Tutorial 1

Section I: Multiple-Choice

Tick √ the box which you think is the best answer.

1. What is the number of degrees of freedom of the block (see figure below) if it is
constrained by a straight rigid rod through a frictionless slot?

Rigid rod

Frictionless slot

2
3
5
None of the above.

2. A DC motor converts:
mechanical energy into electrical energy.
electrical energy into thermal energy.
electrical energy into potential energy.
None of the above.

3. The use of gear transmission in DC motor is to change output:

torque.
position.
power.
None of the above.

4. Which of the followings is(are) the characteristic(s) of a rotation matrix used to represent
the orientation of a 3D object?

I. Its determinant is always equal to one.

II. Its inverse always exists.
III. Its row vectors are all unit vectors.

I and II only.
I and III only.
I, II and III.
None of the above.

1
Section II:

1. How many DOF are there in the human arm (exclude DOF in the hand)? Please break
down the number into shoulder, elbow and wrist joints.
Solution:
Shoulder: 3 DOF, elbow: 1 DOF, wrist: 3 DOF. => 7 DOF in total
Note that this is just an approximate model.

2. Obtain the inverse kinematics equations for the 2 DOF planar manipulator shown below.
That is, to express 𝜃𝜃1 and 𝜃𝜃2 as functions of x and y (desired coordinates of endpoint in
frame xoyo. Comment on the typical number of solutions for the inverse kinematics
problem.

Solution:

From forward kinematic equations (refer to lecture slides):

x(θ1 , θ 2 ) =l1 cos θ1 + l2 cos(θ1 + θ 2 )

y (θ1 , θ 2 ) =l1 sin θ1 + l2 sin(θ1 + θ 2 )

Let 𝑐𝑐1 = 𝑐𝑐𝑐𝑐𝑐𝑐𝜃𝜃1 ; 𝑐𝑐2 = 𝑐𝑐𝑐𝑐𝑐𝑐𝜃𝜃2 ; 𝑠𝑠1 = 𝑠𝑠𝑠𝑠𝑠𝑠𝜃𝜃1 ; 𝑠𝑠2 = 𝑠𝑠𝑠𝑠𝑠𝑠𝜃𝜃2 ; 𝑐𝑐12 = 𝑐𝑐𝑐𝑐𝑐𝑐(𝜃𝜃1 + 𝜃𝜃2 ); 𝑠𝑠12 =
𝑠𝑠𝑠𝑠𝑠𝑠(𝜃𝜃1 + 𝜃𝜃2 ).

Hence, we can rewrite the forward kinematic equations as follows:

𝑥𝑥 = 𝑙𝑙1 𝑐𝑐1 + 𝑙𝑙2 𝑐𝑐12 (1)

𝑦𝑦 = 𝑙𝑙1 𝑠𝑠1 + 𝑙𝑙2 𝑠𝑠12 (2)

2
Taking square for both equations:
𝑥𝑥 2 = 𝑙𝑙12 𝑐𝑐12 + 𝑙𝑙22 𝑐𝑐12
2
+ 2𝑙𝑙1 𝑙𝑙2 𝑐𝑐1 𝑐𝑐12
2 2 2 2 2
𝑦𝑦 = 𝑙𝑙1 𝑠𝑠1 + 𝑙𝑙2 𝑠𝑠12 + 2𝑙𝑙1 𝑙𝑙2 𝑠𝑠1 𝑠𝑠12
Adding the above equations:
𝑥𝑥 2 + 𝑦𝑦 2 = 𝑙𝑙12 𝑐𝑐12 + 𝑙𝑙12 𝑠𝑠12 + 𝑙𝑙22 𝑠𝑠12
2
+ 𝑙𝑙22 𝑐𝑐12
2
+ 2𝑙𝑙1 𝑙𝑙2 𝑐𝑐1 𝑐𝑐12 + 2𝑙𝑙1 𝑙𝑙2 𝑠𝑠1 𝑠𝑠12
2 2
= 𝑙𝑙1 + 𝑙𝑙2 + 2𝑙𝑙1 𝑙𝑙2 𝑐𝑐2
That is,
𝑥𝑥 2 + 𝑦𝑦 2 − 𝑙𝑙12 − 𝑙𝑙22
𝑐𝑐2 =
2𝑙𝑙1 𝑙𝑙2
𝑠𝑠2 = ±�1 − 𝑐𝑐22

