## List of publications

### 2017

• M. Davoodi, A. J. Gmerek, N. Meskin, T. E. S., and R. E. Kearney, "A New Ankle Foot Setup: Modeling, Kinematic Based Optimization and Control," in < ">CoDIT 2017, International Conference on Control, Decision and Information Technologies< >, 2017.
[BibTeX] [Abstract]

This paper presents the design, modeling, optimization and control of a newly developed ankle foot setup designated for testing different gait-assistive compliance control algorithms. This laboratory setup consists of an active ankle foot orthosis (AFO) and an artificial lower leg (ALL). The ALL is designed to emulate the human's ankle joint and muscles and also to simulate different disabilities for an impaired subject. Moreover, the AFO is designed to ameliorate and modify the impaired human walking. It is characterized by a closed-loop chain structure and powered by a linear ServoTube motor making it very fast and responsive. The paper first presents the derivation of the kinematic model for the AFO. Based on the obtained kinematic model, the AFO's range of motion and its size are optimized which allow to choose and mount the actuator in such a way that almost ideally meet predefined requirements. Then, the dynamic model of the AFO complex is derived. Finally, an impedance controller is developed based on the obtained dynamic model. Real-time experimental results of the proposed controller show the performance of the whole setup in simulating abnormal and impaired gaits as well as in supporting the movement of the ankle joint to perform proper gait movements.

@InProceedings{anklefootsetup,
author = {Davoodi, M. and Gmerek, A. J. and Meskin, N. and Tehrani E. S. and Kearney, R.E.},
title = {A {N}ew {A}nkle {F}oot {S}etup: {M}odeling, {K}inematic {B}ased {O}ptimization and {C}ontrol},
booktitle = {Co{DIT} 2017, {I}nternational {C}onference on {C}ontrol, {D}ecision and {I}nformation {T}echnologies},
year = {2017},
abstract = {This paper presents the design, modeling, optimization and control of a newly developed ankle foot setup designated for testing different gait-assistive compliance control algorithms. This laboratory setup consists of an active ankle foot orthosis (AFO) and an artificial lower leg (ALL). The ALL is designed to emulate the human's ankle joint and muscles and also to simulate different disabilities for an impaired subject. Moreover, the AFO is designed to ameliorate and modify the impaired human walking. It is characterized by a closed-loop chain structure and powered by a linear ServoTube motor making it very fast and responsive. The paper first presents the derivation of the kinematic model for the AFO. Based on the obtained kinematic model, the AFO's range of motion and its size are optimized which allow to choose and mount the actuator in such a way that almost ideally meet predefined requirements. Then, the dynamic model of the AFO complex is derived. Finally, an impedance controller is developed based on the obtained dynamic model. Real-time experimental results of the proposed controller show the performance of the whole setup in simulating abnormal and impaired gaits as well as in supporting the movement of the ankle joint to perform proper gait movements.},
}

• A. J. Gmerek, D. M., N. Meskin, and F. Jaber, "An EMG Signal Processing System for Control of an Ankle-foot Orthosis," in < ">CoDIT 2017, International Conference on Control, Decision and Information Technologies< >, 2017.
[BibTeX] [Abstract]

This paper describes an EMG-based digital processing system which estimates force and direction of motion that can be used in control frameworks of ankle-foot orthoses (AFOs). To control an AFO in a reliable way it is usually necessary to estimate ankle's muscles voluntary contraction (VC) and user's intention of motion. Thus, a system was created that can designate these parameters from EMG signals (also known as myopotentials or myosignals). The voluntary contraction was calculated based on the root mean square (RMS) of EMG signals, and a direction of motion was estimated by using a feature-based classifier. The experiments were performed on five healthy, sitting subjects and research includes a selection of discriminative features and comparison of basic classifiers. The results of the study showed that the direction of motion could be estimated in the real time with high accuracy.

