Scribd
Upload
7 views3 pages

Pset 7

Uploaded by

asusephex
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
7 views3 pages

Pset 7

Uploaded by

asusephex
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

UZB 438E: Robotic Control Systems

Fall 2024

Problem Set 7
Due: 8am, Tuesday, December 3, 2024

In the following problems, use the two-link planar elbow arm dynamics:

(m1 + m2 )a21 + m2 a22 + 2m2 a1 a2 cos θ2 m2 a22 + m2 a1 a2 cos θ2 θ̈1


  
+
m2 a22 + m2 a1 a2 cos θ2 m2 a22 θ̈2
−m2 a1 a2 sin θ2 (2θ˙1 θ˙2 + θ̇22 )
 
+ (1)
m2 a1 a2 sin θ2 θ̇12
   
(m1 + m2 )ga1 cos θ1 + m2 ga2 cos(θ1 + θ2 ) τ
= 1 .
m2 ga2 cos(θ1 + θ2 ) τ2

Moreover, consider the filtered-error

e =qd − q,
(2)
r =ė + Λe,

where qd is the desired trajectory and Λ = diag{λ1 , λ2 } is a positive definite matrix.

In the following problems, use the arm parameters m1 = 0.8, m2 = 2.3, a1 = 1, a2 = 1.


Moreover, plot the tracking results for
   
θ1d (t) sin t
qd (t) = = . (3)
θ2d (t) cos t

Problem 1
Consider the two-link planar elbow arm (1), and let f = M (q)(q̈d + Λė) + Vm (q, q̇)(q̇d +
Λe) + G(q). Design and simulate in MATLAB the Computed-Torque Control Variant:

τ = f + Kr, (4)

where K and Λ are design parameters chosen appropriately. Plot the tracking results for
the given qd in eq. (3), by showing θ1 (t), θ1d (t) and θ2 (t), θ2d (t). Plot also the control signal
τ1 and τ2 .
(10 pts)

Problem 2
Reconsider Problem 1, and assume that exact mass values are unknown, and that only the
range info 0 ≤ m1 ≤ 5 and 0 ≤ m2 ≤ 5 is available.

Design and simulate in MATLAB the following robust controller

τ =fˆ + Kr − v,
(5)
v = − (f¯ + η) sgn(r),

1
where fˆ = M̂ (q)(q̈d + Λė) + V̂m (q, q̇)(q̇d + Λe) + Ĝ(q), η > 0, and f¯ is a scalar function
chosen such that ∥f − fˆ∥1 = ∥f˜∥1 ≤ f¯. Both M̂ (q) and V̂m (q, q̇) are as in (1) with m̂1 and
m̂2 used instead. The values for m̂1 and m̂2 are a design choice that you may decide as long
you are within the specified mass ranges.

Plot the tracking results for the given qd in eq. (3), by showing θ1 (t), θ1d (t) and θ2 (t),
θ2d (t). Plot also the control signal τ1 and τ2 .

Hint: You need to find f¯ as follows

∥f˜∥1 ≤ ∥M (q) − M̂ (q)∥1 ∥q̈d + Λė∥1 + ∥Vm (q, q̇) − V̂m (q, q̇)∥1 ∥q̇d + Λe∥1 + ∥G(q) − Ĝ(q)∥1
≤ µ̃2 ∥q̈d + Λė∥1 + ṽB ∥q̇∥1 ∥q̇d + Λe∥1 + g̃b = f¯.
(6)
Check Example 3.3-1 on page 137 of [1] to see the derivation of the following bounds:

∥M (q)∥1 ≤ (m1 + m2 )a21 + 2m2 a22 + 3m2 a1 a2 ≜ µ2


∥Vm (q, q̇)∥1 ≤ m2 a1 a2 ∥q̇∥1 ≜ vb ∥q̇∥1 , (7)
∥G(q)∥1 ≤ (m1 + m2 )ga1 + 2m2 ga2 ≜ gb ,

(10 pts)

Problem 3
Reconsider Problem 2, where in this case we neither know the exact values of m1 and m2
nor do we have range info available.

Design and simulate in MATLAB the following adaptive controller

τ = W (q, q̇, q̇d , q̈d )ϕ̂ + Kr (8)


˙
ϕ̂ = Γ W ⊺ (q, q̇, q̇d , q̈d )r, (9)

where Γ is a design parameter chosen appropriately, and

f = M (q̈d + Λė) + Vm (q, q̇)(q̇d + Λe) + F (q̇) + G(q)


= W (q, q̇, q̇d , q̈d )ϕ, (10)
 
m1
ϕ= . (11)
m2

Plot the tracking results for the given qd in eq. (3), by showing θ1 (t), θ1d (t) and θ2 (t),
θ2d (t). Plot the control signal τ1 and τ2 , and plot the parameter estimates m̂1 (t) and m̂2 (t).
(10 pts)

Problem 4
In estimating the convergence rate for an asymptotically stable linear system, the following
expression applies:
q 1 λmin (Q)
(P ) − 2 λmax (P ) t
∥x(t)∥2 ≤ λλmaxmin (P )
e ∥x(0)∥2 . (12)
Numerically choose an asymptotically stable linear system ẋ = Ax of your choice.
1. Compute the convergence rate estimate for Q = I.

2. Compute the convergence rate estimate for any other Q > 0 of your choice.

2
3. Of the two Q values used, which one returns the largest convergence rate estimate?
(10 pts)

Solution submission:

• It is best to receive your solution in PDF format, in addition to the MATLAB source
files used.
• Upload to: https://www.dropbox.com/request/gMnkOfve53ynjEsKzC8k .

References
[1] F. Lewis, D. Dawson, and C. Abdallah, Robot Manipulator Control: Theory and Prac-
tice (Automation and Control Engineering). CRC Press, 2003, isbn: 9780203026953.
[Online]. Available: https://doi.org/10.1201/9780203026953.

You might also like