Robotics 2

Dynamic model of robots:

Analysis, properties, extensions, uses

Prof. Alessandro De Luca

 Analysis of inertial couplings
#!! 0
n Cartesian robot !=
0 #""

#!! #!"
n Cartesian “skew” robot != # #""
!"

#!! #!" (%" )

n PR robot !=
#!" (%" ) #""

#!! (%" ) #!" (%" )

n 2R robot !=
#!" (%" ) #""

n 3R articulated robot #!! (%" , %# ) 0 0

(under simplifying != 0 #"" (%# ) #"# (%# )
assumptions on the CoMs) 0 #"# (%# ) ###

Robotics 2 2
 Analysis of gravity term
n absence of gravity
n constant !! (motion on horizontal plane)
n applications in remote space
n static balancing ! " ≈0
n distribution of masses (including motors)
n mechanical compensation
n articulated system of springs
n closed kinematic chains

Robotics 2 3
 Bounds on dynamic terms
n for an open-chain (serial) manipulator, there always exist
positive real constants !0 to !7 such that, for any value of
" and "̇
!! ≤ % " ≤ !" + !# " + !$ " # inertia matrix
factorization matrix of
' ", "̇ ≤ !% + !& " "̇ Coriolis/centrifugal terms

) " ≤ !' + !( " gravity vector

n if the robot has only revolute joints, these simplify to

!! ≤ % " ≤ !" ' ", "̇ ≤ !% "̇ ) " ≤ !'
(the same holds true with bounds %$,&$' ≤ %$ ≤ %$,&() on prismatic joints)

NOTE: norms are either for vectors or for matrices (induced norms)
Robotics 2 4
 Robots with closed kinematic chains -1

Comau Smart NJ130 MIT Direct Drive Mark II and Mark III

Robotics 2 5
 Robots with closed kinematic chains -2

MIT Direct Drive Mark IV UMinnesota Direct Drive Arm

(planar five-bar linkage) (spatial five-bar linkage)
Robotics 2 6
 Robot with parallelogram structure
(planar) kinematics and dynamics

4 *+
3 center of mass:
"" − $ arbitrary **$
**# $ **, E-E
5 parallelogram:
1
**!
%# = %$
"#
%" = %%
2 "" #
**"
direct kinematics
%#(# %' cos "" − $ %#(# %'("
'&& = + = −
%#)# %' sin "" − $ %#)# %')"

position of center of masses

**! ,! **" ," *" ," * , *! ,! * ,
+*! = +*" = +*# = + *# ! +*, = − *, "
**! -! **" -" *" -" **# -! *! -! **, -"

Robotics 2 7
 Kinetic energy
linear/angular velocities
−**! -! −**# -! −*" -"
0*! = %̇ ! 0*# = %̇ ! + %̇ " 2! = 2# = %̇ !
**! ,! **# ,! *" ,"
−**" -" −*! -! **, -"
0*" = %̇ " 0*, = %̇ ! + %̇ 2" = 2, = %̇ "
**" ," *! ,! −**, ," "

Note: a (planar) 2D notation is used here!

# " "̇ " + 3 # " # #

*) 0# = " 1#%(# # "̇ 0" = 1 % " "̇ " + 3 "̇ "
" (#,** # " " (" " " (",** "

# ""̇ " " "̇ " # "

0$ = 1 %
" $ " "
+ %($ # + 2%"%($("+#"̇ #"̇ " + 3 "̇
" ($,** #
# #
0% = " 1% %#""̇ #" + %(%
" "̇ " − 2% % (
"
"
# (% "+#"̇ #"̇ " + " 3(%,** "̇ "

Robotics 2 8
 Robot inertia matrix
,
1 .
3 = 4 3$ = %̇ ! % %̇
2
$-!

% +- % %
-!",$$ + /" 0!" !&,$$ + /& 0!& + /' 0" symm
* + = % +- % %
/& 0% 0!& − /' 0" 0!' 5%(" -!%,$$ + /% 0!% !',$$ + /' 0!' + /& 0%

structural condition
+$ ,# ,6$ = +% ," ,6% (*)
in mechanical design

5(") diagonal and constant ⇒ centrifugal and Coriolis terms ≡ 0

mechanically DECOUPLED and LINEAR *"" 0 +̈ " 9"
= 9
dynamic model (up to the gravity term 8(")) 0 *%% +̈ % %

big advantage for the design of motion control laws!

