19
Trajectory Planning for Robots: The Challenges of Industrial Considerations Joonyoung Kim 1,2 and Elizabeth A. Croft 2 1 Hyundai Heavy Industries Co. Ltd., S. Korea 2 The University of British Columbia, Canada 1/19 ICRA 2014

Trajectory Planning for Robots: The Challenges of Industrial

  • Upload
    lythien

  • View
    222

  • Download
    3

Embed Size (px)

Citation preview

Page 1: Trajectory Planning for Robots: The Challenges of Industrial

Trajectory Planning for Robots:The Challenges of Industrial Considerations

Joonyoung Kim1,2 and Elizabeth A. Croft2

1Hyundai Heavy Industries Co. Ltd., S. Korea2The University of British Columbia, Canada

1/19ICRA 2014

Page 2: Trajectory Planning for Robots: The Challenges of Industrial

2

‐ Multi‐purpose and efficient (6DOF, easy to program, fast with good repeatability) ‐> provides various applications (different from other automated machines)

‐ Trends: more intelligence (vision & force sensors), safety (collision, human‐robot interaction)competitive market (cost reduction) → light weight design (less stiff) → the high motion performance is dependent on control methods (the main theme)

Hyundai Motors, Czech Republic LG Display, S. Korea

‐ 1.5 million industrial robots in the world (2013, IFR); used in large industries (automotive, display)

Introduction market, applications, and trends

ICRA 2014

Page 3: Trajectory Planning for Robots: The Challenges of Industrial

ICRA 2014 3

Purpose

‐ To introduce a typical motion planning method for industrial robots

‐ To review trajectory planning methods proposed in the literature(related to motion performance of industrial robots)

‐ To review some common but the most important constraints for generic industrial robot controllers

‐ To explain why some of well‐known methods are not used in practice

Assumptions‐ 6DOF industrial robots‐ A generic robot controller (architecture, servo control loop)‐ A good trajectory: 

; simple (implementation, maintenance); efficient (small CPU burden); being able to address important constraints; maximizes motion performance (high speed and high accuracy)   

Page 4: Trajectory Planning for Robots: The Challenges of Industrial

4

0S1S

2S

3S

S1    MOVE  P, S= 100%,  A=5,  T=1

S2    MOVE  L, S= 80%,  A=3,  T=1

S3    MOVE  L, S=100%,  A=0,  T=1

Teach Pendant

Control module

Planning module(path & trajectory)

S1

S2

S3

Robot Moving

ProgramRunning

Trajectory Generator

Reference Input

Robot ControllerRobot Task Space

HHI (Hyundai Heavy Industries Co.., Ltd.) robot system

Introduction motion generation

ICRA 2014

Page 5: Trajectory Planning for Robots: The Challenges of Industrial

5

fast (time-optimal)

Control

accurate tracking

Equal at all times!

planning

V

t

low speed

Introduction An ideal motion control for industrial robots

Path‐Invariant Time‐Optimal Motion

Requirements for trajectories‐ time‐optimal‐ generate a path independent of trajectories‐ band‐limited   

Page 6: Trajectory Planning for Robots: The Challenges of Industrial

6

HHI’s HS165 (6DOF, payload: 165kg) 

Task space

joint space

3 (3)SO

6

vector space curved space (Lie Group)(not a vector space)

3: , (3)nf SO

Multi‐dimensional nonlinear map

f

n 1f

3 (3)SO

Challenges kinematics

Multi‐dimensional, nonlinear

ICRA 2014

‐ Need to address both positions and orientations‐ Orientations: much more complex than positions 

Kinematic constraints: path, maximum speed/acceleration/jerk  

Page 7: Trajectory Planning for Robots: The Challenges of Industrial

7

Multi‐dimensional, nonlinear, highly coupled 

12( ) ( , ) ( ) τ M q q + B q q q + N q

m mJ

mq rqr

k rJ

mb

each joint

Equation of motion

Inertia‐ position‐dependent‐ coupled

Coriolis‐Centrifugal and friction‐ Pos. & vel. dependent‐ coupled

Joint compliance,gravity‐ position dependent‐ coupled

Natural frequencies at steady‐state: 3 ~ 25 Hz (1~2nd mode) (very low, position dependent)

Joint (link) positions: not measurable!