𝑥𝑥 2 + 𝑦𝑦 2 − 𝑙𝑙12 − 𝑙𝑙22
𝜃𝜃2 = 𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴2(𝑠𝑠2 , 𝑐𝑐2 ) = 𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴2 �±�1 − 𝑐𝑐22 , �
2𝑙𝑙1 𝑙𝑙2
(3)

Remark: Atan2(y,x) computes tan-1(y/x) and uses both x and y to determine the
quadrant in which the angle lies. E.g. Atan2(-1, -1) = - 135o, whereas Atan2(1,1) = 45o.

Now, assume 𝜃𝜃2 can be found from equation (3), we can obtain 𝜃𝜃1 from equations (1)
and (2):
𝑥𝑥 = 𝑙𝑙1 𝑐𝑐1 + 𝑙𝑙2 𝑐𝑐12 = 𝑙𝑙1 𝑐𝑐1 + 𝑙𝑙2 (𝑐𝑐1 𝑐𝑐2 − 𝑠𝑠1 𝑠𝑠2 ) = (𝑙𝑙1 + 𝑙𝑙2 𝑐𝑐2 )𝑐𝑐1 − 𝑙𝑙2 𝑠𝑠1 𝑠𝑠2 (4)
𝑦𝑦 = 𝑙𝑙1 𝑠𝑠1 + 𝑙𝑙2 𝑠𝑠12 = 𝑙𝑙1 𝑠𝑠1 + 𝑙𝑙2 (𝑠𝑠1 𝑐𝑐2 + 𝑐𝑐1 𝑠𝑠2 ) = 𝑙𝑙2 𝑠𝑠2 𝑐𝑐1 + (𝑙𝑙1 + 𝑙𝑙2 𝑐𝑐2 )𝑠𝑠1 (5)

In matrix-vector form:
𝑥𝑥 −𝑙𝑙2 𝑠𝑠2 𝑙𝑙1 + 𝑙𝑙2 𝑐𝑐2 𝑠𝑠1
�𝑦𝑦� = � � �𝑐𝑐 �
𝑙𝑙1 + 𝑙𝑙2 𝑐𝑐2 𝑙𝑙2 𝑠𝑠2 1
Solving:

𝑠𝑠1 1 𝑙𝑙2 𝑠𝑠2 −(𝑙𝑙1 + 𝑙𝑙2 𝑐𝑐2 ) 𝑥𝑥

�𝑐𝑐 � = � � �𝑦𝑦�
1 −(𝑙𝑙2 𝑠𝑠2 ) − (𝑙𝑙1 + 𝑙𝑙2 𝑐𝑐2 ) −(𝑙𝑙1 + 𝑙𝑙2 𝑐𝑐2 )
2 2 −𝑙𝑙2 𝑠𝑠2
𝑠𝑠1 1 𝑙𝑙2 𝑠𝑠2 −(𝑙𝑙1 + 𝑙𝑙2 𝑐𝑐2 ) 𝑥𝑥
�𝑐𝑐 � = � � �𝑦𝑦�
1 −(𝑙𝑙1 + 𝑙𝑙2 + 2𝑙𝑙1 𝑙𝑙2 𝑐𝑐2 ) −(𝑙𝑙1 + 𝑙𝑙2 𝑐𝑐2 )
2 2 −𝑙𝑙2 𝑠𝑠2

Hence,
−𝑙𝑙2 𝑠𝑠2 𝑥𝑥 + (𝑙𝑙1 + 𝑙𝑙2 𝑐𝑐2 )𝑦𝑦
𝑠𝑠1 =
𝑙𝑙12 + 𝑙𝑙22 + 2𝑙𝑙1 𝑙𝑙2 𝑐𝑐2
𝑙𝑙2 𝑠𝑠2 𝑦𝑦 + (𝑙𝑙1 + 𝑙𝑙2 𝑐𝑐2 )𝑥𝑥
𝑐𝑐1 = 2
𝑙𝑙1 + 𝑙𝑙22 + 2𝑙𝑙1 𝑙𝑙2 𝑐𝑐2