Robotics 2 9
 Potential energy and gravity terms
from the 8 -components of vectors +*$
.)
!# = 1#8,%(#)# !" = 1"8,%(")"
!$ = 1$8, %")" + %($)# !% = 1%8, %#)# − %(%)"
,

9 = 4 9$
$-! gravity
.
;9 :/ #! **! + ## **# + #, *! ,! : % components
: % = = = ! ! are always
;% :/ #" **" + ## *" − #, **, ," :" %"
“decoupled”

in addition, +"" "̈ " + )" "" = 0" <$ are

(non-conservative) torques
when (*) holds +## "̈ # + )# "# = 0# performing work on %$

further structural conditions in the mechanical design lead to 8(") ≡ 0!!

Robotics 2 10
 Adding dynamic terms ...
1) dissipative phenomena due to friction at the joints/transmissions
n viscous, Coulomb, stiction, Stribeck, LuGre (dynamic)...

n local effects at the joints

n difficult to model in general, except for:

0:,) = −2:,) "̇ ) 0<,) = −2<,) sgn "̇ )

=* Ċ

in general: =) >?@(C)
̇
0
;-.// "̇ < 0
(component-wise too) =+,-.,-/0

Robotics 2 11
 Adding dynamic terms ...
2) inclusion of electrical actuators (as additional rigid bodies)
n motor = mounted on link = − 1 (or before) , with very few exceptions

n often with its spinning axis aligned with joint axis =

n (balanced) mass of motor included in total mass of carrying link

n (rotor) inertia is to be added to robot kinetic energy

n transmissions with reduction gears (often, large reduction ratios)

n in some cases, multiple motors cooperate in moving multiple links:

use a transmission coupling matrix > (with off-diagonal elements)

Unimation PUMA family
JT2 Motor

JT3 Motor
'"%"( 3
JT1 Motor
!"#$% 3

'"%"( 2
!"#$% 2

Mitsubishi RV-3S
Robotics 2 12
 Placement of motors along the chain
?̇&0
?̇&! rotor N
frame
motor 1 motor N

rotor 1
frame joint 2
link 1 link 2 link N
RFW RFN-1
(world frame)

RFN

RF0 RF1
link N - 1

link 0 motor 2 joint N ?̇&$ = @1$ ?̇$

(base) joint1
A$ = @1$ A&$

?̇&"
Robotics 2 13
 Resulting dynamic model
n simplifying assumption: in the rotational part of the kinetic
energy, only the “spinning” rotor velocity is considered
0
1 1 1 1 .
3&$ ̇
= B&$ ?&$ = B&$ @1$ %̇ $ = C&$ %̇ $"
" " "
3& = 4 3&$ = %̇ C& %̇
2 2 2 2
$-!
diagonal, > 0
n including all added terms, the robot dynamics becomes
moved to
the left ...
% " + 6G "̈ + 7 ", "̇ + ) " + 2: "̇ + 2< sgn "̇ = 8
motor torques
constant does NOT F* > 0, F1 > 0 (after
contribute to 5 diagonal reduction gears)

n scaling by the reduction gears, looking from the motor side

diagonal 7
/33 (+) 1 motor torques
-2 + diag ̈
M2 + diag P5 (+)+̈ 5 + Q +, +̇
O* = R2 (before
% L43
L43 reduction gears)
56"
Robotics 2 except the terms %!! 14
 Special actuation and associated coordinates
planar 2R robot with remotely driven forearm