Harmonic Drive (r=100 ~ 200)

Challenges dynamics

ICRA 2014

Constraints‐maximum torques (hard constraint)‐ not exciting natural modes 

Page 8: Trajectory Planning for Robots: The Challenges of Industrial

8

Complex geometry + complex dynamics (too complicated)

Divide & Conquer (Planning / Control) 

‐ Increase speed; time‐optimal trajectory planning

‐ minimize vibrations; smooth motion trajectory planning

‐ Track accurately, respond quickly, ‐ Obtain stable motions

; various control methods (linear, nonlinear, SISO, MIMO) 

Planning Task (goals) Control Task (goals)

Traditional Approach

ICRA 2014

Page 9: Trajectory Planning for Robots: The Challenges of Industrial

9

Time‐Optimal Trajectory Planning

Shin and McKay (1985) Bobrow et al. (1985) 

Constantinescu and Croft (2000)  

3( ) [ , ]{ | 0 1}

f ss s

A parameterized path

s

s

( , )o os s ( , )f fs sforward in time backward in time

Phase Plane

switching point

Find the time‐optimal trajectory

Pros: problem simplified by the path parameter very close to true time‐optimal trajectories

Cons: high CPU burden, infinite jerk (in general),dynamic model never perfect 

not considered

Well‐Known Methods Time‐optimal trajectories

ICRA 2014

EOM ( ) ( , ) ns s s s τ m c

max max( ) ( , )s s s s τ m c τ Constraints

max,max,

( , )min

( )i i

acci

c s ss

m s

Page 10: Trajectory Planning for Robots: The Challenges of Industrial

10

Meckl and Woods (1985)Singer (1990)

Macfalane and Croft (2003)Kroger and Walh (2010) 

‐ Argument ; time‐optimal: too complicated, high frequencies (advanced control methods required); smooth trajectories can obtain faster settling times due to small residual vibrations. 

Pros: small oscillation, no special controllers neededCons: slow (no dynamics) 

too many cases for non‐zero boundary conditionscannot completely suppress high freq.natural freq. must be known 

0 0.2 0.4 0.6 0.8 1 1.2 1.40

200

400

600

800

1000

1200

1400t-dp trajectory

t [s]

dp [mm/s]

0 5 10 15 20 25 300

1

2

3

4

Frequency (Hz)

|Y(f)

|

TVP

S‐curveTVP

S‐curve

Well‐Known Methods Smooth‐Trajectories

ICRA 2014

Page 11: Trajectory Planning for Robots: The Challenges of Industrial

Challenges online planning

Online trajectory planning: A robot’s trajectory must be planned (replanned) during motion for any (stationary) target given kinematic and dynamic constraints.

Non-zero boundary conditions

S1

replan

torquesaturation

S2 input

11

Trapezoid Velocity

S‐curves or higher polynomials?‐ too many cases to consider‐ prone to generate s/w errors‐> difficult to implement and maintain 

ICRA 2014

Page 12: Trajectory Planning for Robots: The Challenges of Industrial

bad good

V 10 %

V 100 %

V 10 %

V 100 %

Speed change = collision

12

Challenges path and speed

ICRA 2014

Path must be the same regardless of speed changes.

Page 13: Trajectory Planning for Robots: The Challenges of Industrial

S1 S2

S3Stop/Restart inManual Playback Mode

StopMinimum distance to the original path

Robot position at power‐failure

Motion command at power‐failure

Return to the path

S1

S2

Position after Power‐failure

Power‐failure during motion

Path unchanged

return

jogging

Planned path

Real path

13

Challenges path invariance for all operation conditions

ICRA 2014

A programmed path must not be changed under any circumstances unless such arequest is made by the user.

Page 14: Trajectory Planning for Robots: The Challenges of Industrial

No powerfailure

Power failure Path Recovery

Path RecoveryImagine you are in charge of this factory!And there is a chance of black-out!

