the operation, observer roles, visual-contact requirements, and any authorisation conditions.
- EASA: The competent member state, operating category, current rule set, and authority outcome determine the evidence required; a SighThor model is not that decision.
- FAA: Verify current Part 107, waiver, airspace, and operational requirements separately. Draft geometric LOS material does not create an authorisation.
Geometric viewshed output can help a team document its assumptions, but it is not proof of legal VLOS. Source resolution, aircraft conspicuity, conditions, observer competence, field checks, and the current authority process all remain material.
The Core Challenge
Terrain decides where your observer plan breaks
A flat map makes every observer position look viable. Real terrain is less forgiving.
Consider two observer positions: one on a hilltop at 180 metres elevation, the other in a valley at 95 metres. On a 2D map, both appear to have similar range to the drone route. In reality, the hilltop observer has unobstructed sightlines across kilometres of terrain, while the valley observer loses visibility behind a ridge just 400 metres away. This is not an edge case — it is the normal condition in any terrain that is not completely flat.
SighThor calculates the viewshed from each observer position using real elevation data. The analysis does not simply draw a straight line between two points. It evaluates a dense set of sample points along the route, checking each one against the terrain surface between the observer and the drone.
The calculation accounts for several physical effects that naive line-of-sight tools ignore:
- Terrain masking:The primary obstruction source. Even modest ridges (10–20 m above the surrounding terrain) can completely block an observer's view of a drone flying at 50 m AGL two kilometres away.
- Fresnel zone clearance:Visual observation degrades before the geometric line of sight is fully blocked. When terrain approaches the sightline without intersecting it, atmospheric effects and diffraction reduce the observer's ability to judge the aircraft's position and heading. SighThor models the first Fresnel zone to flag these marginal cases.
- Earth curvature correction: At distances beyond roughly 3 km, the curvature of the earth begins to lower distant terrain relative to the observer. Over 10 km, this effect amounts to approximately 8 metres — enough to change a marginal model finding. The engine applies a standard 4/3 earth radius refraction model.
What the analysis includes
- Terrain masking against the selected source
- First Fresnel zone clearance modelling
- Earth curvature + atmospheric refraction (4/3 R)
- Per-observer viewshed across the full route
- Marginal clearance warnings (near-miss zones)
- Composite coverage map across all observers
Common failure pattern
An observer is placed at a road junction with good vehicle access. On the map it looks adequate. But the terrain profile reveals a low ridge 800 m north that blocks visibility to a 1.2 km section of the route. The team only discovers this during the field survey — after committing time, travel, and coordination costs. Terrain-aware planning can surface this class of risk before fieldwork, so the team can verify it rather than discovering it only on site.
Multi-Observer Network
Plan the full observer network, not just individual positions
EVLOS operations rarely depend on a single observer. SighThor models the entire network — pilot, relays, and the gaps between them.
Pilot + relay observer placement
Place the remote pilot and relay observers on the map. Each position is simulated against the selected terrain source to flag candidate visible, blocked, and unknown route sections.
Coverage overlap between observers
The beta compares geometric line of sight from each observer. Overlap zones help plan redundancy, but continuous legal VLOS still requires field and operational confirmation.
Handoff zones along the route
Candidate handoff zones flag where one simulated observer view ends and another begins. Treat their locations as planning prompts until field-checked.
Corridor analysis
A corridor-width simulation explores lateral route offsets. It does not ensure that an observer can maintain visual contact under real operating conditions.
The practical value of multi-observer analysis is that it turns observer placement from a judgment call into an engineering problem. Instead of guessing whether three observers are enough, the team can compare which route segments are modelled for zero, one, or multiple observers — and flag positions and gaps for field review.
For higher-friction operations, a colour-coded map and terrain profile can make observer assumptions easier for a competent person to challenge. The beta output remains draft planning material and must not be treated as confirmation of continuous coverage or authority acceptance.
Workflow
From map to evidence in four steps
The workflow prepares traceable geometric LOS assumptions for competent-person review.
Place observers on the map
Drop the pilot position and relay observers onto the selected terrain source, then review its grade, resolution, and limitations before relying on the comparison.
Draw your drone route
Trace the planned flight path as waypoints and record the intended altitude reference. The beta altitude model has known limitations and must be checked before relying on AGL or AMSL results.
Run geometric LOS simulation
The engine samples line-of-sight rays along the route using the available terrain model and Fresnel assumptions. Results remain planning findings and require source and field review.
Prepare draft evidence for review
Prepare watermarked PDF and KML planning outputs with maps, profiles, and segment findings. They do not constitute authority evidence, permission, or an official SORA outcome.
Why It Matters
Flat-map planning is the most common source of EVLOS field failures
Most drone planning tools treat the world as flat. They calculate range circles from observer positions and declare coverage based on distance alone. This works acceptably in genuinely flat terrain — agricultural plains, coastal areas, open desert. But the moment the terrain has any meaningful relief — and most operational environments do — flat-map planning produces observer networks that fail in the field.
The cost of that failure is not just a wasted field day. It is a delayed project timeline, planning material that must be reworked during review, or an operation that proceeds with unrecognised gaps in visual coverage. In regulated airspace, that last outcome is the one that keeps safety managers awake.
SighThor exists because the terrain question should be answered before anyone drives to the site. The elevation data is available. The mathematics of line-of-sight analysis are well understood. The only reason operators still discover terrain problems in the field is that their planning tools do not model terrain. That is the gap this platform fills.
Turn a site brief into an inspectable planning draft.
Start with terrain-aware geometry, then review every source, limitation and field check.