n motor 1 moves link 1 by '#

n motor 2 at the base moves the absolute
angle '" of link 2
n derive the dynamic model from scratch
98% using the = coordinates
% : :̈ + 7 :, :̇ + ) : = 0T
98"
<" − <$ <# 7#S"
%(:) = < 7 <$
# #S"
−<# =#S" :̇## no more
7(:, :)̇ =
<# =#S" :̇"# Coriolis forces!
<% 7"
)(:) = < 7
& #
5" = cos W" 5% = cos W%
5%(" = cos W% − W" X%(" = sin W% − W"
Robotics 2 15
 Including joint elasticity
n in industrial robots, use of motion transmissions based on
n belts
n harmonic drives
n long shafts
introduces flexibility between actuating motors (input) and driven
links (output)
n in research robots, compliance in transmissions is introduced on
purpose for safety (human collaboration) and/or energy efficiency
n actuator relocation by means of (compliant) cables and pulleys
n harmonic drives and lightweight (but rigid) link design
n redundant (macro-mini or parallel) actuation, with elastic couplings
n in both cases, flexibility is modeled as concentrated at the joints
n in most cases, assuming small joint deformation (elastic domain)
Robotics 2 16
 Robots with joint elasticity

DLR LWR-III
with harmonic drives
Dexter
with cable transmissions

elastic
motor load/link
spring
(stiffness K)

Quanser Flexible Joint video Stanford DECMMA

(1-dof linear, educational) with micro-macro actuation
Robotics 2 17
 Dynamic model
of robots with elastic joints
n introduce 2> generalized coordinates
n % = D link positions
n ? = D motor positions (after reduction, ?$ = ?&$ /@1$ )
1 0
n add motor kinetic energy 0! to that of the links 01 = "̇ 5(")"̇
0 2
1 1 1 1 .
3&$ = B&$ ?&$ = B&$ @1$ ?$ = C&$ ?̇$"
̇ " " ̇"
3& = 4 3&$ = ?̇ C& ?̇
2 2 2 2
$-! diagonal, > 0
n add elastic potential energy !" to that due to gravity !! (")
n F = matrix of joint stiffness (diagonal, > 0)
" 0
1 ?&$ 1 "
1
92$ = F$ %$ − = F$ %$ − ?$ 92 = 4 92$ = % − ? . F % − ?
2 @1$ 2 2
$-!
n apply Euler-Lagrange equations w.r.t. (", @)
no external torques
2[ 2nd-order % " "̈ + 7 ", "̇ + ) " + > " − ? = 0 performing work on &
differential
equations 6G ?̈ + > ? − " = 8
Robotics 2 18
 Use of the dynamic model
inverse dynamics
n given a desired trajectory "- (B)
n twice differentiable (∃ %̈ 3 (J))

n possibly obtained from a task/Cartesian trajectory K3 (J), by

(differential) kinematic inversion
the input torque needed to execute this motion (in free space) is

8\ = % "\ + 6G "̈ \ + 7 "\ , "̇ \ + ) "\ + 2: "̇ \ + 2< sgn "̇ \

`
(in contact, with an external wrench) … − B]^_ "\ 2]^_,\
n useful also for control (e.g., nominal feedforward)
n however, this way of performing the algebraic computation (∀B) is
not efficient when using the Lagrangian modeling approach
n symbolic terms grow much longer, quite rapidly for larger D
n in real time, numerical computation is based on Newton-Euler method

Robotics 2 19
 State equations
direct dynamics

Lagrangian > differential

dynamic model % " "̈ + ( ", "̇ + ! " = , 2nd order
equations
C" " #a
defining the vector of state variables as C = = ∈ ℝ
C# "̇
state equations

Ċ " C# 0
Ċ = = −%S" (C ) 7 C , C + )(C ) + %S" (C ) 0
Ċ # " " # " "

= F(C) + G(C)0 2> differential

1st order
2D × 1 2D × D equations

" generalized
another choice... ED = 5(")"̇ EḊ = ... (do it as exercise)
momentum
Robotics 2 20
 Dynamic simulation
Simulink here, a generic 2-dof robot
block 7(", ")
̇ ", "̇
scheme %̇ ! (0) %! (0)
"̈ " "̇ "
""
input torque _
command +
(open-loop 0 !!" (#)
or in _ "̈ # "̇ #
feedback) "#