14ICRA 2014

Page 15: Trajectory Planning for Robots: The Challenges of Industrial

15

TVP S‐curve

0 t

( )r t

aT aT

A A

0 t

( )r t

jT jT

J J

2( ) 1 ajTTVP

AR j e

3( ) 1 jjTS curve

JR j ej

Frequency Analysis via Fourier Transform

( ) 0m x k x r

2

2 2

( ) ( )( )

n

n

X j H jR j

2

2 2 2

( ) ( ) ( )

1 a

TVP TVP

jTn

n

X j H j R j

A e

2

3 2 2

( ) ( ) ( )

1 j

S curve S curve

jTn

n

X j H j R j

J ej

‐ Big A, Big J ‐> large Oscillation‐ TVP Larger vibration than S‐curve‐ Vibration can be 0 at some ‐ S‐curves still have high freq.

0 5 10 15 20 25 30 35 40 45 50-160

-140

-120

-100

-80

-60

-40

-20

0

frequency [Hz]

power

spe

ctru

m o

f R(jw

)

,a jT T

FRF of TVP

Trajectories Vs. Residual Vibrations

ICRA 2014

Page 16: Trajectory Planning for Robots: The Challenges of Industrial

16

,1rq

,2rq1L

,2cL

2L

,1cL

2m

1m

X

Y

,1cJ

,2cJm mJ

mq rqr

k rJ

mb

Parameter Joint 1 Joint 2 Parameter Joint 1 Joint 2

Mass [kg] 80 170 k [Nm/rad] 1010700 505350

Jr [kg.m2] 9.97 13.7 bm [Nm s/rad] 0.0281 0.0088

L [m] 0.87 1.05 Max. motor torque [Nm] 49 35.86

Lc [m] 0.38 0.2 r (gear ratio) 201 145

Robot: 2DOF industrial Robot with flexible jointsInitial position: qr0 = [90, ‐90] [deg]Natural frequencies: 7 Hz, 21 Hz at qr0

Simulation dynamic model

Joint model

ICRA 2014

Page 17: Trajectory Planning for Robots: The Challenges of Industrial

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 189

90

91

92

93

94

95

96

97

time [s]

qr1 [d

eg]

TVP (infinite jerk) 

S‐curve (large jerk – tuned at the lowest natural frequency, 7 Hz)

S‐curve (small jerk)

0 0.2 0.4 0.6 0.8 1-50

0

50

m,cm

d 1 [Nm

]

Torque saturation

Oscillation at 7 Hz (1st mode)

‐ Control performance is affected by trajectories.‐ Trajectories must be band‐limited.‐ Such limitation of the band cannot be   determined arbitrarily.

Simulation trajectories Vs. vibrations

Controller: Decentralized PD controllers

First joint position

17ICRA 2014

Page 18: Trajectory Planning for Robots: The Challenges of Industrial

Simulation trajectories Vs. controller

0 0.1 0.2 0.3 0.4 0.588

90

92

94

96

time [s]

q r1 [d

eg]

0 0.1 0.2 0.3 0.4 0.5-20

0

20

40

60

80

time [s]

dqr1

[deg

/s]

0 0.1 0.2 0.3 0.4 0.5-96

-94

-92

-90

-88

time [s]

q r2 [d

eg]

0 0.1 0.2 0.3 0.4 0.5-80

-60

-40

-20

0

20

time [s]

dqr2

[deg

/s]

‐ Large tracking errors‐ Large oscillations ‐> Decentralized and linear control methods do not work well.‐> A more advanced control method is needed.

18

Trajectory: TVPController: decentralized

& linear  full‐state feedback (linear state‐observer)

fixed gain by pole‐placement

Page 19: Trajectory Planning for Robots: The Challenges of Industrial

19

Conclusions

‐ Trajectory planning for industrial robots is challenging.; online planning, safety, efficiency, complex kinematics and dynamics 

‐ Traditionally, a good trajectory is regarded as either time‐optimal or smooth.

‐ Time‐optimal trajectories; high CPU burden and high frequency components

‐ Smooth trajectories; too slow, still complex (high order polynomials), vague jerk limitation 

‐ What is the best trajectory for industrial robots?; not clear but TVP is still widely used!; it depends on servo controller. 

ICRA 2014