1
00:00:00,620 --> 00:00:03,620
(graphic whooshing)

2
00:00:07,470 --> 00:00:08,400
- [Narrator] Although completely

3
00:00:08,400 --> 00:00:11,730
autonomous self-driving
cars are still years away,

4
00:00:11,730 --> 00:00:15,450
semi-autonomous vehicles
already travel our roadways.

5
00:00:15,450 --> 00:00:17,970
To prepare for fully-autonomous vehicles,

6
00:00:17,970 --> 00:00:21,720
digitizing the locations of
traffic control infrastructure,

7
00:00:21,720 --> 00:00:24,900
like traffic signals,
signs, and intersections,

8
00:00:24,900 --> 00:00:29,460
is essential to safely ushering
in that future reality.

9
00:00:29,460 --> 00:00:32,700
The Texas Department of
Transportation funded research

10
00:00:32,700 --> 00:00:36,030
led by the Texas A&M
Transportation Institute

11
00:00:36,030 --> 00:00:38,730
to study the use of emerging data sources

12
00:00:38,730 --> 00:00:41,250
in mapping TCI locations

13
00:00:41,250 --> 00:00:44,280
with the ultimate goal of
improving the safe interaction

14
00:00:44,280 --> 00:00:47,550
of automated vehicles with
the driving environment.

15
00:00:47,550 --> 00:00:51,690
- Autonomous vehicles heavily
rely on the sensing devices

16
00:00:51,690 --> 00:00:54,030
to operate safely on our roads.

17
00:00:54,030 --> 00:00:56,291
However, when the sensors fail,

18
00:00:56,291 --> 00:01:00,060
having access to
digitized data about roads

19
00:01:00,060 --> 00:01:02,610
or traffic signs becomes critical

20
00:01:02,610 --> 00:01:05,220
for the vehicle's continued operation.

21
00:01:05,220 --> 00:01:08,755
The main goal of this project
was to develop a framework

22
00:01:08,755 --> 00:01:12,240
for digitizing and providing TCI data

23
00:01:12,240 --> 00:01:14,130
to autonomous vehicles.

24
00:01:14,130 --> 00:01:15,780
- Basically, the research team focused

25
00:01:15,780 --> 00:01:18,270
on enhancing the traffic
control infrastructure

26
00:01:18,270 --> 00:01:22,530
to provide a digital
access to automating codes

27
00:01:22,530 --> 00:01:25,110
and adding some redundancy

28
00:01:25,110 --> 00:01:28,230
to their perception system
for improved safety.

29
00:01:28,230 --> 00:01:31,170
- The most important
thing we have accomplished

30
00:01:31,170 --> 00:01:34,348
in this project is that we have built up

31
00:01:34,348 --> 00:01:39,345
additional map layer with
up-to-date TCI information,

32
00:01:39,345 --> 00:01:42,810
so the AV sensors can be augmented

33
00:01:42,810 --> 00:01:45,720
by this digital TCI map layer,

34
00:01:45,720 --> 00:01:49,530
and also this map layer
can also provide redundancy

35
00:01:49,530 --> 00:01:51,930
for AV perception and safety.

36
00:01:51,930 --> 00:01:54,540
- The research also explored
the various, you know,

37
00:01:54,540 --> 00:01:57,600
traffic control infrastructure
digitization approaches,

38
00:01:57,600 --> 00:01:59,643
including like data collection,

39
00:02:00,690 --> 00:02:02,910
the artificial intelligence solutions,

40
00:02:02,910 --> 00:02:05,580
and dashcam image processing.

41
00:02:05,580 --> 00:02:07,560
The research team also acquired

42
00:02:07,560 --> 00:02:10,560
some sample traffic control
infrastructure datasets

43
00:02:10,560 --> 00:02:12,513
from some of the commercial companies.

44
00:02:12,513 --> 00:02:16,920
- We evaluate and collect
this kind of TCI dataset.

45
00:02:16,920 --> 00:02:20,460
Then we do the quality
check and the data fusion.

46
00:02:20,460 --> 00:02:23,940
We get the final, more complete,

47
00:02:23,940 --> 00:02:25,860
more accurate TCI layer.

48
00:02:25,860 --> 00:02:30,300
Then we pass this layer
to our UT data team.

49
00:02:30,300 --> 00:02:33,240
They can use this layer
to do the simulation

50
00:02:33,240 --> 00:02:37,080
on their AV simulation
platform called the MATISSE.

51
00:02:37,080 --> 00:02:40,966
- The simulations clearly
demonstrated the positive impact

52
00:02:40,966 --> 00:02:45,060
of TCI-digitized data on AV operations.

53
00:02:45,060 --> 00:02:47,940
However, the simulations also unveiled

54
00:02:47,940 --> 00:02:49,263
that in certain scenarios,

55
00:02:50,370 --> 00:02:53,160
AVs cannot solely rely on digitized data

56
00:02:53,160 --> 00:02:55,470
to ensure safety operations.

57
00:02:55,470 --> 00:02:58,586
- This team also addressed
the potential legal issues

58
00:02:58,586 --> 00:03:03,586
and that created a data
management plan for the agency.

59
00:03:04,140 --> 00:03:08,310
This project offers the
department a streamlined framework

60
00:03:08,310 --> 00:03:10,440
for textiles, traffic control,

61
00:03:10,440 --> 00:03:12,810
infrastructure digitization and sharing.

62
00:03:12,810 --> 00:03:16,980
And I think this traffic
control infrastructure layer

63
00:03:16,980 --> 00:03:20,940
enhances automated vehicle
perception and safety,

64
00:03:20,940 --> 00:03:23,850
and it presents a robust solution

65
00:03:23,850 --> 00:03:28,200
for improved AV integration
into Texas roadway system.

66
00:03:28,200 --> 00:03:29,490
- [Narrator] For more information

67
00:03:29,490 --> 00:03:32,490
and to find the publications
for this project,

68
00:03:32,490 --> 00:03:35,160
please visit the text at research library

69
00:03:35,160 --> 00:03:36,903
at the link shown below.