A General Atomics Aeronautical Systems Avenger unmanned air vehicle (UAV) was recently autonomously flown by an artificially intelligent software program developed by the US Defense Advanced Research Projects Agency (DARPA).
DARPA’s Collaborative Operations in Denied Environment (CODE) software program controlled the manoeuvring of the Avenger UAV for more than two hours without traditional pilot input, General Atomics said on 4 December.
The experimental CODE program is designed to help the US military fly one or more UAVs without needing an always-connected communications link or dedicated pilot and sensor operator. In theory, autonomous flight would allow many more UAVs to be flown simultaneously and protect the aircrafts’ incoming control signals from being jammed or spoofed.
“This represents a big step on the path to more-sophisticated autonomous missions for unmanned aircraft, where operator input can be minimised to support optimal manning of multiple products for complex air battles,” says General Atomics Aeronautical Systems president David Alexander. “For this initial flight, we used Avenger as the flight surrogate for the Skyborg capability set, which is a key focus for [General Atomics’] emerging air-to-air portfolio.”
Skyborg is a parallel effort by the US Air Force Research Laboratory to develop an autonomous flight and mission control program for a variety of UAVs.
The CODE program is also designed to allow UAVs to autonomously work in coordination with other UAVs, in part by relying on a mesh network – a communications system where signals are passed around using multiple nodes, not just broadcast from a single transmission point. With multiple UAVs in the sky sharing information, the network of aircraft becomes smarter, and blocking all broadcast points at once becomes harder for enemies.
DARPA envisions that a human supervisor, perhaps flying in a nearby fighter aircraft, would watch over the UAVs as they executed missions. That supervisor would approve flight movements and subsystem actions as recommended by the artificially intelligent software program.
“Using collaborative autonomy, CODE-enabled unmanned aircraft would find targets and engage them as appropriate under established rules of engagement, leverage nearby CODE-equipped systems with minimal supervision, and adapt to dynamic situations such as attrition of friendly forces or the emergence of unanticipated threats,” says the research agency.
General Atomics says it was able to integrate its Avenger with a mesh network and show the UAV autonomously working with five simulated aircraft as part of an air-to-air search mission demonstration. It is thought UAVs might be better equipped than manned aircraft for a number of missions, including air-to-air search, because of their longer flight endurance and expendability.