Mecatrónica y Robótica de Servicio: Teoría y Aplicaciones

226 Pages·2016·13.05 MB·English

(cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:9)(cid:3)(cid:4)(cid:10)(cid:11)(cid:10)(cid:12)(cid:13)(cid:14)(cid:7)(cid:5)(cid:9)(cid:3)(cid:4)(cid:10)(cid:15)(cid:2)(cid:10)(cid:16)(cid:2)(cid:6)(cid:17)(cid:9)(cid:3)(cid:9)(cid:13)(cid:18)
(cid:19)(cid:2)(cid:13)(cid:6)(cid:20)(cid:4)(cid:10)(cid:11)(cid:10)(cid:21)(cid:22)(cid:23)(cid:9)(cid:3)(cid:4)(cid:3)(cid:9)(cid:13)(cid:8)(cid:2)(cid:24)
(cid:25)(cid:26)(cid:10)(cid:27)(cid:4)(cid:24)(cid:5)(cid:9)(cid:23)(cid:23)(cid:13)(cid:10)(cid:27)(cid:4)(cid:24)(cid:5)(cid:4)(cid:28)(cid:2)(cid:15)(cid:4)(cid:29)(cid:10)(cid:30)(cid:26)(cid:21)(cid:26)(cid:10)(cid:31)(cid:9)(cid:28)(cid:13)(cid:10)(cid:16) !(cid:6)(cid:2)"(cid:29)(cid:10)
(cid:25)(cid:26)(cid:10)(cid:1)(cid:13)(cid:6)(cid:4)(cid:23)(cid:2)(cid:24)(cid:10)(cid:16)!(cid:8)(cid:3)#(cid:2)"(cid:29)(cid:10)$(cid:26)(cid:10)(cid:25)(cid:26)(cid:10)%(cid:4)(cid:6)&(cid:4)(cid:24)(cid:10)(cid:16)(cid:13)(cid:5)(cid:13)(cid:10)
(cid:11)(cid:10)$(cid:26)(cid:1)(cid:26)(cid:10)(cid:12)(cid:4)’(cid:13)(cid:24)(cid:10)(cid:21)(cid:6)(cid:6)(cid:2)& (cid:20)(cid:8)
(cid:21)(cid:24)(cid:13)(cid:3)(cid:9)(cid:4)(cid:3)(cid:9)(cid:7)(cid:8)(cid:10)(cid:1)(cid:2)((cid:9)(cid:3)(cid:4)(cid:8)(cid:4)(cid:10)(cid:15)(cid:2)(cid:10)(cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:9)(cid:3)(cid:4)(cid:10)(cid:21)(cid:26)(cid:27)(cid:26)
Rob(cid:243)tica y Mecatr(cid:243)nica de Servicios: Teor(cid:237)a y Aplicaciones.
Editado por Eduardo Castillo Castaæeda, Paola Andrea Niæo SuÆrez,
Eduardo Morales SÆnchez, JosØ Emilio Vargas Soto
y Juan Manuel Ramos Arregu(cid:237)n.
Publicado por
Asociaci(cid:243)n Mexicana de Mecatr(cid:243)nica, A. C.
' Los editores y autores
Rob(cid:243)tica y Mecatr(cid:243)nica de Servicios: Teor(cid:237)a y Aplicaciones, es un libro digital autorizado por el
Instituto Nacional de Derechos de Autor bajo el nœmero de radicaci(cid:243)n 291032 a la Asociaci(cid:243)n
Mexicana de Mecatr(cid:243)nica A. C., Calle Fonolog(cid:237)a, No. 116, Col. Tecnol(cid:243)gico C.P. 76158,
QuerØtaro Qro. Tel.[phone]. www.mecamex.net, las opiniones y la informaci(cid:243)n que
se muestran en los cap(cid:237)tulos del libro son exclusivas de los autores y no representan la postura
de la Asociaci(cid:243)n Mexicana de Mecatr(cid:243)nica A.C. Fecha de la œltima modificaci(cid:243)n: 29 de
Noviembre del 2016. Esta obra es una publicaci(cid:243)n de acceso abierto, distribuido bajo los
tØrminos de la Asociaci(cid:243)n Mexicana de Mecatr(cid:243)nica A. C., la cual permite el uso, distribuci(cid:243)n y
reproducci(cid:243)n sin restricciones por cualquier medio, siempre y cuando los trabajos estØn
apropiadamente citados, respetando la autor(cid:237)a de las personas que realizaron los cap(cid:237)tulos.
Primera edici(cid:243)n, Noviembre 2016
Impreso en QuerØtaro, MØxico.

