
The robotic-rabbit experiment is testing whether a python can be lured close enough to be detected and then captured. Nearly 90 devices are in the Everglades, equipped with fans that disperse mammal scent, a second scent source about 10 feet away, and cameras with updated detection software. So far, the project has not recorded a python capture.
A University of Florida progress report found 79 of the devices functional as of last summer, according to the Palm Beach Post. The newspaper reported that the setup now uses a fan to distribute mammal scent near each rabbit, a second scent source roughly 10 feet away, and upgraded camera software intended to detect pythons more quickly. The South Florida Water Management District has earmarked $327,535 for the experiment and related work dating back to 2022; the district said the program had not yet produced any python captures.
Nearly 90 Decoys Already in the Field
Researchers told NBC 6 South Florida in September that about 90 decoys were deployed in the Everglades. Each can be programmed to produce heat, scent, vibration, or movement, in combinations researchers can vary. The devices are meant to imitate marsh rabbits, a known python prey item, and draw snakes toward researchers rather than require hunters to search the wetlands for them.
Researchers can control the decoys' body temperature remotely and are testing warmth levels that mimic wild rabbits, including adjustments of one or two degrees Celsius. They are also varying the duration and strength of vibrations. The tests are aimed at finding combinations that will attract a python and keep it near long enough for a camera to detect it and alert someone who can capture it.
Field testing is mostly confined to warmer months, roughly June through September, according to the same NBC 6 report, and researchers hope to eventually build a University of Florida lab that would allow year-round testing instead of a seasonal window.
Bunny Eyes and a Cellular Cost Problem
Part of the upgrade effort is a smaller, low-powered camera system researchers are developing called Bunny Eyes, which uses a filter designed to distinguish the reflective light that bounces off python scales from ordinary background light. The system includes an onboard AI processor meant to help determine whether whatever triggered the camera is actually a python, and it is designed to send an alert and prompt a photograph the moment it suspects one is nearby. Researcher Robert McCleery has said that AI-based python detection can identify the snakes with about 95 percent accuracy.
Image transmission is another constraint. Researchers have deployed roughly 90 traditional game cameras across the Everglades that take high-resolution photographs every minute, but sending all that data by cellular connection is prohibitively expensive, according to 10 Tampa Bay's report. That cost is one reason for developing Bunny Eyes. Researchers are also considering a floating version of the robotic rabbits because parts of the Everglades are underwater during portions of the year.
McCleery did not provide fresh project numbers when the Palm Beach Post sought an update, saying the data was still being reviewed. He has described the broader difficulty as searching a vast area for a well-camouflaged animal. McCleery has also said that stopping pythons from spreading farther north could offer a better return on investment than removing them where they are already abundant; the team plans to identify areas where the invasive population is expanding and focus removal efforts there, according to 10 Tampa Bay's report.
What's Already Known to Work
The decoys are being tested against a behavioral benchmark from earlier research: pythons drawn to live rabbits in pens stayed an average of 80 minutes. Non-robotic removal efforts have also produced results. The Conservancy of Southwest Florida has reported removing 177 invasive pythons weighing a combined 8,080 pounds, largely with trained scout snakes used to track wild ones.
A separate UF study compared wildlife observed in and around Big Cypress National Preserve with records from 12 years earlier and found that some previously documented animals were no longer present at those locations. McCleery has said he has no data supporting the idea of a wildlife rebound. The Florida Fish and Wildlife Conservation Commission notes that pythons prey on native species and may reduce their populations locally, and that research into the scope of that impact is ongoing. The agency considers the snakes established across a range stretching from just south of Lake Okeechobee to Key Largo, and from western Broward County west to Collier County.
The water management district has confirmed that its Python Elimination Agents, who physically remove snakes, are not yet using the robotic rabbits, according to Yahoo News. Under current FWC rules, Burmese pythons can be captured and humanely killed year-round without a permit or hunting license on 32 commission-managed lands in south Florida. The species is on Florida's Prohibited species list, and the U.S. Fish and Wildlife Service lists it as an Injurious Species under the Lacey Act, which blocks its importation into the country. Researchers at the University of Florida and the South Florida Water Management District are still working to establish whether a robotic decoy can hold a python's attention long enough to bring it within reach, as live bait has, 10 Tampa Bay reported.
What the removal records show
The district's Python Elimination Program operates separately from the robotic-rabbit trial. The South Florida Water Management District says its removal agents have access to designated lands including Everglades National Park, Big Cypress National Preserve, and Everglades and Francis S. Taylor Wildlife Management Area. The Florida Fish and Wildlife Conservation Commission reported that more than 23,500 pythons had been removed from Florida natural areas as of April 2025, including more than 11,000 by FWC PATRIC and SFWMD PEP contractors since their programs began. Those figures count removals; on their own, they do not establish how the python population has changed.