@InProceedings{emg_ankle_orthosis,
author = {Gmerek, A. J. and Davoodi M. and Meskin, N. and Jaber, F},
title = {An {EMG} {S}ignal {P}rocessing {S}ystem for {C}ontrol of an {A}nkle-foot {O}rthosis},
booktitle = {Co{DIT} 2017, {I}nternational {C}onference on {C}ontrol, {D}ecision and {I}nformation {T}echnologies},
year = {2017},
abstract = {This paper describes an EMG-based digital processing system which estimates force and direction of motion that can be used in control frameworks of ankle-foot orthoses (AFOs). To control an AFO in a reliable way it is usually necessary to estimate ankle's muscles voluntary contraction (VC) and user's intention of motion. Thus, a system was created that can designate these parameters from EMG signals (also known as myopotentials or myosignals). The voluntary contraction was calculated based on the root mean square (RMS) of EMG signals, and a direction of motion was estimated by using a feature-based classifier. The experiments were performed on five healthy, sitting subjects and research includes a selection of discriminative features and comparison of basic classifiers. The results of the study showed that the direction of motion could be estimated in the real time with high accuracy.},
}

### 2016

• A. Gmerek, N. Meskin, and F. Jaber, "Development of LPV models and switching LPV-H-inf controller for a hydraulic system," in < ">Proc. Decision and Information Technologies (CoDIT) 2016 Int. Conf. Control< >, 2016, pp. 484-489. doi:10.1109/CoDIT.2016.7593610
[BibTeX]
@InProceedings{lpv_switching_controller,
author = {A. Gmerek and N. Meskin and F. Jaber},
title = {Development of {LPV} models and switching {LPV-H}-inf controller for a hydraulic system},
booktitle = {Proc. {D}ecision and {I}nformation {T}echnologies ({CoDIT}) 2016 {I}nt. {C}onf. {C}ontrol},
year = {2016},
month = apr,
pages = {484--489},
doi = {10.1109/CoDIT.2016.7593610},
__markedentry = {[Artur:6]},
keywords = {H\textsuperscript{$\infty$} control, electrohydraulic control equipment, hydraulic actuators, hydraulic systems, linear parameter varying systems, servomechanisms, switching systems (control), valves, LPV model, black-box modelling, electrohydraulic system, hydraulic actuator, linear parameter varying model, servovalve, switching LPV-H\textsubscript{$\infty$} controller, white box modelling, Adaptation models, Analytical models, Control systems, Pistons, Robustness, Valves},
}

• A. Gmerek, N. Meskin, E. S. Tehrani, and R. Kearney, "Design of a hydraulic ankle-foot orthosis," in < ">Proc. 6th IEEE Int. Conf. Biomedical Robotics and Biomechatronics (BioRob)< >, 2016, pp. 1041-1048. doi:10.1109/BIOROB.2016.7523768
[BibTeX]
@InProceedings{hydraulic_ankle_foot2016,
author = {A. Gmerek and N. Meskin and E. S. Tehrani and R. Kearney},
title = {Design of a hydraulic ankle-foot orthosis},
booktitle = {Proc. 6th {IEEE} {I}nt. {C}onf. {B}iomedical {R}obotics and {B}iomechatronics ({B}io{R}ob)},
year = {2016},
month = jun,
pages = {1041--1048},
doi = {10.1109/BIOROB.2016.7523768},
__markedentry = {[Artur:6]},
keywords = {hydraulic actuators, minimisation, motion control, orthotics, accumulator, compactness maximization, force minimization, human walking, hydraulic actuator, hydraulic ankle-foot orthosis, hydraulic power supply, intrinsic ankle joint stiffness, kinematic parameter, reflex ankle joint stiffness, servo-valve, Force, Hydraulic actuators, Kinematics, Legged locomotion, Pistons, Torque},
}

• A. Gmerek, N. Meskin, E. S. Tehrani, and R. E. Kearney, "The characterization of the kinematic and dynamic properties of the ankle joint for an artificial ankle joint design," in < ">Proc. 6th IEEE Int. Conf. Biomedical Robotics and Biomechatronics (BioRob)< >, 2016, pp. 710-716. doi:10.1109/BIOROB.2016.7523710
[BibTeX]
@InProceedings{anklecharacterization,
author = {A. Gmerek and N. Meskin and E. S. Tehrani and R. E. Kearney},
title = {The characterization of the kinematic and dynamic properties of the ankle joint for an artificial ankle joint design},
booktitle = {Proc. 6th {IEEE} {I}nt. {C}onf. {B}iomedical {R}obotics and {B}iomechatronics ({B}io{R}ob)},
year = {2016},
month = jun,
pages = {710--716},
doi = {10.1109/BIOROB.2016.7523710},
__markedentry = {[Artur:6]},
keywords = {acceleration, artificial limbs, design engineering, legged locomotion, patient rehabilitation, probability, robot dynamics, robot kinematics, acceleration, ankle joint dynamic properties, ankle joint kinematic properties, ankle-foot orthoses, artificial ankle joint design, below-knee prostheses, exoskeletons, motion, probability density function, rehabilitation robots, torque-velocity curve, velocity, Dynamics, Foot, Joints, Kinematics, Legged locomotion, Muscles, Torque},
}

