Autonomous Flight Safety System (AFSS)

ABOUT AUTONOMOUS FLIGHT SAFETY SYSTEM (AFSS)

An Autonomous Flight Safety System is a small package of electronics that rides on a rocket and decides, all by itself and in real time, whether the rocket has become a danger to people on the ground. If it has, the system destroys the rocket. It does the job a human safety officer once did from the ground.

Quick facts

• Also called: Autonomous Flight Termination System (AFTS). The two terms are used largely interchangeably; “AFSS” names the whole onboard safety system, while “AFTS” emphasizes the part that ends the flight.
• Developed by: the U.S. Government version by NASA Goddard’s Wallops Flight Facility and Kennedy Space Center, partnering with the U.S. Department of Defense; work began in the early 2000s. Key NASA authors: James B. Bull and Raymond J. Lanzi.
• First operational orbital use: SpaceX Falcon 9, February 19, 2017.
• Core sensors: redundant GPS receivers and inertial measurement units (IMUs) — devices that sense motion and rotation.
• Built-in redundancy: multiple flight computers, batteries, and “dead man” switches that guard against an accidental destruct.

What it is and how it works

Before liftoff, engineers load the unit with mission-specific “flight rules”: the corridor the rocket is allowed to fly through, keep-out zones, speed and impact limits, and what a healthy ascent should look like. The system then runs self-checks while ground operators watch.

In flight, the Autonomous Flight Termination Unit (AFTU) — the decision computer — constantly reads its position, velocity, and orientation from the redundant GPS receivers, aided by the IMUs (together a GPS-aided inertial navigation system). Several independent flight processors each work out where the rocket is and predict its Instantaneous Impact Point (IIP): where the debris would land if the engines stopped right now.

Think of it like a self-driving car that already knows every street it is allowed to use. It is not waiting for a dispatcher on a radio; it watches its own position and acts the instant it strays. Each processor checks the rocket against the loaded rules, and the redundant copies must agree (voting logic) before acting, so one faulty sensor or computer cannot wrongly trigger — or wrongly suppress — a destruct.

If the rocket leaves its boundary, fails to reach a safe orbit, or otherwise becomes an immediate threat, the unit fires Electronic Safe and Arm Devices and initiators (such as low-energy exploding foil initiators) to set off the destruct hardware — rupturing tanks, venting motors, or cutting the vehicle — and the flight ends.

Why it matters

The traditional alternative was a large, costly ground setup: tracking radars, command-destruct radio transmitters, safety-critical telemetry, and human Range Safety Officers watching screens and ready to press a destruct button. Every launch had to share and schedule that apparatus.

By moving the navigation and the destruct decision onto the rocket, ranges can support many more launches at lower cost and recycle faster between flights. According to the U.S. Air Force in 2017, the cost of range services per launch fell roughly 50 percent, and the Eastern Range targeted up to 48 launches per year by trimming ground radar and telemetry infrastructure. U.S. regulation (the FAA’s Part 450 framework) pushed in the same direction, making onboard autonomy the modern standard.

It also enables flights that were physically impossible before — such as polar “dogleg” corridors where the rocket’s own exhaust plume would block a ground command link, demonstrated by SpaceX’s SAOCOM 1B polar launch from Florida in August 2020. And it lets rockets fly from new or austere sites that lack built-up range infrastructure. The trade-off: every rule and safeguard must be correct, because there is no human to override in the moment.

Where it is used

• SpaceX Falcon 9 / Falcon Heavy — flew the first operational orbital AFTS on February 19, 2017; now standard, and enabled the August 2020 SAOCOM 1B polar launch from Cape Canaveral.
• NASA developmental flights — the Wallops/Goddard and Kennedy prototype made its first flight demonstration (an ATK-built system) on November 19, 2013, aboard an Orbital Sciences Minotaur I (the ORS-3 mission) from Wallops.
• Rocket Lab Electron — a fully autonomous flight-termination system (debuted December 2019) that shuts down the engines if the vehicle goes off course.
• ESA Ariane 5 / Ariane 6 — the European “KASSAV” autonomous safety kit (first flown on Ariane 5 in August 2020; KASSAV 2 adds autonomous termination authority for Ariane 6).
• Next-generation U.S. vehicles — ULA Vulcan Centaur, Blue Origin New Glenn, and NASA’s Space Launch System (SLS) all adopt autonomous flight-safety systems.

OPERATING PRINCIPLE

VEHICLES USING THIS SYSTEM

SPECIFICATIONS

DETAILED SPECS