−𝑙𝑙2 𝑠𝑠2 𝑥𝑥 + (𝑙𝑙1 + 𝑙𝑙2 𝑐𝑐2 )𝑦𝑦 𝑙𝑙2 𝑠𝑠2 𝑦𝑦 + (𝑙𝑙1 + 𝑙𝑙2 𝑐𝑐2 )𝑥𝑥
𝜃𝜃1 = 𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴2(𝑠𝑠1 , 𝑐𝑐1 ) = 𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴2 � , 2 �
𝑙𝑙12 + 𝑙𝑙22 + 2𝑙𝑙1 𝑙𝑙2 𝑐𝑐2 𝑙𝑙1 + 𝑙𝑙22 + 2𝑙𝑙1 𝑙𝑙2 𝑐𝑐2
(6)

In general, there are 2 solutions for a given set of x and y. This is due to the fact that
equation (3) typically yields two solutions for 𝜃𝜃2 and each of which will yield a value
for 𝜃𝜃1 based on equation (6).

3
3. A robot weighs 1 kg and is driven by two independent fixed wheels (each with radius, r
= 2cm) on the same axle. Assume that the robot needs to climb a slope with 20 degree
gradient. If it also needs to achieve a constant speed of 0.2m/s while climbing up the
slope, calculate the required torque for each motor when the robot is climbing the slope.
Based on a factor of 3 for power requirement, compute the desired power rating of each
motor. Assume that frictional forces are negligible for both wheel mechanisms.

Solution:

Using the same model covered in the lecture notes:

The required applied force (from the motors) on the robot, Fapp, is given by (assume frictional
force Ff = 0):

𝐹𝐹𝑎𝑎𝑎𝑎𝑎𝑎 = 𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚 = 1 × 9.81 × 𝑠𝑠𝑠𝑠𝑠𝑠20𝑜𝑜 = 3.36𝑁𝑁

Torque requirement for each motor (at the gear output shaft):

𝐹𝐹𝑎𝑎𝑎𝑎𝑎𝑎 𝑟𝑟 3.36 × 0.02

𝑇𝑇 = = = 0.0336𝑁𝑁𝑁𝑁
2 2

The mechanical power requirement for each motor:

𝐹𝐹𝑎𝑎𝑎𝑎𝑎𝑎 3.36
𝑃𝑃𝑚𝑚 = 𝑣𝑣 = × 0.2 = 0.336𝑊𝑊
2 2

With a factor 3, the power rating for each motor = 3 × 𝑃𝑃𝑚𝑚 = 1.008𝑊𝑊

4
4. Given the specification data of a PMDC motor as below, complete the table below
and sketch the current-torque curves for the motor before and after a geartrain with
gear ratio G equal to 9.

Parameters (with units) Value

Rated (measuring) voltage, V (in V) 6
No-load speed, No (in rpm)
No-load speed, No (in rad/s)
Stall torque, Ts (in Nm)
No-load current, Io (in A)
Torque constant, kt (in Nm/A)
Back-emf constant, ke (in V/1000rpm)
Back-emf constant, ke (in Vs/rad)
Stall current, Is (in A)

5
Solution:

Parameters (with units) Value

Rated (measuring) voltage, V (in V) 6
No-load speed, No (in rpm) 8400
No-load speed, No (in rad/s) 880
2𝜋𝜋
(= 𝑁𝑁𝑜𝑜 (𝑖𝑖𝑖𝑖 𝑟𝑟𝑟𝑟𝑟𝑟) × )
60
Stall torque, Ts (in Nm) 3x10-3
No-load current, Io (in A) 7x10-3
Torque constant, kt (in Nm/A) 6.7x10-3
Back-emf constant, ke (in V/1000rpm) 0.7
Back-emf constant, ke (in Vs/rad) 6.68x10-3
𝑘𝑘 (𝑖𝑖𝑖𝑖 𝑉𝑉/1000𝑟𝑟𝑟𝑟𝑟𝑟)
(= 𝑒𝑒 2𝜋𝜋 )
1000×
60
Stall current, Is = Ts/kt (in A) 0.448

I (A)

0.448

Before
geartrain

After geartrain

0.007
3 10 20 27 30
T (mNm)

Remark: The stall torque after the gearbox is typically obtained from the stall torque from
the motor multiplied by the gear ratio and also the gearbox efficiency (assumed to be 100% in
this problem).