### 2015

• A. Gmerek, "Ogólna charakterystyka prototypu robota ARR przeznaczonego do fizjoterapii stawu barkowo-obojczykowego i \lokciowego," < ">Niepełnosprawność - zagadnienia, problemy, rozwiązania< >, vol. 16, iss. 3, pp. 7-22, 2015.

The paper describes the design and operation of a semi-exoskeleton arm rehabilitation robot dedicated for upper limb rehabilitation. The robot was created at Lodz University of Technology. It analyzes electromyography signals generated by patient in order to use them in control and to assess physiotherapy process. The control system of the robot is also able to work in advanced control framework related to virtual reality and compliance control, thereby patient can interact with the robot, what has positive influence on the course of therapy. In the beginning of the article there is the description of current state of the art in Poland and in the world. Next there is the description of various robotic subsystems. In the end of the article there are conclusions, which are in facts guidelines for engineers, who are working under similar projects.

@Article{gmerekpfron,
Title = {Ogólna charakterystyka prototypu robota {ARR}
przeznaczonego do fizjoterapii stawu
barkowo-obojczykowego i {\l}okciowego},
Author = {Gmerek, A.},
Journal = {Niepełnosprawność - zagadnienia, problemy, rozwiązania},
Year = {2015},
Number = {3},
Pages = {7-22},
Volume = {16},
Abstract = {The paper describes the design and operation of a semi-exoskeleton arm rehabilitation robot dedicated for upper limb rehabilitation. The robot was created at Lodz University of Technology. It analyzes electromyography signals generated by patient in order to use them in control and to assess physiotherapy process. The control system of the robot is also able to work in advanced control framework related to virtual reality and compliance control, thereby patient can interact with the robot, what has positive influence on the course of therapy. In the beginning of the article there is the description of current state of the art in Poland and in the world. Next there is the description of various robotic subsystems. In the end of the article there are conclusions, which are in facts guidelines for engineers, who are working under similar projects.},
Owner = {Artur},
Timestamp = {2016.01.01},
Url = {http://www.pfron.org.pl/kn/poprzednie-numery?page=1}
}

### 2014

• A. Gmerek, D. Pakulski, M. Balcerzak, T. Soghbatyan, and P. Dawid, "AutoLUT - autonomiczny robot mobilny oparty na systemie ROS," in < ">Prace Naukowe Politechniki Warszawskiej. Elektronika< >, Politechnika Warszawska, 2014, pp. 155-164.
[BibTeX] [Abstract]

This paper describes the autonomous mobile robot which was created for Robotour competition. The aim of this competition was to reach desired GPS coordinates. The only permissible path motion was on the road (riding on grass was forbidden). The electric quad was used as a construction skeleton of the robot. The low-level controller was based on the Arduino technology and the high-level controller was programmed in the ROS platform. This paper contains the description of the developed technology as well as clues for future autonomous mobile robots designers.

@InCollection{autolutrobotour,
Title = {Auto{LUT} - autonomiczny robot mobilny oparty na systemie {ROS}},
Author = {Gmerek, A. and Pakulski, D. and Balcerzak, M. and Soghbatyan, T. and Dawid, P.},
Booktitle = {Prace {N}aukowe {P}olitechniki {W}arszawskiej. {E}lektronika},
Publisher = {Politechnika Warszawska},
Year = {2014},
Pages = {155-164},
Abstract = {This paper describes the autonomous mobile robot which was created for Robotour competition. The aim of this competition was to reach desired GPS coordinates. The only permissible path motion was on the road (riding on grass was forbidden). The electric quad was used as a construction skeleton of the robot. The low-level controller was based on the Arduino technology and the high-level controller was programmed in the ROS platform. This paper contains the description of the developed technology as well as clues for future autonomous mobile robots designers.},
Owner = {Artur},
Timestamp = {2014.08.23}
}

### 2013

• A. Gmerek, "Mechanical and hardware architecture of the semi-exoskeleton Arm Rehabilitation Robot," < ">The archive of mechanical engineering< >, vol. 60, iss. 4, pp. 557-574, 2013.

