1
00:00:00,120 --> 00:00:02,560
Speaker 1: Bit of tech, speaking of tech to make life easier.

2
00:00:02,560 --> 00:00:04,320
Speaker 1: This time on the farm got a new drone with

3
00:00:04,360 --> 00:00:07,400
Speaker 1: a bit of AI technology makes counting stock easier. Apparently

4
00:00:07,440 --> 00:00:11,680
Speaker 1: they call it Skycount project. Stephen Gerrin is PGG writes

5
00:00:11,720 --> 00:00:15,920
Speaker 1: and CEO and he's with as Stephen morning for hand

6
00:00:16,239 --> 00:00:18,799
Speaker 1: not at all. So you're involved or your companies involved

7
00:00:18,840 --> 00:00:20,959
Speaker 1: in this? So is this an expansion into tech for

8
00:00:21,040 --> 00:00:23,200
Speaker 1: your company? So it's more than gum boots and feed

9
00:00:23,320 --> 00:00:23,560
Speaker 1: or what.

10
00:00:24,760 --> 00:00:26,520
Speaker 2: We've always been a bit more than gun boots and

11
00:00:26,560 --> 00:00:29,760
Speaker 2: feed as time has developed, Mike, But this is a

12
00:00:29,760 --> 00:00:33,200
Speaker 2: new innovation for us because we can previously our innovations

13
00:00:33,240 --> 00:00:37,159
Speaker 2: around online training for our lavestock and our application. But

14
00:00:37,200 --> 00:00:40,279
Speaker 2: this is a new piece of kit for us and

15
00:00:41,080 --> 00:00:45,040
Speaker 2: we've developed in a partment within the and Microsoft. So

16
00:00:45,280 --> 00:00:47,000
Speaker 2: we're really pleased to launch it to the market at

17
00:00:47,040 --> 00:00:48,400
Speaker 2: the phield days and June.

18
00:00:48,680 --> 00:00:51,720
Speaker 1: How much drone and how much AI.

19
00:00:52,640 --> 00:00:56,040
Speaker 2: The drone does the initial count so we using high

20
00:00:56,080 --> 00:00:59,160
Speaker 2: definition cameras there. We've tried a number of technologies, but

21
00:01:00,040 --> 00:01:01,959
Speaker 2: READ doesn't seem to do the work. You can't really

22
00:01:02,000 --> 00:01:05,080
Speaker 2: distinguish between sometimes between a tractor sitting in a field

23
00:01:05,120 --> 00:01:06,920
Speaker 2: and a water trough, et cetera versus a sheep, and

24
00:01:06,920 --> 00:01:08,440
Speaker 2: that's quite important for the farm, and they don't want

25
00:01:08,440 --> 00:01:12,600
Speaker 2: to count the sheep, not the tractors. So we've we

26
00:01:12,680 --> 00:01:16,920
Speaker 2: take the fly the drone across the paddock. You've got

27
00:01:16,959 --> 00:01:19,600
Speaker 2: to be very conscious about a lot disturbing the animals,

28
00:01:19,800 --> 00:01:22,800
Speaker 2: and that's really what this is about, making not interrupting

29
00:01:23,040 --> 00:01:26,080
Speaker 2: the workflow what's going on on farm. And then the

30
00:01:26,720 --> 00:01:29,399
Speaker 2: drone comes back to base and then we run it

31
00:01:29,480 --> 00:01:35,280
Speaker 2: through a scanner AI kit. It is counting as it goes,

32
00:01:35,319 --> 00:01:37,360
Speaker 2: but we do have to verify some things because the

33
00:01:37,400 --> 00:01:40,040
Speaker 2: AI is still learning technology.

34
00:01:40,160 --> 00:01:43,600
Speaker 1: That was next question, where do you reconniize that right here,

35
00:01:43,680 --> 00:01:45,560
Speaker 1: right now versus where it might be next year or

36
00:01:45,600 --> 00:01:46,119
Speaker 1: the year after.

37
00:01:47,120 --> 00:01:49,800
Speaker 2: In terms of this, we're getting up to ninety seven

38
00:01:49,840 --> 00:01:54,080
Speaker 2: percent accuracy in the first passes. We getting up. We're

39
00:01:54,120 --> 00:01:57,040
Speaker 2: climbing up to ninety nine in some areas. But sheep

40
00:01:57,080 --> 00:01:58,640
Speaker 2: seemed to be a bit harder because they tend to

41
00:01:59,040 --> 00:02:02,559
Speaker 2: hide and scrub and so forth. So but dery cattle

42
00:02:02,600 --> 00:02:04,440
Speaker 2: we can get you know, pretty really close to one

43
00:02:04,520 --> 00:02:08,680
Speaker 2: hundred percent accuracy, you know, ninety nine percent plus right now.

44
00:02:08,760 --> 00:02:13,519
Speaker 2: So that's that's you questioned about a year's time. We'll

45
00:02:13,560 --> 00:02:16,760
Speaker 2: advance even further beyond that as we do more, more

46
00:02:16,760 --> 00:02:20,680
Speaker 2: more runs across farms, Plectuan hill country areas fantastic.

47
00:02:20,680 --> 00:02:22,560
Speaker 1: We'll stay in touch on it. Well done, Stephen Garan,

48
00:02:22,600 --> 00:02:25,760
Speaker 1: who is worth PGG writing for more from the mic

49
00:02:25,840 --> 00:02:28,919
Speaker 1: Asking Breakfast Listen live to news talks that'd be from

50
00:02:29,000 --> 00:02:32,440
Speaker 1: six am weekdays, or follow the podcast on iHeartRadio