1 00:00:01,100 --> 00:00:04,160 covering the week's top text doors it's 2 00:00:04,160 --> 00:00:04,170 covering the week's top text doors it's 3 00:00:04,170 --> 00:00:06,680 covering the week's top text doors it's like Linux bias researchers have 4 00:00:06,680 --> 00:00:06,690 like Linux bias researchers have 5 00:00:06,690 --> 00:00:08,690 like Linux bias researchers have developed an algorithm that could stop 6 00:00:08,690 --> 00:00:08,700 developed an algorithm that could stop 7 00:00:08,700 --> 00:00:10,520 developed an algorithm that could stop self-driving vehicles from getting into 8 00:00:10,520 --> 00:00:10,530 self-driving vehicles from getting into 9 00:00:10,530 --> 00:00:14,030 self-driving vehicles from getting into crashes and traffic jams the algorithm 10 00:00:14,030 --> 00:00:14,040 crashes and traffic jams the algorithm 11 00:00:14,040 --> 00:00:16,939 crashes and traffic jams the algorithm divides the ground beneath the machines 12 00:00:16,939 --> 00:00:16,949 divides the ground beneath the machines 13 00:00:16,949 --> 00:00:19,910 divides the ground beneath the machines into a grid the robots learn their 14 00:00:19,910 --> 00:00:19,920 into a grid the robots learn their 15 00:00:19,920 --> 00:00:22,070 into a grid the robots learn their position through technology similar to 16 00:00:22,070 --> 00:00:22,080 position through technology similar to 17 00:00:22,080 --> 00:00:25,670 position through technology similar to GPS and coordinate their own movements 18 00:00:25,670 --> 00:00:25,680 GPS and coordinate their own movements 19 00:00:25,680 --> 00:00:28,939 GPS and coordinate their own movements together through sensors that assess 20 00:00:28,939 --> 00:00:28,949 together through sensors that assess 21 00:00:28,949 --> 00:00:32,500 together through sensors that assess where there's free space to move 22 00:00:32,500 --> 00:00:32,510 where there's free space to move 23 00:00:32,510 --> 00:00:34,880 where there's free space to move Northwestern engineers Michael 24 00:00:34,880 --> 00:00:34,890 Northwestern engineers Michael 25 00:00:34,890 --> 00:00:39,380 Northwestern engineers Michael Rubenstein says the robots refused to 26 00:00:39,380 --> 00:00:39,390 Rubenstein says the robots refused to 27 00:00:39,390 --> 00:00:41,900 Rubenstein says the robots refused to move to a spot until that spot is free 28 00:00:41,900 --> 00:00:41,910 move to a spot until that spot is free 29 00:00:41,910 --> 00:00:45,470 move to a spot until that spot is free and until they know that no other robots 30 00:00:45,470 --> 00:00:45,480 and until they know that no other robots 31 00:00:45,480 --> 00:00:48,650 and until they know that no other robots are moving to that same spot they are 32 00:00:48,650 --> 00:00:48,660 are moving to that same spot they are 33 00:00:48,660 --> 00:00:52,729 are moving to that same spot they are careful and reserve a space ahead of 34 00:00:52,729 --> 00:00:52,739 careful and reserve a space ahead of 35 00:00:52,739 --> 00:00:57,080 careful and reserve a space ahead of time Rubenstein's team tested their 36 00:00:57,080 --> 00:00:57,090 time Rubenstein's team tested their 37 00:00:57,090 --> 00:01:00,770 time Rubenstein's team tested their algorithm on a swarm of 100 robots set 38 00:01:00,770 --> 00:01:00,780 algorithm on a swarm of 100 robots set 39 00:01:00,780 --> 00:01:03,290 algorithm on a swarm of 100 robots set up in their lab to cut out any 40 00:01:03,290 --> 00:01:03,300 up in their lab to cut out any 41 00:01:03,300 --> 00:01:05,390 up in their lab to cut out any distractions the robots were only 42 00:01:05,390 --> 00:01:05,400 distractions the robots were only 43 00:01:05,400 --> 00:01:08,359 distractions the