This paper deals with mechanical and hardware design of a robot, used for the rehabilitation of upper extremities. It has been called ARR-1 (Arm Rehabilitation Robot). The robot has a semi-exoskeleton structure. This means that some parts of the robot fit closely to the human arm (an orthosis), but the weight of the construction does not load patients body. The device is used for the whole arm rehabilitation, but active joints are only situated in glenohumeral and elbow joints. The robot is electrically actuated.

@Article{amegmerek,
Title = {Mechanical and Hardware Architecture of the Semi-Exoskeleton {A}rm {R}ehabilitation {R}obot},
Author = {Gmerek, A.},
Journal = {The Archive of Mechanical Engineering},
Year = {2013},
Number = {4},
Pages = {557-574},
Volume = {60},
Abstract = {This paper deals with mechanical and hardware design of a robot, used for the rehabilitation of upper extremities. It has been called ARR-1 (Arm Rehabilitation Robot). The robot has a semi-exoskeleton structure. This means that some parts of the robot fit closely to the human arm (an orthosis), but the weight of the construction does not load patients body. The device is used for the whole arm rehabilitation, but active joints are only situated in glenohumeral and elbow joints. The robot is electrically actuated.},
Owner = {Artur},
Timestamp = {2013.10.30},
Url = {http://ame.czasopisma.pan.pl/}
}

• A. Gmerek, "Design of the robotic exoskeleton for upper-extremity rehabilitation," < ">Pomiary, Automatyka, Robotyka< >, vol. 17, pp. 97-101, 2013.
[BibTeX] [Abstract]

This paper touches upon the issue of designing of an upper-limb exoskeleton used for rehabilitation. Mainly there are presented the results concerning the mechanical design. The exoskeleton, also known as orthosis, has 12 degrees of freedom (DOF). Among 12 DOF there can be distinguished 7 DOF which are actively controlled during rehabilitation, 3 passive DOF which are responsible for wrist movements, and 2 adjustable DOF used for links elongation. An orthosis has been specifically designed to accomplish requirements of medical applications. Described exoskeleton is a key-component of the upper-limb rehabilitation robot.

@Article{gmerek2013design,
Title = {Design of the robotic exoskeleton for upper-extremity rehabilitation},
Author = {Gmerek, A.},
Journal = {{P}omiary, {A}utomatyka, {R}obotyka},
Year = {2013},
Pages = {97--101},
Volume = {17},
Abstract = {This paper touches upon the issue of designing of an upper-limb exoskeleton used for rehabilitation. Mainly there are presented the results concerning the mechanical design. The exoskeleton, also known as orthosis, has 12 degrees of freedom (DOF). Among 12 DOF there can be distinguished 7 DOF which are actively controlled during rehabilitation, 3 passive DOF which are responsible for wrist movements, and 2 adjustable DOF used for links elongation. An orthosis has been specifically designed to accomplish requirements of medical applications. Described exoskeleton is a key-component of the upper-limb rehabilitation robot.}
}

• A. Gmerek and A. Kawka, "STEFAN - samostabilizujący się egzoszkielet kończyn dolnych," in < ">Publikacja projektu Najlepsi Doktoranci Politechniki ??ódzkiej< >, Politechnika {\'L}ódzka, 2013.
[BibTeX] [Abstract]

The article describe potential problems of a mechanical design and control system design of self-stabilizing exoskeleton of lower extremity. In the beginning, there is a short review of literature of passive and active exoskeletons of lower extremity. Next, the article centers on the novel proposal of a design, which is able to maintain balance, despite its structural instability. This theoretical robot was called STEFAN (Self-sTabilizing Exoskeleton For Assistive Neurotherapy). Such an exoskeleton could assist in locomotion or be used for medical rehabilitation.

• A. Gmerek, T. Soghbatyan, M. Sudowski, A. Grzelka, and M. Jarz{k{e}}bski, "Projekt i budowa robota planarnego o dwóch stopniach swobody przeznaczonego do zastosowań dydaktycznych," in < ">Publikacja projektu Najlepsi Doktoranci Politechniki \'Lódzkiej< >, Politechnika {\'L}ódzka, 2013.
[BibTeX] [Abstract]

This article describes the project of creating the planar 2-DoF robot. The robot is designated for education purposes. Authors describe the design of a manipulator, as well as outcome from experimental research.

