S28-3

STUDY ON MODELING AND CONTROL OF EXCAVATOR

Le Quang Hoan1, Chan Se Jeong1, Hack Sun Kim1, He Lim Yang1, and Soon Yong Yang2*
1

Graduate School of Mechanical and Automotive Engineering, University of Ulsan, Ulsan, Korea
2
Department of Mechanical and Automotive Engineering, University of Ulsan, Ulsan, Korea
* Corresponding author (soonyy@ulsan.ac.kr)

ABSTRACT: An excavator is a typical hydraulic heavy-duty human-operated machine used in general versatile
construction operations, such as digging, carrying loads, grounding and dumping loads. However, operating excavators in
the hazard environment such as earthquake, nuclear decomposition etc is not suitable for human to operate on site.
Therefore, automation in excavator has been investigated to protect the operators from the harmful environments. To find a
feasible way to control excavator, firstly the Full Kinematic and dynamic model of the excavators manipulator which three
degrees of freedom were studied. The inverse kinematic equations are used to determine the joint angles and the actuator
lengths corresponding to a specific position and orientation of the bucket, given in the base coordinate system. To control
the excavators manipulator, the computed torque control are using and some simulation results are given.
Keywords: Excavator, Modeling, Automation, Control, Manipulator
1. INTRODUCTION

bucket tracking the desired trajectory. In this paper, the

Excavation is an importance work in mining, earth removal

computed torque control and PID control are designed.

and general earthworks. Excavation productivity (amount

Some simulations are proposed to verify the controller.

of work done), efficiency (cost of work done in terms of

labor and machinery) and operator safety, particularly in

2.

underground mining or during the removal of hazardous

MAINIPULATIOR

waste, are constantly under pressure from industry. After

decades of increases in machine size and power, practical

2.1. Kinematic
The kinematic and dynamic models of the experimental D

limits are now being approached and automation is being

Company solar 105 hydraulic excavator will be presented.

sought for further improvements. Furthermore, computing

These models have been built to describe the experimental

and sensing technologies have reached to a stage where

excavator as a robotic manipulator with 3 degrees of

they can be applied to automatic excavation. Nowadays,

MODELING

OF

THE

EXCAVATORS

freedom. In computer-controlled motion, the bucket of the

automation in excavators has been studied by several

excavator follows a path specified by its position and angle,

researchers and so on. Among the fundamental element

i.e. the pose of the bucket, which corresponds to specific

used as a basis for developing complicated tasks, the

Control of excavators manipulator is an important and

values of angular positions of the joint shafts. The values

basic task in autonomous research of hydraulic excavator.

of these joint variables are determined by the lengths of

To control an excavators manipulator, firstly, the Full

hydraulic actuators. The mathematical relations between

Kinematic and dynamic model of the excavators

these variables are given by the kinematic relations of the

manipulator which three degrees of freedom were studied

excavator arm.

in [1]. The trajectory of the bucket is designed and the

desired angle joint of each links are determined by using
Inverse Kinematic. The controller is designed to control the

969

S28-3

O
1 2 O
P0 2 A0 A1 P2 2
a2c1c2 a1c1 a2c2 s1 a1s1 a2 s2

(4)

Fig. 1

Schematic diagram of an excavator

The forward kinematics is used to describe the

positions and orientations of the points on the excavator in
the Cartesian coordinate for the given joint positions
during the digging operation. The problem can be

c
1
O3
O
s
1 2 3
P0 A0 A1 A2 P3 3 1

a3c23 a2c2 a1
a3c23 a2c2 a1

c
1
O4
O
s
1 2 3 4
P0 A0 A1 A2 A3 P4 4 1

a4c234 a3c23 a2c2 a1

a4c234 a3c23 a2c2 a1

Where ci = cos

summarized as below:
For the given 1 2 3 4 , find the coordinate

a3 s23 a2 s2
1

a3 s23 a3 s23 a2 s2
1

(5)

(6)

I, si = sin I, 23= 2+ 3,

234= 2+ 3+ 4.

To determine the positions of the points on the

excavator in the base Cartesian coordinate frame, the
relations between the fixed coordinate system and other
coordinate

systems

is

necessary.

Therefore,

the

transformation matrix relating two adjacent coordinate

frames was studied by Koivo [2] as follows:

Aii1

cos i
sin
i

cos i sin i

sin i sin i

cos i cos i
sin i

sin i cos i
cos i

Where

ai cos i
ai sin i
di

1 (1)

is the twist angle of link I, ai is the

length of link I, and di is the offset distance in link I, I = 1,

Fig. 2

2, 3, 4.

The mainipulators workspace of the Solar 105

2.2 Inverse kinematics

By the given coordinates of the origin in each

The inverse kinematic equations are used to

coordinate frame Oi, the coordinates of points Oi in the

determine the joint angles and the actuator lengths

base coordinate frame can be described as follows using

corresponding to a specific position and orientation of the

the equation (1):

bucket, given in the base coordinate system.

