bioRxivJClub
A Smart Bait Dispenser with Real-time Monitoring for Species-Specific Bait Delivery
bioRxiv · · Preprint · Open access
Chaturvedi, Connelly + 4 more
Abstract ↗AI summary
The abstract is read at the publisher; the summary is JClub's.
An open-source smart bait dispenser for Tasmanian devil Sarcophilus harrisii achieved 98.6% classification accuracy and 75.6% overall target-specific bait delivery success across 62 nights in Tasmania.
- Why it matters: Non-target bait consumption limits oral bait vaccination and other wildlife baiting approaches, and existing delivery devices often lack species-specific targeting.
- What they did: The system used field-based image classification with near real-time remote telemetry to trigger bait release; four dispensers ran 62 nights at three Tasmanian field sites, recording 2106 motion triggers.
- The result: Of 876 animal-triggered events, the dispenser released 41 placebo baits, with 31 target-specific distributions when Tasmanian devils were present, and no animals forcibly retrieved baits; average battery life was 19.2 nights.
The findingWhy it mattersWhat they didThe result
- Open access