@InCollection{gmerek2dof,
Title = {Projekt i budowa robota planarnego o dwóch stopniach swobody przeznaczonego do zastosowań dydaktycznych},
Author = {Gmerek, A. and Soghbatyan, T. and Sudowski, M. and Grzelka, A. and Jarz{\k{e}}bski, M.},
Booktitle = {Publikacja projektu {N}ajlepsi {D}oktoranci {P}olitechniki {\'L}{\'o}dzkiej},
Publisher = {Politechnika {\'L}{\'o}dzka},
Year = {2013},
Abstract = {This article describes the project of creating the planar 2-DoF robot. The robot is designated for education purposes. Authors describe the design of a manipulator, as well as outcome from experimental research.},
Owner = {Artur},
Timestamp = {2016.01.02}
}

• E. Jezierski and A. Gmerek, "Impedance controllers for electric-driven robots," < ">Journal of Automation, Mobile Robotics & Intelligent Systems (JAMRIS)< >, vol. 7, pp. 13-20, 2013.

This article presents a proposal of impedance controller, which is able to influence the compliance of a robot, based on information obtained from a low level controller about robot interaction with the environment. The proposed system consists of a low level controller with proximity sensor, based on which mechanical impedance is adjusted. The intention of the creators was to develop a universal impedance regulator in the sense of hardware and software layers. Because of TCP/IP architecture, the designed regulator can be easily adapted to different robot controllers. Experiments were conducted on a real 1-DOF manipulator driven by BLDC motor, as well as using simulation on a 2DOF planar robot.

@Article{gmerekjamris,
Title = {Impedance controllers for electric-driven robots},
Author = {Jezierski, E. and Gmerek, A.},
Journal = {Journal of {A}utomation, {M}obile {R}obotics \& {I}ntelligent {S}ystems ({JAMRIS})},
Year = {2013},
Pages = {13--20},
Volume = {7},
Abstract = {This article presents a proposal of impedance controller, which is able to influence the compliance of a robot, based on information obtained from a low level controller about robot interaction with the environment. The proposed system consists of a low level controller with proximity sensor, based on which mechanical impedance is adjusted. The intention of the creators was to develop a universal impedance regulator in the sense of hardware and software layers. Because of TCP/IP architecture, the designed regulator can be easily adapted to different robot controllers. Experiments were conducted on a real 1-DOF manipulator driven by BLDC motor, as well as using simulation on a 2DOF planar robot.},
Url = {http://www.jamris.org/archive/37-issue-04-2013}
}

### 2012

• A. Gmerek, "Przegląd rozwiązań z zakresu robotyki rehabilitacyjnej i protetyki - wykorzystanie sygnałów bioelektrycznych w sterowaniu robotów," in < ">Transfer Innowacyjnych Technologii w Obszarach Automatyki i Robotyki< >, M. l, Ed., Oficyna Wydawnicza {PIAP}, 2012.
[BibTeX] [Abstract]

This article presents an overview of the latest rehabilitation robots and intelligent prostheses. In particular robots are discussed, that will be implemented soon. Very important in control of rehabilitation robots and advanced prostheses is the technology of increasing human interaction with the robot. This can be achieved with the use of bioelectric signals, in particular the electromyographic signals (EMG). Using it can be accurately estimated the activity of individual motor units and strength and fatigue of the muscle.

@InCollection{gmerek2012przeglkad,
Title = {Przegląd rozwiązań z zakresu robotyki rehabilitacyjnej i protetyki - wykorzystanie sygnałów bioelektrycznych w sterowaniu robotów},
Author = {Gmerek, A.},
Booktitle = {Transfer {I}nnowacyjnych {T}echnologii w {O}bszarach {A}utomatyki i {R}obotyki},
Publisher = {Oficyna Wydawnicza {PIAP}},
Year = {2012},
Editor = {Ma{\l}gorzata Kaliczy{\'n}ska},
Abstract = {This article presents an overview of the latest rehabilitation robots and intelligent prostheses. In particular robots are discussed, that will be implemented soon. Very important in control of rehabilitation robots and advanced prostheses is the technology of increasing human interaction with the robot. This can be achieved with the use of bioelectric signals, in particular the electromyographic signals (EMG). Using it can be accurately estimated the activity of individual motor units and strength and fatigue of the muscle.}
}