ISBN 978-607-9394-06-6
Rob(cid:243)tica y Mecatr(cid:243)nica de Servicios: Teor(cid:237)a y Aplicaciones.

ISBN 978-607-9394-06-6, Asociaci(cid:243)n Mexicana de Mecatr(cid:243)nica, A. C.

˝ndice

Cap(cid:237)tulo 1.
An estimation acoustic transducer based on sliding modes control. 1
Julio CØsar Tovar Rodr(cid:237)guez, Floriberto Ort(cid:237)z Rodr(cid:237)guez y
Carlos RomÆn Mariaca Gaspar.

Cap(cid:237)tulo 2
Regularized divide and conquer training for dendrite morphological 9
neurons.
Erik Zamora y Humberto Sossa.

Cap(cid:237)tulo 3
Wall-bot: the hexapod robot for inspection. 18
E. C. Orozco Magdaleno, E. Castillo Castaæeda, F. A. Aguirre Cerrillo,
J. Franco Acuæa, K. H. FloreÆn Aquino y E. Francisco Agust(cid:237)n.

Cap(cid:237)tulo 4
Effects of a reconfiguration on kinematic performance of a 33
delta-type parallel robot.
Albert L. Balmaceda-Santamar(cid:237)a y Eduardo Castillo-Castaæeda.

Cap(cid:237)tulo 5
Robot b(cid:237)pedo antropomØtrico con dieciocho grados de libertad. 41
Enfoque mecÆnico.
D. Alvarado Rivera, L. G Corona Ram(cid:237)rez y J. S. Muæoz Reina.

Cap(cid:237)tulo 6
Sliding mode tracking control for an inertia wheel pendulum 51
around its unstable open-loop equilibrium point.
Luis T. Aguilar, Andhers N. Piæa y Daniel I. Aparicio.

Cap(cid:237)tulo 7
On adaptive control of a permanent magnet synchronous motor. 59
J. Moreno(cid:150)Valenzuela, Y. Quevedo-Pillado, R. PØrez-Aboytes
y L. GonzÆlez(cid:150)HernÆndez.

Cap(cid:237)tulo 8
A master-slave synchronization of 3 dof robot manipulators based on super- 70
twisting algorithm.
Jesœs Mart(cid:237)nez, Omar Amaya, Oscar Salas, Susana GutiØrrez,
Salvador A. Rodr(cid:237)guez y Jesœs de Le(cid:243)n.
Rob(cid:243)tica y Mecatr(cid:243)nica de Servicios: Teor(cid:237)a y Aplicaciones.

ISBN 978-607-9394-06-6, Asociaci(cid:243)n Mexicana de Mecatr(cid:243)nica, A. C.

˝ndice

Cap(cid:237)tulo 9
Integraci(cid:243)n de una plataforma y una grœa como prototipo para
rehabilitaci(cid:243)n de niæos con PCI.
Rodr(cid:237)guez-MØndez Jessica E., Ramos-Saavedra Kassandra G., 89
Ambriz-Sandoval Karen A., HernÆndez-Oliva Noemi
y Alejandre-Flores Marisol.

