Introduction to Humanoid Robotics by Dr. Rawichote Chalodhorn (Choppy)

Transcription

1 Introduction to Humanoid Robotics by Dr. Rawichote Chalodhorn (Choppy) Humanoid Robotics Lab, Neural System Group, Dept. of Computer Science & Engineering, University of Washington.

2 RoboCup soccer

3 The brain controlled humanoid robot

4 Outline History of humanoid robots Humanoid robots today Androids Analytical approaches of bipedal locomotion Learning to walk through imitation Future of humanoid robotics

5 Atomaton: Leonardo's robot 1495

6 Atomaton: The Japanese tea serving doll 18th century to 19th century

7 Honda E series

8 Honda P series P1 P2 P

9 AIST / Kawada Industries : HRP series HRP-4C

10 Androids A robot that closely resembles a human Human robot interaction Prof. Hiroshi Ishiguro

11 Zero-Moment Point (ZMP) [Vukobratovic et al. 1972]! ZMP d R! eaves the support polygon!

12 Derivation of the ZMP [Takanishi et al., 1989] 2 % 3 #! $ :! 9 8 & 6,!& & 0 &. ', ' * ' ' % % 3 : 9 # 9, + $ ( 5 &! ' ( * 6 + % ; 5 ; - A&E234(42$&9'0(25&2=&' *, : & 3! )$! & ' ' 3! ($ *% % + %. ( A&(2(#)&(25?'&#0(4$%&#(&'- 6 5 % 4 * + : : :... % :, ( &A&'F('5$#)&727'$(&:( * < < < % 9 + ( &A&'F('5$#)&=250'&9( # ' * = = = % 9 * ( A&E234(42$&B1'5'&+ 9 &43&#0(4$%& 6 5 4

13 Derivation of the ZMP $ +.$ + + ' ' : +.$ +.$ #! ) (! ) $ ) ) % * & ! 2 9!! ( ' : % 2 8 # 63! 4! 3 & 7 '!!! 4 63 & $ 8!!! 4 & 7! ' 0 35) & ( 2 % 2 8 ' 3 : %!!! 2 ) ( ' ! 2! ) 9 (!!! 7 ) # 5!! $ 3 4! 3 & 7 '!! 4 53 & % 83!!! 43 & 7! ' 0 36) & ( ' ' 1 5 # $ ) ) % * & ! 2. ( ' ' 3 : '!!! 2 ) ( ' ' 1 7!! [Takanishi et al., 1989] [Huang et al., 2001]

14 ZMP Jacobian Compensation [Sobota et al., 2003; Wollherr et al., 2003] $ % &!#$%&'%#()!! # +,$!! +,$! % # # +,$ 4! +,$!! +,$ $! +,$ 4! +,$ % 4! $!!! #! #! +,$ $ +,$ 4!! % 0$! $! +,$ +,$ $ +,$ :.! + / $ ; M $ ; 8, % - :!!! $ # U #!! +,$ 5 $ +,$ +,$!! 4 $ - +,.# +,$. 7 8.#. 7 +,$ 6 * ( )! $! 4 $ , * ( ( ( ( )! $! 4

15 The Cart-Table Model τ zmp = mg(x p x ) mẍz c = 0 [Kajita et al. 2003] m ẋ z c O p x x τ p y = y z c g ÿ, z p x = x z c g ẍ ÿ = g z c y 1 mz c τ x, ẍ = g z c x + 1 mz c τ y he mass of the pen p y = τ x mg p x = τ y mg τ

16 ZMP Model-based Feedback controller configuration Balancer Module [Kajita et al. 2003] Gyro Accelerometer Kalman filter Body inclination Force/torque sensor encoder ZMP calculator ZMP Stabilizer Joint servo References of Joint angles, body inclination and ZMP

17 Learning Approach [Chalodhorn et al. 2006]

18 Dimension reduction [Grochow et al. 2004]

19 ^V& =+$-+1K$56& *$-3'-/& #+$*#& \& +0-',20& :& -$*$+$%56& +'& 1/*-'K$&+0$&/'%$5&*-$%1.+'-J&(%&.01$K$&1(.-$#1(256& '*+1/5&/'+1'(8& & /#! 012&!(3&24251&2(6!.%78*&7! & 2 1 Learning through low-dimensional spaces Gyroscope signal of X-axis Original : RMS = Optimized : RMS = [Chalodhorn et al. 2005] 0 4 Original data Search space Embed function Optimized data 2 1 Gyroscope signal of Y-axis Original : RMS = Optimized : RMS = Gyroscope signal of Z-axis Original : RMS = Optimized : RMS = & 9128&^8&W'/*-1#'(&'3&26-'#.'*$&#12(5#&3-'/&1(1+15&(%&'*+1/14$%&)57&E0$& *5'+#&3-'/&+'*&+'&!'++'/&#0')&+0$&26-'#.'*$&#12(5#&3'-&+0$&Z$#&@J_J&(%&`& -$.'-%$%&%,-1(2&1(1+15&(%&'*+1/14$%&)571(2&/'+1'(#8&B''+&/$(&#P,-$%& &

20 Learning by imitation results

21 Learning by imitation results

22 Intrinsically low-dimensional humanoid motion

23 Conclusion Humanoid robots are used as a research tool in several scientific areas. Although the initial aim of humanoid research was to build better orthosis and prosthesis for human beings.

24 Future of Humanoid robots

25 THANK YOU