Hi Jacob,
Thank you for reaching out and sharing such a detailed and fascinating use case. We truly appreciate the effort you’ve put into developing this free/open-source livestock tracking solution using Milesight gateways and trackers. Your bachelor's thesis and the camera-based predator deterrent system are incredibly impressive!
The feature you requested—using the AT101 as a BLE scanner to detect nearby lamb BLE tags and reporting them alongside regular uplinks—is highly insightful. You clearly outlined the pain point (the inability to track small, fast-growing lambs and the high loss rate) and how this specific BLE implementation can solve it efficiently.
Your secondary suggestion of using BLE scanning for positioning as a low-power alternative to Wi-Fi is also a very valuable perspective.
Here is how we are handling your suggestions:
Feature Request Escalation: I have officially submitted your detailed use case and feature request to our Product Management and R&D teams for evaluation. They will assess the hardware feasibility and firmware development required to support BLE scanning on the AT101.
Follow-up: We are currently reviewing our roadmap to see how and when we might be able to incorporate this functionality. I will keep you updated as soon as I have concrete feedback from our engineers regarding the technical feasibility and potential timeline.
We love seeing our products being used in such meaningful, open-source projects that solve real-world problems.
Thank you again for your valuable feedback and for helping us improve our solutions. I will be in touch soon.
Best regards,
Aslan
Jacob Ludvigsen
I've developed a free/libre/open source livestock tracking solution using Milesight AT101, UG67, Chirpstack v4 and Home Assistant. Details here, Home Assistant automations here.
A highly requested feature among the sheep farms who test and deploy the solution, is functionality for knowing "when a lamb was last detect nearby its mother / the last known location of a lamb".
Of the 1.1 million sheep & lambs released yearly to graze in mountains and remote pastures of Norway, more than 100 thousand are lost. Finding the dead sheeps/lambs quickly and accurately determining their causes of death is important to reduce dangers to sheep+lambs, ensure farmers are properly compensated, and reduce tensions between conservation of predators and farming.
If only mother sheep are tracked, it can take a long time before the loss of a lamb is noticed by farmers. When untracked lambs are lost/taken by predators while out grazing, it's very difficult to find them again because their last position isn't known. If and when they're found, it's usually too late to determine their cause of death.
For more details, you can read my bachelor's thesis on the subject, where I developed a camera-based system to detect and scare predators from attacking livestock. Thesis and code.
Lambs are too small to carry a full-size tracker, and grow too fast for carrying a tracker or similar around their neck. Any item for keeping track of lambs must be small and lightweight enough to attach to the ear of the lambs, like ordinary ear tags used for visual identification.
Other livestock tracking solutions solve this pretty well, using the tracker attached to the mother sheep as a BLE scanner, and equipping the lambs with small BLE tags (something like airtags). When the mother tracker sends an uplink reporting its position, it also scans for nearby BLE tags and sends a list of detected BLE tags.
I'd like to use AT101 similarly, such that each AT101 scans for BLE tags and includes a list of detected BLE tags along with ordinary uplink location reports.
A very different and also valid use case could be to use BLE scanning as a positioning strategy, where AT101 scans for stationary sources of BLE signal, and uses signal strength to determine position. This would be similar to wifi positioning strategy, but might use less power and be simpler to set up for users.
Best regards,
Jacob