Cap(cid:237)tulo 10
Control por par calculado para un seguidor solar de dos 98
grados de libertad.
Sergio Isai Palomino-Resendiz, Diego Alonso Flores HernÆndez,
Alberto Luviano JuÆrez, Norma Lozada Castillo e Isaac Chairez.

Cap(cid:237)tulo 11
Synthesis and analysis of a fractional-order sliding mode control for 111
mechanical systems.
Carlos A. Rodriguez, Luis T. Aguilar, Alejandra Ferreira y and Eusebio Bugar(cid:237)n.

Cap(cid:237)tulo 12
Micro algoritmos genØticos en arquitecturas de sistemas 121
embebidos para la autonom(cid:237)a de marcha en robots.
F. A. ChÆvez Estrada, J. C. Herrera Lozada, J. Sandoval GutiØrrez
y M. I. Cervantes Valencia.

Cap(cid:237)tulo 13
Sistema prototipo de adquisici(cid:243)n de imÆgenes estereosc(cid:243)picas. 133
J. A. Olvera-Balderas, J. C. Sosa-Savedra y M. A. Oloæo Garc(cid:237)a.

Cap(cid:237)tulo 14
Desarrollo de un sistema para comunicaci(cid:243)n utilizando EOG. 140
Miguel I. Ceballos PØrez y Eduardo Morales SÆnchez.

Cap(cid:237)tulo 15
Diseæo de un robot m(cid:243)vil para desplazamiento en terreno agreste. 147
J. Alejandro Aguirre Anaya y O. Octavio GutiØrrez Fr(cid:237)as.

Cap(cid:237)tulo 16
Determinaci(cid:243)n anal(cid:237)tica del balance estÆtico de un robot b(cid:237)pedo. 157
Christian Alberto Matilde Dom(cid:237)nguez y Eduardo Morales SÆnchez.
Cap(cid:237)tulo 17
Algoritmo de visi(cid:243)n embebido para condiciones de iluminaci(cid:243)n 163
no controladas.
J. Contreras, J. `lvarez y J. Herrera.
Rob(cid:243)tica y Mecatr(cid:243)nica de Servicios: Teor(cid:237)a y Aplicaciones.

ISBN 978-607-9394-06-6, Asociaci(cid:243)n Mexicana de Mecatr(cid:243)nica, A. C.

˝ndice

Cap(cid:237)tulo 18
Exoesqueleto para asistencia en terapias de rehabilitaci(cid:243)n de movimientos 169
de los dedos de la mano.
Aguilar-Pereyra J. Felipe y Castillo-Castaæeda Eduardo.

Cap(cid:237)tulo 19
Reference trajectory generation for a knee assistive device. 181
B. Chaparro-Rico y E. Castillo-Castaæeda.

Cap(cid:237)tulo 20
Propuesta de un robot m(cid:243)vil para la detecci(cid:243)n de fuentes radioactivas. 192
A. De la Barrera Gonzalez, E. G. HernÆndez Mart(cid:237)nez, J. V. Cervantes Bazan
y J. A. Monterrubio Suarez.

Cap(cid:237)tulo 21
Towards an adaptable human-machine interface for autonomous 200
visual navigation for UAV in unstructured environments.
Cesar Omar Orozco L(cid:243)pez y JosØ Gabriel Ram(cid:237)rez Torres.