• A. Gmerek, "ARR - robot do rehabilitacji kończyny górnej," in < ">Postępy robotyki 2012< >, K. Tchoń and C. Zielińskiego, Eds., Politechnika Warszawska, 2012.
[BibTeX] [Abstract]

ARR (Arm Rehabilitation Robot) is a robot designed for upper limb rehabilitation. The active part of the robot is an exoskeleton, which is fastened to an arm and a forearm of the patient�s limb. The electromyographic system cooperate with the robot. Based on EMG signals, the force is estimated, which is then used in control algorithms. The paper presents the general characteristics of the robot. The article briefly describes the design issues, the functioning of low and high-level controllers of the robot and the associated support systems, which facilitate operate and simplify therapy (e.g. electromyographic system, virtual reality system dedicated for rehabilitation, audible feedback, voice control subsystem and others).

@InCollection{gmerekkkr2012,
Title = {{ARR} - Robot do Rehabilitacji kończyny górnej},
Author = {Gmerek, A.},
Booktitle = {Postępy robotyki 2012},
Publisher = {Politechnika Warszawska},
Year = {2012},
Editor = {Tchoń, K. and Zielińskiego, C.},
Abstract = {ARR (Arm Rehabilitation Robot) is a robot designed for upper limb rehabilitation. The active part of the robot is an exoskeleton, which is fastened to an arm and a forearm of the patient�s limb. The electromyographic system cooperate with the robot. Based on EMG signals, the force is estimated, which is then used in control algorithms. The paper presents the general characteristics of the robot. The article briefly describes the design issues, the functioning of low and high-level controllers of the robot and the associated support systems, which facilitate operate and simplify therapy (e.g. electromyographic system, virtual reality system dedicated for rehabilitation, audible feedback, voice control subsystem and others).}
}

• A. Gmerek, "Wykorzystanie quasi-chaotycznych oscylatorów do generowania rytmu chodu robotów kroczących," in < ">Postępy robotyki 2012< >, K. Tchoń and C. Zielińskiego, Eds., Politechnika Warszawska, 2012.
[BibTeX] [Abstract]

The aim of this study was to develop a control system of autonomous walking robots, that is able to efficiently control the gait rhythm of walking robots, in terms of various properties of the substrate and environment conditions. Experiments were conducted on models of insect walking robots. Gait were generated using the coupled oscillators (van der Pol and Recursive Artificial Neural Networks). It was also carried out simulation studies on the impedance control of pedipulators. With the use of developed system robots were able to work effectively, e.g. to get out of low.

@InCollection{gmerekkroczace,
Title = {Wykorzystanie quasi-chaotycznych oscylatorów do generowania rytmu chodu robotów kroczących},
Author = {Gmerek, A.},
Booktitle = {Postępy robotyki 2012},
Publisher = {Politechnika Warszawska},
Year = {2012},
Editor = {Tchoń, K. and Zielińskiego, C.},
Abstract = {The aim of this study was to develop a control system of autonomous walking robots, that is able to efficiently control the gait rhythm of walking robots, in terms of various properties of the substrate and environment conditions. Experiments were conducted on models of insect walking robots. Gait were generated using the coupled oscillators (van der Pol and Recursive Artificial Neural Networks). It was also carried out simulation studies on the impedance control of pedipulators. With the use of developed system robots were able to work effectively, e.g. to get out of low.}
}

• A. Gmerek, "The virtual reality system used for upper extremity rehabilitation," in < ">17th International Conference on Methods and Models in Automation and Robotics (MMAR)< >, 2012, pp. 312-314.
[BibTeX] [Abstract]

This paper presents a complete Virtual Reality (VR) system used for upper extremity rehabilitation. The system consists of sensors, the computer application and stereoscopic display glasses. The system is designed to cooperate with special mechanical orthosis used for arm rehabilitation. Patients can see their movements in the 3D virtual space during exercises. The system has a key role in development of rehabilitation robot, because numerous researches show that exercises in VR decreases the time of rehabilitation.