)(") " " %̇ " (0) %" (0)

including “inv(M)”
§ initialization (dynamic coefficients and initial state)
§ calls to (user-defined) Matlab functions for the evaluation of model terms
§ choice of a numerical integration method (and of its parameters)
e.g., 4th-order Runge-Kutta (ode45)
Robotics 2 21
 Approximate linearization
n we can derive a linear dynamic model of the robot, which is valid
locally around a given operative condition
n useful for analysis, design, and gain tuning of linear (or of the

linear part of) control laws

n approximation by Taylor series expansion, up to the first order

n linearization around a (constant) equilibrium state or along a

(nominal, time-varying) equilibrium trajectory

n usually, we work with (nonlinear) state equations; for mechanical

systems, it is more convenient to directly use the 2nd order model

n same result, but easier derivation

equilibrium state (", ")

̇ = ("2 , 0) [ "̈ = 0 ] ) "] = 0]
equilibrium trajectory (", ")
̇ = ("- (B), "̇ - (B)) [ "̈ = "̈ - (B) ]
% "\ "̈ \ + 7 "\ , "̇ \ + ) "\ = 0\
Robotics 2 22
 Linearization at an equilibrium state
n variations around an equilibrium state
" = "2 + Δ" "̇ = "̇ 2 + Δ"̇ = Δ"̇ "̈ = "̈ 2 + ∆"̈ = ∆"̈ ; = ;2 + Δ;

n keeping into account the quadratic dependence of % terms

on velocity (thus, neglected around the zero velocity)
M8
5 "2 ∆"̈ + 8 "2 + L ∆" + o ∆" , ∆"̇ = ;2 + ∆;
M" 131
9
infinitesimal terms
of second or higher order
N("2 )
∆"
n in state-space format, with ∆C =
∆"̇
̇ 0 I 0
∆C = S" ∆C + S" ∆0 = J ∆C + 6 ∆0
−% "] G("] ) 0 % ("] )
Robotics 2 23
 Linearization along a trajectory
n variations around an equilibrium trajectory
" = "- + Δ" "̇ = "̇ - + Δ"̇ "̈ = "̈ - + ∆"̈ ; = ;- + Δ;

n developing to 1st order the terms in the dynamic model ...

5("- + ∆") "̈ - + ∆"̈ + (("- + ∆", "̇ - + ∆")
̇ + 8 "- + ∆" = ;- + ∆;
4 (-th row of the
M5. identity matrix
5 "- + ∆" ≅ 5 "- +P L Q.0 ∆"
M" 131
.3# :

8 "- + ∆" ≅ 8 "- + N("- )∆" M! (%3 , %̇ 3 )

M( M(
( "- + ∆", "̇ - + ∆"̇ ≅ ( "- , "̇ - + L ∆" + L ∆"̇
M" 131: M"̇ 131:
13
̇ 1̇ : 13
̇ 1̇ :

M" (%3 , %̇ 3 )
Robotics 2 24
 Linearization along a trajectory (cont)

n after simplifications …
%("! )∆"̈ + 0" ("! , "̇ ! )∆"̇ + 1 "! , "̇ ! , "̈ ! ∆" = ∆,
with a
O%)
K "\ , "̇ \ , "̈ \ = G "\ + L" "\ , "̇ \ +M N "̈ \ P)`
O" cbc
)b" #
n in state-space format
0 3
∆2̇ = ∆2
−%#$ "! 1 "! , "̇ ! , "̈ ! −%#$ "! 0" "! , "̇ !
0
+ #$ ∆, = 5(6) ∆2 + 7(6) ∆,
% ("! )
a linear, but time-varying system!!

Robotics 2 25
 Coordinate transformation
" ∈ ℝ% % " "̈ + ( ", "̇ + ! " = % " "̈ + : ", "̇ = ,& 1

if we wish/need to use a new set of generalized coordinates '

; ∈ ℝ% ; = <(") " = < #$ (;) by duality

(principle of virtual work)
=<
;̇ = "̇ = >(")"̇ "̇ = >#$ (");̇ ,& = >' ("),( 1
="
̇ "̇
;̈ = > " "̈ + >(") ̇
"̈ = >#$ (") ;̈ − >(")> #$ (");̇

̇
% " >#$ " ;̈ − %(")>#$ (")>(")> #$ (");̇ + : ", "̇ = >' ("),
(

>#' (") ? pre-multiplying the whole equation...