Cap(cid:237)tulo 22
Dispositivo rob(cid:243)tico coadyuvante para la rehabilitaci(cid:243)n 211
de dedos de la mano
A. Zapatero GutiØrrez, J. F. Rodr(cid:237)guez Le(cid:243)n,
J. F. Aguilar Pereyra y E. Castillo Castaæeda.
Rob(cid:243)tica y Mecatr(cid:243)nica de Servicios: Teor(cid:237)a y Aplicaciones. Cap(cid:237)tulo 1, pÆginas 1-8.
ISBN 978-607-9394-06-6, Asociaci(cid:243)n Mexicana de Mecatr(cid:243)nica, A. C.
An estimation acoustic transducer based on sliding modes control
Julio CØsar Tovar Rodr(cid:237)guez, Floriberto Ort(cid:237)z Rodr(cid:237)guez, Carlos RomÆn Mariaca Gaspar
Abstract (cid:151) Electronic sensor devices in geophysical processes are required to measure and automate different tasks. Throughout
history, people have created multiple type devices, but acoustics have an important application such as the content form
description in deep wells, watersheds, lakes, caves, among others. The acoustic signal is capable of reflecting where other types of
signals cannot operate, either by drawbacks or where fluid is displaced.

A mathematical model is presented in this paper described in state space as a basic acoustic sensor description. The objective is
to adjust the parameters allowing the acoustic device to describe a signal in its trajectory, representing in geophysical manner the
cavity form. Therefore, the control is performed on the response of the acoustic sensor model, adjusted with a parameter
estimation process. The simulation results counts convergence between the reference and identified signals.

Keywords (cid:150) Sliding modes, estimation, acoustic transducer, control.
I. INTRODUCTION
Acoustic processes are of particular interest in geophysical processes because they describe different phenomena in nature,
such as the Doppler effects with its relative velocity between two bodies [4]; stochastic entropy [10] with its pernicious effects
when the signal exceeds the allowable thresholds [7]. The resonance commonly applies different methodologies describing the
motion of an embryo in a womb or an egg, or the liquid levels in an oil well. Physically, the identification process is based on
electronic sensors measuring and building the forms, allowing industrial automatic inspection where man has no access, or the
eye cannot see, as the water depth sensing is performed by a sonar [6], [8], [9].

To meet the demands of science, multiple types of sensors such as photoelectric, magnetic, inductive, nuclear, acoustic,
among other have been created. The acoustic sensor has a great potential still to be exploited in applications such as: content
description in deep wells, watersheds, lakes, caverns, applied in places where solutions have a very rough shape container [6].
In general, the challenges of all the acoustic sensors are:
a) The signal type must be consistent with the environment, where the moves measure permits describing the contents,
b) All types of sensors are designed for a range defined without self-adjusting.
An acoustic signal can be reflected in areas where other signal types cannot operate, either by the fluid drawbacks
displacement or because they would require hard-reflected signals. For example, a distance of at least 500m in aqueous
media, operating with power lines at different speeds. Large distances require high capacity signals: In this paper a control
model acoustic sensor response with a parameter estimator as an adaptive technique, solves the problem considered in the
first application point; such as, the reflection intensity offering a high capacity self-regulation descriptor distance sensing the
distance range without, previously described. In some sense only a range is identified, hence the device needs self-regulation
range distance with a limited level of uncertainty, adjusting the piezoelectric reference parameters system, which is firstly
compared with a piezoelectric model tracking. In the event, where the results need the self-adjusting and controlling law, the
parameters set is used within the model like an adjustable parameters description for long distances and rough cavities
considering the initial conditions reference system [6], [8] and, [9].

To overcome current limiting, the control action regulating the acoustic signal level emitted by a piezoelectric device
ensures that the reflected echo signal requires that the acoustic sensor intensity reading, analyzes the reflected signal [1]. The
model and the control system were the basis for a particular container description, without presupposing the distance, adjusting
the parameters through the measurement system and control action.
Therefore, the piezoelectric model actuator according to [1] is simplified into the form:

