Introduction to Artificial Intelligence and Robots

Introduction to Artificial Intelligence

1. AI for Robotics HS1501 • Lecture B Tan Minh Nguyen Department of Mathematics

2. Review: AI and NLP

3. What is AI? Artificial intelligence is the development and study of computer systems to address complex real-world problems typically associated with some form of intelligence

4. What is NLP?

5. Robotics

6. Rocky area next to river bed Recovery from leg obstruction

7. External Disturbance

8. What is a robot?

9. What is a robot? A robot is an intelligent system that interacts with the physical environment through sensors and effectors. Program module? Web crawling ‘bot? Robot Environment sensors effectors

10. History of Robotics

11. The First Wave of Robots

12. The Second Wave of Robots DARPA Grand Challenge PR1 Robot

13. Robots Today Robot Hands Boston Dynamics

14. Robot Learning

15. What is Robot Learning

16. What is Robot Learning Definition 1 : The study of AI algorithms and principles with their applications to robotics problems Definition 2 : The study of methods and principles that make robot learn from data Definition 3 : The research field at the intersection of AI and robotics

17. Self-driving cars are still not a product Yet a 16 year old can learn to drive a car in less than 20 hours

18. A twelve year old can do all these kitchen tasks, today no robot can

19. Challenges: What is the problem here?

20. Challenges: Moravec’s paradox Moravec (1988): It is comparatively easy to make computers exhibit adult level performance on intelligence tests or playing checkers, and difficult or impossible to give them the skills of a one-year-old when it comes to perception and mobility. Pinker (1994): The main lesson of thirty-five years of AI research is that the hard problems are easy and the easy problems are hard.... As the new generation of intelligent devices appears, it will be the stock analysts and petrochemical engineers and parole board members who are in danger of being replaced by machines. The gardeners, receptionists, and cooks are secure in their jobs for decades to come.

21. But why? Moravec’s argument was that we will have more difficulty reverse-engineering skills that are the result of hundreds of millions of years of evolution I think a better argument is that of available digitized data on the web for training machine learning models

22. Robot Learning: Current Progress

23. Major Capabilities of Robotics/Robot Learning

24. Sensing: Navigation Tasks for Mobile Robots

25. Sensing

26. Sensing

27. Sensing

28. Navigation in Cluttered Indoor Setting

29. Navigation in Outdoor Setting

30. Navigation While indoors, AI can use Bluetooth signals, Wi-Fi signals, and mobile phone signals for navigation. While outdoors, AI can use satellite signals for navigation via the Global Positioning System (GPS).

31. Physical Activities

32. Interacting with Other Machines

33. Interacting with Human

34. Teleoperation from Human Video

35. When NOT to Make Robots Learn? Learning is not a solution to every problem in robotics

36. When to Make Robots Learn? Learning is critical for taking robots to the real world

37. Now is a good time to study and work on Robot Learning

38. Now is a good time to study and work on Robot Learning

39. Robot Learning as a Growing Research Community

40. Some Challenges 1. To protect the safety and the privacy of people, governments have strict regulations on where robots can operate. 2. In unfamiliar situations, the behavior of robots may be hard to predict. 3. The technology for imitating emotions is still not mature.

41. Reflection We saw some current capabilities of AI robots and how powerful/restricted they are. Supposing that you are to get a physical AI robot for your home free-of-charge next month, what capabilities would you like it to have and not have? The automation of which operations within the NUS using AI-powered RPA would be most beneficial to you now? Do you think robots will be a threat to humans (like in the films and otherwise)? In what sense?