Robotics 2 26
 Robot dynamic model
after coordinate transformation

̇
R+0 " 5 " R+# " '̈ + R+0 " S ", "̇ − 5(")R+#(")R(")R +#(")'̇ = ;
6

for actual computation,

these inner substitutions (", ")
̇ → (:, :)̇
"→: are not strictly necessary
non-conservative
generalized forces
!# & &̈ + %# &, &̇ + +# & = -# performing work on +

symmetric,
%T = S` S"
B %B positive definite )T = BS` )
(out of singularities)
quadratic
7T = BS` 7 − %BS" B ̇ BS" :̇ = BS` 7 − %T B ̇ BS" :̇ dependence on +̇

7T (:, :)̇ = 'T (:, :)̇ :̇ %̇ T − 2'T skew-symmetric

when ' = E-E pose, this is the robot dynamic model in Cartesian coordinates
NOTE: in this case, we have implicitly assumed than ! = D (no redundancy!)
Robotics 2 27
 Example of coordinate transformation
planar 2R robot using absolute coordinates

n motor 1 at joint 1, motor 2 at joint 2

n in place of DH angles T, use the absolute
angles '# = "# and '" = "# + ""

:= 1
&" 0 "=B" a linear
<4" 1 1 transformation

<4! 1 0 1 −1
BS" = BS` =
= &# −1 1 0 1
n from 5 " "̈ + ( ", "̇ + 8 " = ;1
obtained with DH relative coordinates
blue terms are the same found in a direct way in slide #15
U# − U$ U"("+# U%(#
56 (') = R+0 5 R+# = U ( U$ 86 (') = R+0 8
= U (
" "+# ' "
" ;1# − ;1"
−U )
" "+# "' ̇
(6 (', ')̇ = R+0 ( = +0
;6 = R ;1 = ;1"
U")"+# '̇ #"
Robotics 2 28
 Robot dynamic model
in the task/Cartesian space, with redundancy

dynamic model in the joint space " ∈ ℝ4 second-order task kinematics

5 " "̈ + S ", "̇ = ] X = Y " ∈ ℝ7 ̇ "̇
Ẍ = R " "̈ + R(")
5<> N is full rank = !

1) isolate the joint acceleration from the dynamics "̈ = 5+# " ] − S ", "̇
2) decompose the joint torques in two complementary spaces
] = R0 " ^ + 3 − R0 " \(") ], O is a generalized inverse of N.
∈ ℛ R0 ∈ [ R0 \ N. ON. = N.
torques coming from … and joint torques R; ∉ ℛ f<
generalized forces F
in the task space … A/ = N. % P, ∀P ∈ ℝ5 ⇒ B − N. % O % N. % P = 0
3) substitute 1) and 2) in the differential task kinematics
Ẍ = R " 5+# " R0 " ^ + 3 − R0 " \(") ], − S ", "̇ ̇ "̇
+ R(")
4) isolate on the right-hand side the generalized forces P in the task space …
Robotics 2 29
 Robot dynamic model
in the task/Cartesian space, with redundancy
6! . 6!
N % ! % N % K̈ = P +
6!
N % ! 6! % N. % N % ! 6! % B − N. % O(%) A/ − @ %, %̇ ̇ %̇
+ N(%)
6! . 6! 6! # .
5) choose as generalized inverse O = N! N N! = N5 ,
i.e., the transpose
of the inertia-weighted pseudoinverse of the task Jacobian (see block of slides #2)
in this way, the joint torque component R; will NOT affect the task acceleration ḧ
6! 6!
N % ! 6! % N. % K̈ = P + N % ! 6! % N. % N ̇ % %̇ − N % ! 6! % @ %, %̇

6) the resulting (! −dimensional) task dynamics is then

external forces can be added
%g " T̈ + Ug ", "̇ = 2 … + 2]^_ on the rhs of the equations in
with a dynamically consistent way!
+#
58 " = R " 5+# " "R0 task inertia matrix for * = [ , these terms
S8 ", "̇ = 58 " ̇ "̇
R " 5+# " S ", "̇ − R(") are identical to slide #27