*Research supported by National Polytechnic Institute, ESIME Zacatenco.
Julio CØsar Tovar Rodr(cid:237)guez is with the National Polytechnic Institute, Mechanical and Electrical School, IPN av., MØxico City 07738, MØxico.
(corresponding author to provide phone: [phone] ext. 54660; e-mail: jctovar77@ hotmail.com).
Floriberto Ort(cid:237)z Rodr(cid:237)guez, is with the National Polytechnic Institute, Mechanical and Electrical School, , IPN av., MØxico City 07738, MØxico (e-mail:
[email protected]).
Carlos RomÆn Mariaca Gaspar is with the National Polytechnic Institute, Mechanical and Electrical School, IPN av., MØxico City 07738, MØxico (e-mail:
[email protected]).
1
Rob(cid:243)tica y Mecatr(cid:243)nica de Servicios: Teor(cid:237)a y Aplicaciones. Cap(cid:237)tulo 1, pÆginas 1-8.
ISBN 978-607-9394-06-6, Asociaci(cid:243)n Mexicana de Mecatr(cid:243)nica, A. C.
•
X = AX +Bw
t t t (1)
Y =CX +Dw
t t t
Where the matrix order correspond to the differential equation order with !"[&(cid:215)&], ’!"[&(cid:215)(], )!"[&(cid:215)&], *!
[#,$%[ [#,([ [(,$([
"[&(cid:215)(],+ !./0 ,34 <(cid:127)5
[#,([ - 12 12
Theorem 1. Let the model considered in in state space have the form (1). In agreement to (1) and [3], the recursive form
is:
•
Yt =GYt +HVt (2)
Where: , ! are matrices bounded with "#[’(cid:215)’]and, !=()*,+,-,./, 0 "234 ,78 <(cid:127)9 Proof (See Appendix).
[$,%&[ 1 56 56

II. CONTROL LAW

Theorem 2. The control law system " with respect to (2) has the form:
!

V* =H+(E -GY ) (3)
t t t

With #$the pseudo-inverse matrix , the innovation process ! considered in (2), and #, as a unknown matrix having
"
the form$# %&[*(cid:215)*]. Proof (See Appendix).
[’,()[

The piezoelectric device has a main problem which corresponds to parameter distance description [1], [4], described as
#+ . This is solved estimating the gain with a lower uncertainty in almost all points that make up its surface.
"
III. PARAMETERS ESTIMATION
Applying the control law (3) into the model described in (2), converges to reference system ( ) only if the matrix "[1],
!
[3] is known. Unfortunately, the reference system viewed as a Black Box scheme [3]; the matrix gain " is unknown,
because it corresponds to the internal system description. Consequently, the estimation process is required describing the
internal matrix gain through the time process [2], [3], [5].

Theorem 3. Let the recursive model (2), with answer have the stochastic matrix estimation based on [2], [3], [13]-
[17]:
Gt =PtQt (4)

With #,$% covariance and variance matrices, with respect to (2). Proof (See Appendix).
! !

!
Theorem 4. Let " be a stationary process, the recursive form in discrete manner is (5) and is viewed in figure 1, based
on [2], [3].

Gt =atGt-1+bt (5)

Figure 1 Recursive form of #$ " considering stationary conditions.
2
Rob(cid:243)tica y Mecatr(cid:243)nica de Servicios: Teor(cid:237)a y Aplicaciones. Cap(cid:237)tulo 1, pÆginas 1-8.
ISBN 978-607-9394-06-6, Asociaci(cid:243)n Mexicana de Mecatr(cid:243)nica, A. C.

A piezoelectric device developed as a mathematical model viewed as (1), considering the Black Box properties, only is
observes input and output signals without knowing exactly the internally system operations. The control law does not affect
the reference system, but the model through the parameters is estimated in the probability sense affecting the model
converging to the reference piezoelectric device answer. All results are described in real numbers ( ), specifically over the
hypothetical line that describes the container form.

The piezoelectric block diagram using the control action with adaptation into the model system is shown in figure 2.

Figure 2 Piezoelectric system viewed as a control block diagram with parameters adjusted dynamically.

