Part IV Of A LinkedIn Series By Mike Richards
Based on Drone America’s 16 years of real-world experience in the UAS industry

Not everything that flies itself is autonomous. In fact, most commercial drones in operation today are automated, not truly autonomous. The difference? Automation follows pre-set instructions. Autonomy interprets real-world input and makes decisions. One is a script; the other is situational awareness.

Here’s a simple analogy: a microwave is automated, it knows what to do when you hit “Start.” A chef is autonomous, they adapt if the stove malfunctions, if ingredients change, or if the fire alarm goes off. One executes; the other evaluates.

In UAS operations, true autonomy is still emerging. We rely on automation for commercial survey work, such as powerline inspections to execute precision flight paths, trigger sensors, and capture high-resolution data. But autonomy is what we need when the unexpected happens: when GPS drifts, when an obstacle appears, or when the weather shifts mid-flight.

This is where AI has the potential to transform flight safety, not by replacing the operator, but by augmenting the aircraft’s ability to detect and avoid, to re-route, and to escalate decisions based on rules and risk.

Autonomy isn’t a shortcut. It’s a layered capability that demands testing, transparency, and trust. And as with every part of aviation, we build that capability step by step with the same rigor we apply to the rest of the aircraft.

Are you working on AI-driven detect-and-avoid systems or autonomy for UAS? We’d love to hear what challenges you’re facing, how you’re testing them, and what success looks like in real-world conditions. Let’s start a conversation because safe autonomy isn’t a solo effort.

Up Next: Heavier UAS, Higher Stakes