robots were only allowed to sense three or four of their 44 00:01:08,359 --> 00:01:08,369 allowed to sense three or four of their 45 00:01:08,369 --> 00:01:11,660 allowed to sense three or four of their closest neighbors this restricted their 46 00:01:11,660 --> 00:01:11,670 closest neighbors this restricted their 47 00:01:11,670 --> 00:01:15,200 closest neighbors this restricted their vision it and it made the system easier 48 00:01:15,200 --> 00:01:15,210 vision it and it made the system easier 49 00:01:15,210 --> 00:01:18,320 vision it and it made the system easier to scale as the robots can interact 50 00:01:18,320 --> 00:01:18,330 to scale as the robots can interact 51 00:01:18,330 --> 00:01:21,469 to scale as the robots can interact locally without needing global 52 00:01:21,469 --> 00:01:21,479 locally without needing global 53 00:01:21,479 --> 00:01:23,660 locally without needing global information think about the impact of 54 00:01:23,660 --> 00:01:23,670 information think about the impact of 55 00:01:23,670 --> 00:01:26,749 information think about the impact of that so the advantage of a swarm of 56 00:01:26,749 --> 00:01:26,759 that so the advantage of a swarm of 57 00:01:26,759 --> 00:01:29,510 that so the advantage of a swarm of robots is that there is no centralized 58 00:01:29,510 --> 00:01:29,520 robots is that there is no centralized 59 00:01:29,520 --> 00:01:32,420 robots is that there is no centralized controller that can disrupt the whole 60 00:01:32,420 --> 00:01:32,430 controller that can disrupt the whole 61 00:01:32,430 --> 00:01:35,359 controller that can disrupt the whole system this allows them to work together 62 00:01:35,359 --> 00:01:35,369 system this allows them to work together 63 00:01:35,369 --> 00:01:38,929 system this allows them to work together to accomplish any task even if one of 64 00:01:38,929 --> 00:01:38,939 to accomplish any task even if one of 65 00:01:38,939 --> 00:01:41,080 to accomplish any task even if one of them breaks down 66 00:01:41,080 --> 00:01:41,090 them breaks down 67 00:01:41,090 --> 00:01:43,490 them breaks down this gives the system and obvious 68 00:01:43,490 --> 00:01:43,500 this gives the system and obvious 69 00:01:43,500 --> 00:01:45,410 this gives the system and obvious application in warehouse robots but 70 00:01:45,410 --> 00:01:45,420 application in warehouse robots but 71 00:01:45,420 --> 00:01:47,690 application in warehouse robots but Rubinstein believes it could actually 72 00:01:47,690 --> 00:01:47,700 Rubinstein believes it could actually 73 00:01:47,700 --> 00:01:51,410 Rubinstein believes it could actually also cut traffic and collisions for 74 00:01:51,410 --> 00:01:51,420 also cut traffic and collisions for 75 00:01:51,420 --> 00:01:54,889 also cut traffic and collisions for self-driving vehicles on the road he 76 00:01:54,889 --> 00:01:54,899 self-driving vehicles on the road he 77 00:01:54,899 --> 00:01:57,230 self-driving vehicles on the road he said quote by understanding how to 78 00:01:57,230 --> 00:01:57,240 said quote by understanding how to 79 00:01:57,240 --> 00:02:00,950 said quote by understanding how to control our swarm robots to form shapes 80 00:02:00,950 --> 00:02:00,960 control our swarm robots to form shapes 81 00:02:00,960 --> 00:02:03,830 control our swarm robots to form shapes we can understand how to control fleets 82 00:02:03,830 --> 00:02:03,840 we can understand how to control fleets 83 00:02:03,840 --> 00:02:07,399 we can understand how to control fleets of autonomous vehicles as they interact 84 00:02:07,399 --> 00:02:07,409 of autonomous vehicles as they interact 85 00:02:07,409 --> 00:02:10,009 of autonomous vehicles as they interact with each other