Now, the control law system V!" described in (3) considering (5) has the form:

(cid:230)(cid:230) (cid:246)(cid:246)
Vt* =H+(cid:231)ŁEt -GtYtł(cid:247) (6)
With #$the pseudo-inverse matrix%#, the innovation process &’ considered in (2), and ()’, as a matrix estimation.
IV. SIMULATION
The piezoelectric signal ( ) with stochastic properties tracking the model output answer ( " ) with parameters estimation
! !
affecting the control law action. Both considered in (2) is illustrated in figure 3.

y vs "
t t
0.23
y
0.22 t
0.21
0.2
0.19
0.18
0.17 "
t
0.16
0 10 20 30 40 50 60 70 80 90 100 t

Figure 3 Piezoelectric temporal signal and its tracking.

3
Rob(cid:243)tica y Mecatr(cid:243)nica de Servicios: Teor(cid:237)a y Aplicaciones. Cap(cid:237)tulo 1, pÆginas 1-8.
ISBN 978-607-9394-06-6, Asociaci(cid:243)n Mexicana de Mecatr(cid:243)nica, A. C.
The adaptation scheme viewed in figure 1, was applied in the control law and into the model, observing that the answer is
very narrow with respect to the real piezoelectric results. In figure 4, is observed the innovation process.

Innovation Process
0.025
0.02
0.015
0.01
0.005
0
-0.005
-0.01
-0.015
-0.02
-0.025
0 10 20 30 40 50 60 70 80 90 100 t

FigFigure 4 Innovation process through the time t

The convergence rate generated between the reference signal and the tracking output model is measured in decibels based
on stochastic entropy and has the form ="20[# $%(# )"(1/20) ], with # *+ "+, and the result is viewed in
! ! ! !&’ ! ! !
figure 4.

H
t
x 105
0.5
0
-0.5
-1
-1.5
-2
-2.5
-3
-3.5
-4
-4.5
0 10 20 30 40 50 60 70 80 90 100 t

Figure 5. Stochastic entropy with respect to difference
between the tracking and reference signal.

4
Rob(cid:243)tica y Mecatr(cid:243)nica de Servicios: Teor(cid:237)a y Aplicaciones. Cap(cid:237)tulo 1, pÆginas 1-8.
ISBN 978-607-9394-06-6, Asociaci(cid:243)n Mexicana de Mecatr(cid:243)nica, A. C.
V. CONCLUSION
This paper presented a model considering into the references as a piezoelectric device description. The control system over
the model required matrix gain parameters estimation, having an adaptive movement using into the tracking operations. The
theoretical results were developed by the dynamical model properties and the control action. The simulation tracking was
acceptable in probability sense, with convergence rate measured in decibels in stochastic description manner. The output
piezoelectric device had an evolution with innovation signal. The tracking permitted a great convergence with stationary
conditions affecting the control action over the model in positive form, minimizing the convergence error near to
piezoelectric answer with rough conditions. The control law depended on the internal parameters, with adjustable gains
estimation. The simulation results described the proposal answer, with a convergence level bounded in its movements as
shown in the entropy.
APPENDIX

Proof (Theorem 1).
Let the model (1), with the first derivate described in:
Yt =CXt+Dwt (7)

Substituting in (7) to X! de (1), has:

Yt =CAXt +CBw+Dwt (8)

The internal state X! in agreement to (1) is described in (9) with respect to observable signal:

Xt =C+Yt -C+Dwt (9)

(9) in (8), has:

Y =CAC+Y -CAC+Dw +CBw +Dw (10)
t t t t t

In symbolic form (10) is described in (11) with G CAC!, H [("CAC!+CB) D], and V [w w$ ]%.
# # #

Y =GY +HV (11)
t t t

Corresponding to (2).(cid:127)

Proof (Theorem 2). Let the system (2) accomplish with:

M MT <0 (12)
t t

Where the trajectory region with respect to the gain matrixM, is described in:

M =-F (13)
t t

With F, a continuous function bounded by intervals with uniform measure t, accomplishing with the innovation process:

(cid:217)
Y =Y +F (14)
t t t

5