Outline
00:00 Introduction
01:24 Background
02:38 Getting into robotics
04:55 Programming at 12 years old
05:52 Getting into Stanford
07:30 Robotics at Stanford
12:19 From computer vision to motion planning
13:59 RRT-Connect algorithm
20:31 Is research in motion planning still valuable?
22:48 Will LLMs deprecate planning?
26:23 Timeline for truly intelligent robots
28:10 Breakthroughs required for AI
32:30 Cloud robotics
34:47 Going to Japan
37:04 Technological progress and craftsmanship in Japan
39:34 Work culture in Japan
43:49 Professor at Carnegie Mellon University
46:03 Collaborations at CMU
47:25 Self-driving project at Google
51:53 Coding mastery at google
53:51 Code search tool
55:09 Vi versus Emacs
56:27 Full self-driving
58:44 Who is ahead in self-driving technology?
01:01:06 Achieving level 5 autonomy
01:04:51 Behavior of robot cars and the median driver
01:06:30 Impact of self-driving on labor
01:10:37 Biggest problems for society
01:13:36 Woven City by Toyota
01:15:41 Job loss after Fukushima earthquake
01:18:01 A city from scratch
01:19:29 Climbing mount Fuji
01:20:42 Inventor garage
01:23:11 Startup incubator and beyond
01:23:57 Success definition
01:26:18 Daily habits
01:27:59 Pulling all-nighters
01:30:00 Advanced Shakespeare
01:30:58 Advice for early career researchers
01:31:59 Value of books
01:32:31 Linus Pauling: Change fields every 7 years
01:33:37 Next big project
01:34:34 Influential books
01:36:59 The value of biographies
01:38:23 Personal goal of 50 books per year
01:39:43 Biggest contribution to science