06-15 Is Localization Necessary Solution
0:00 - 0:02The answer is yes.
0:02 - 0:06In nearly all cases of mapping, we have robot uncertainty in motion.
0:06 - 0:08That uncertainty might grow over time.
0:08 - 0:12We need to address this; otherwise the map looks really bad.
0:12 - 0:14Let me give an example .
0:14 - 0:18Suppose a robot drives down a corridor, and it senses surrounding walls.
0:18 - 0:23If this robot has a drift problem and because of uncertainty it its motion,
0:23 - 0:26it actually believes it drives a trajectory like this.
0:26 - 0:29Then the surrounding map would look very much like that.
0:29 - 0:33Now, these might be indistinguishable at first glance,
0:33 - 0:35but if this robot ever comes back to the same place,
0:35 - 0:38then it has an opportunity to correct all this.
0:38 - 0:43A good SLAM technique is able to understand not just the fact that the environment
0:43 - 0:48is uncertain but also the robot itself runs on an uncertain trajectory.
0:48 -That makes it hard.
- 06-15 Is Localization Necessary Solution
Other units in this course below:
Unit 1: http://www.youtube.com/playlist?list=PL1EF620FCB11312A6
Unit 2: http://www.youtube.com/playlist?list=PL107FD47786234011
Unit 3: http://www.youtube.com/playlist?list=PL5493E5D24A081719
Unit 4: http://www.youtube.com/playlist?list=PLAADAB4F235FE8D65
Unit 5: http://www.youtube.com/playlist?list=PL1B9983ACF22B1920
Unit 6: http://www.youtube.com/playlist?list=PLC9ED5AC39694C141
To gain access to interactive quizzes, homework, programming assignments and a helpful community, join the class at http://www.udacity.com
- Video Language:
- CS373 - Artificial Intelligence
|Amara Bot added a translation|