Queensland's Cyborg Cockroaches Hit 72% on a Rescue Task

Researchers at the University of Queensland's Biorobotics Lab, working with biomedical engineers at the University of New South Wales, said a cyborg-equipped cockroach completed a full navigate-and-inject rescue sequence in 18 of 25 trials, a 72 percent success rate, according to a study published in the journal Advanced Science and detailed in a University of Queensland announcement on Aug. 27. The insect, a North Queensland giant burrowing cockroach fitted with a needle-tipped injector, is a proof of concept for delivering medication to a person a search-and-rescue team cannot physically reach.

In the test, the roach-mounted robots, which the researchers call Paraborgs, navigated a 2.5-meter course past three checkpoints before launching an injector at an 8-by-10-centimeter silicone target standing in for a trapped person's arm. Every one of the 25 trials completed the course; the lower 72 percent figure covers the harder combined task of positioning precisely enough to fire the needle into the target as well. Isolated to just the close-range injection step, within 15 centimeters of the target, the success rate rose to 95 percent, according to the University of Queensland's account of the results.

"We wanted to take the next step. Once they find someone, can they actually help?" said T. Thang Vo-Doan, who directs the University of Queensland's Biorobotics Lab and led the study. Hai Nhan Le, a PhD candidate on the project, said the engineering problem was mostly about stability rather than sensing: "The cyborg insect has to navigate to the target, position itself accurately and remain stable enough to perform the injection."

Why a cockroach carries the payload

The species matters as much as the electronics. A North Queensland giant burrowing cockroach can weigh up to 35 grams unequipped, and the research units in the study weighed roughly 40 grams once fitted with navigation electronics and an injector; the researchers say a healthy specimen can still carry a payload of more than half its own body weight, which is what makes it a chassis and not just a subject. The team deliberately split functions across separate Paraborgs rather than combining a camera and an injector on one insect, reasoning that the added weight and power draw of both systems together would cut mobility and battery life on the same animal.

The injector concept sits next to two other robotic-medicine bets filed the same month: Vitestro's FDA-cleared Aletta blood-draw robot, built for a hospital room rather than a disaster site, and Icarus Robotics' zero-gravity test of its Joy robot bound for the International Space Station, another machine designed to operate somewhere a human technician cannot easily go.

What five to ten years still has to solve

Vo-Doan told the University of Queensland that real deployment is five to 10 years away, contingent on resources, and the researchers listed the open problems rather than glossing over them: reliable wireless communication inside a collapsed structure, compensating for a trapped person who moves, navigating rubble more complex than a lab course, and coordinating multiple Paraborgs autonomously. On the medical side, the team said drug choice, dosage, sterility, needle safety and regulatory approval all remain unresolved, and said plainly that the current system "is not a medical device ready to be used on people today."

What would change that timeline is a field trial outside a lab course, in an actual rubble or cave environment closer to the disaster scenario the project is built for. None has been scheduled, and the 72 percent figure so far describes a fixed 2.5-meter course under lab conditions, not an unpredictable one.

Paraborg trial result (n=25)	Rate
Course completion (2.5 m, 3 checkpoints)	100%
Full navigate-and-inject task	72%
Close-range injection (within 15 cm)	95%