P0Oi A0i Pi Oi

(2)

Where P0O 0 0 0 1T specifies point Oi in the

i

ith coordinate frame. From equation (2) we can describe

the origin of each coordinate frame Oi in the base
coordinate frame as follows:
(3)

970

S28-3

Where
is the vector of measured joint angles as
defined in Fig.2
represents inertia
represents Coriolis and centripetal effects
represents gravity forces
( ) represents frictions
is the corresponding input matrix
specifies the torques acting on the
Fig. 3

joint shafts

Projection of the links onto the vertical plane

It is assumed that the coordinate of point O4 N

represents the interactive torques between the bucket

and the environment during the digging operation.

are given in the coordinate system. According to Tafazoli

[3], the inverse kinematic model of the excavator is given
as follows:

3. CONTROL OF EXCAVATOR

1 b
l1

(7)

rb a1

to make the end of effecter track desired trajectory. Using

2
zb

(8)

1 a tan zb / ( rb a1 )

The control system shown in figure 4 is designed

2
2
l1 a4 2la4 cos 1

the dynamic model of excavator in (15), the primary

controller is given below:

(9)

U d D v C , G B

(10)

(15)

Where
2 a sin ( zb a4 sin 4 ) / l

2 2
2
2 2 a cos a2 l a3 / 2a2l

(11)

2
2 2
3 a cos a2 a3 l / 2a2 a3

v d kv e k p e ; e d is the error kv and kp are

the linear gains to be determined.

(12)

D is the estimated inertia, C , is the estimated

(13)

4 23

Coriolis and centripetal effects, G is the estimated

(14)

gravity forces; B is the estimated friction effects; Ud is

2.3 Modeling of the Excavators manipulator

The Dynamic model of excavator can be

expressed concisely using the form of the well-known

the desired joint link angle, angular velocity and angular

rigid-link manipulator equations of motion:

D C , G B FL

the desired torque provided to the system; d , d , d are

acceleration.
(15)

A secondary controller is designed to compensate

for derivations of the actual motion from the desired

971

S28-3

trajectory [2].
Inverse dynamics
Ud(t)

COMPUTED TORQUE
CONTROL
d(t)

+
+

EXCAVATOR

(t)

Upid(t)

Fig.4

U(t)

PID CONTROL

Control architecture
(c)

4. SIMULATION
Fig.5

A simulation study on an excavator was

Horizontal movement (m): (a) Trajectory of the

performed in Matlab/Simulink. A desired trajectory is first

bucket; (b) x-coordinate Error of bucket Position versus

designed in the Casterian space. The joint angles of the

time; (c) z-coordinate Error of bucket Position versus time.

boom, arm and bucket are determined by using Inverse

The error of the angular joint of boom, arm and

bucket are shown in the figure 6. The errors of all the links

Kinematic.

of system when the bucket is moving are less than 1 degree.

In figure 5 shows the result of the simulation

when the trajectory of the bucket is moving horizontal.

(a)

(a)

(b)

(b)

972

S28-3

(c)
Fig. 6

The error of the angular joint

(c)

(a) Boom; (b) Arm; (c) Bucket (degree)

Fig.7
Another simulation is shown in the figure 7 when

Vertical movement (m): (a) Trajectory of the

bucket; (b) x-coordinate Error of bucket Position versus

the end of effecter is moving vertical.

time; (c) z-coordinate Error of bucket Position versus time.

(a)
(a)

(b)
(b)

973

S28-3

[4]Tafazoli, S., Salcudean, S.E., Hashtrudi-Zaad, K. and

Lawrence,

P.D.

(2002)

Impedance

control

of

teleoperated excavator, IEEE Trans. on Control Systems

Technology, May, Vol. 10, No. 3, pp.355367.
[5] Hongnian Yu*, Yang Liu and Mohammad Shahidul
Hasan, Review of modelling and remote control for
Excavators, Int. J. Advanced Mechatronic Systems, Vol. 2,
Nos. 1/2, 2010.
[6] Yang Liu, Mohammad Shahidul Hassan, Hong-Nian Yu,
Modeling

and

remote

control

of

an

Excavator,

International Journal of Automation and Computing, 7(3),

August 2010, 349-358.

(c)

[7] Ha, Q., Santos, M., Nguyen, Q., Rye, D. and DurrantFig. 8

Whyte, H. (2002) Robotic excavation in construction

The error of the angular joint (a) Boom; (b) Arm;

(c) Bucket (degree)

automation, IEEE Robotics & Automation Magazine,

The error of the angular joint of the boom, arm and bucket

March, Vol. 9, No. 1, pp.2028.

when the trajectory of the end of effecter is moving vertical

[8] Ha, Q.P., Nguyen, Q.H., Rye, D.C. and Durrant-Whyte,

is shown in figure 8. The errors of all the links of system

H.F. (2000) Impedance control of a hydraulically-actuated

when the bucket is moving are less than 0.5 degree. The

robotic excavator, Automat. Construction, Vol. 9, No. 5,

simulations illustrate that the controller performs very well

pp.421435.
[9] ZHANG Da-qing, HE

during the system is operating.

Qing-hua,

HAO

Peng.

Trajectory tracking control of hydraulic excavator bucket

5. CONCLUSION

[J]. Journal of Jilin University: Engineering and

The paper has reported the review of the Kinematic and

Technology Edition, 2005, 35(5): 490-494.

modeling of the Excavators Manipulator. The basic

controller is computed torque control and PID control are
applied to the system and some simulations are proposed to
test the controller.
References
[1] Koivo, A.J. (1994) Kinematics of excavators
(backhoes) for transferring surface material, J. Aerosp.
Eng., Vol. 7, No. 1, pp.1732.
[2] Koivo, A.J., Thoma, M., Kocaoglan, E. and AndradeCetto, J. (1996) Modeling and control of excavator
dynamics during digging operation,

J. Aerosp. Eng.,

January, Vol. 9, No. 1, pp.1018.

[3] Tafazoli. S, P. D. Lawrence, S. E. Salcudean.
Identification of inertial and friction parameters for
excavator arms. IEEE Transactions on Robotics and
Automation, vol. 15, no.5, pp.996-971, 1999.

974