7) an additional D − ! -dimensional second-order dynamics is needed to describe

the full robot!
Robotics 2 30
 Dynamic scaling of trajectories
uniform time scaling of motion

n given a smooth original trajectory "- (B) of motion for B ∈ [0, 0]

n suppose to rescale time as B → X(B) (a strictly increasing function of J )

K
X(0)
B=0 K = 2J
(uniformly)
slower!
K(3) K = K(J)
3 K=J
generic X(B)
B
K = 0.5J
(uniformly)
faster!
0 3 J

Robotics 2
BX=00 31
 Dynamic scaling of trajectories
uniform time scaling of motion

n in the new time scale, the scaled trajectory "/ (X) satisfies
W"\ W"i WT
"\ V = "i T(V) "̇ \ V = = = "ij (T) T(V)
̇
same path executed
WV WT WV
(at different instants of time)

W"̇ \ W"ij WT j
WṪ
"̈ \ V = = Ṫ + "i = "ijj (T)Ṫ # (V) + "ij (T)T(V)
̈
WV WT WV WV
n uniform scaling of the trajectory occurs when X B = `B

"̇ \ V = !"ij (!V) "̈ \ V = ! # "ijj (!V)

Q: what is the new input torque needed to execute the scaled trajectory?
(suppose dissipative terms can be neglected)

Robotics 2 32
 Dynamic scaling of trajectories
inverse dynamics under uniform time scaling

n the new torque could be recomputed through the inverse dynamics, for
every X = `B ∈ 0, 0/ = [0, `0] along the scaled trajectory, as
8i !V = % "i "ijj + 7 "i , "ij + )("i )
n however, being the dynamic model linear in the acceleration and
quadratic in the velocity, it is
]- B = 5 "- "̈ - + ( "- , "̇ - + 8 "- = 5 "/ ` ""/99 + ( "/ , `"/9 + 8("/ )
= ` " 5 "/ "/99 + ( "/ , "/9 + 8 "/ = ` " ]/ `B − 8 "/ + 8 "/
n thus, saving separately the total torque ]- (B) and gravity torque 8- (B)
in the inverse dynamics computation along the original trajectory, the
new input torque is obtained directly as
* > 1: slow down
1 ⇒ reduce torque
@) A6 = " @! 6 − ! "! (6) + ! "! (6) * < 1: speed up
A ⇒ increase torque
gravity term (only position-dependent): does NOT scale!
Robotics 2 33
 Dynamic scaling of trajectories
numerical example

n rest-to-rest motion with cubic polynomials for planar 2R robot under gravity
(from downward equilibrium to horizontal link 1 & upward vertical link 2)
n original trajectory lasts 3 = 0.5 s (but say, it violates the torque limit at joint 1)
only joint 1 torque is shown
torque only due
to non-zero initial
acceleration

total torque gravity torque

component

at equilibrium
= zero gravity
torque

for both joints

Robotics 2 34
 Dynamic scaling of trajectories
numerical example

original gravity torque n scaling with k = 2 (slower) → l = = 1 s

total to sustain the link
torque at steady state

8! 0.1 − = >! 0.1 = 20 Nm

#$
8" 2 A 0.1 − = >" 2 A 0.1 = = 5 Nm
#"
$ = 0.5 s
*
gravity torque
component * *
does not scale 0 Nm
scaled
total
torque 1
4

+ = 0.5 s 1=2 +=1s

$! = 1 s
Robotics 2 35
 Optimal point-to-point robot motion
considering the dynamic model

§ given the initial (⇒ A) and final (⇒ B) robot configurations (at

rest) and the actuator torque bounds, find
n the minimum-time Tmin motion

n the (global/integral) minimum-energy Emin motion

and the associated command torques needed to execute them

§ a complex nonlinear optimization problem solved numerically
video video

Tmin= 1.32 s, E = 306 T = 1.60 s, Emin = 6.14

Robotics 2 36