@InProceedings{gmerekvr,
Title = {The virtual reality system used for upper extremity rehabilitation},
Author = {Gmerek, A.},
Booktitle = {17th {I}nternational {C}onference on {M}ethods and {M}odels in {A}utomation and {R}obotics ({MMAR})},
Year = {2012},
Pages = {312-314},
Abstract = {This paper presents a complete Virtual Reality (VR) system used for upper extremity rehabilitation. The system consists of sensors, the computer application and stereoscopic display glasses. The system is designed to cooperate with special mechanical orthosis used for arm rehabilitation. Patients can see their movements in the 3D virtual space during exercises. The system has a key role in development of rehabilitation robot, because numerous researches show that exercises in VR decreases the time of rehabilitation.},
Keywords = {medical robotics;orthotics;patient rehabilitation;virtual reality;3D virtual space;arm rehabilitation;computer application;mechanical orthosis;rehabilitation robot;sensor;stereoscopic display glass;upper extremity rehabilitation;virtual reality system;Computers;Extremities;Games;Joints;Robots;Virtual reality}
}

• A. Gmerek and E. Jezierski, "Admittance control of a 1-DoF robotic arm actuated by BLDC motor," in < ">17th International Conference on Methods and Models in Automation and Robotics (MMAR)< >, 2012, pp. 633-638.
[BibTeX] [Abstract]

The aim of this paper is to analyze the admittance control of a single-axis mechanism actuated by brushless DC (BLDC) motor. There is also presented an examination of applied control architecture. Experiments consisted of the study of high-speed contact of the manipulator with infinitely rigid environment, as well as slow increase in force, which affected the manipulator. There are presented simulation results, as well as the practical study. The impedance controller drives the torque controlled manipulator system. Force between the arm and the environment was estimated based on motor current. The paper shows from practical point of view how to efficiently steer BLDC motors with admittance controllers.

@InProceedings{gmerek2012admittance,
Title = {Admittance control of a 1-{D}o{F} robotic arm actuated by {BLDC} motor},
Author = {Gmerek, A. and Jezierski, E.},
Booktitle = {17th {I}nternational {C}onference on {M}ethods and {M}odels in {A}utomation and {R}obotics ({MMAR})},
Year = {2012},
Organization = {IEEE},
Pages = {633--638},
Abstract = {The aim of this paper is to analyze the admittance control of a single-axis mechanism actuated by brushless DC (BLDC) motor. There is also presented an examination of applied control architecture. Experiments consisted of the study of high-speed contact of the manipulator with infinitely rigid environment, as well as slow increase in force, which affected the manipulator. There are presented simulation results, as well as the practical study. The impedance controller drives the torque controlled manipulator system. Force between the arm and the environment was estimated based on motor current. The paper shows from practical point of view how to efficiently steer BLDC motors with admittance controllers.}
}

• E. Jezierski and A. Gmerek, "Regulator impedancyjny dla elektrycznego zespołu napędowego robota," in < ">Post\kepy robotyki 2012< >, K. Tchoń and C. Zielińskiego, Eds., Politechnika Warszawska, 2012.
[BibTeX] [Abstract]

This article presents a proposal of impedance controller, which is able to influence the compliance of a robot based on information obtained from a low-level controller, about robot interaction with the environment. The proposed system consists of a low-level controller with proximity sensor, based on which mechanical impedance is adjusted. The intention of the creators was to develop a universal impedance regulator in the sense of hardware and software layers. Because of TCP/IP architecture, the designed regulator can be easily adapted to different robot controllers. Experiments were conducted on a real 1-DOF manipulator driven by BLDC motor.

@InCollection{gmerekregkkr2012,
Title = {Regulator impedancyjny dla elektrycznego zespołu napędowego robota},
Author = {Jezierski, E. and Gmerek, A.},
Booktitle = {Post{\k{e}}py robotyki 2012},
Publisher = {Politechnika Warszawska},
Year = {2012},
Editor = {Tcho{\'n}, K. and Zieli{\'n}skiego, C.},
Abstract = {This article presents a proposal of impedance controller, which is able to influence the compliance of a robot based on information obtained from a low-level controller, about robot interaction with the environment. The proposed system consists of a low-level controller with proximity sensor, based on which mechanical impedance is adjusted. The intention of the creators was to develop a universal impedance regulator in the sense of hardware and software layers. Because of TCP/IP architecture, the designed regulator can be easily adapted to different robot controllers. Experiments were conducted on a real 1-DOF manipulator driven by BLDC motor.}
}

• D. Kamińska and A. Gmerek, "The open-source speech recognition system for robots control," < ">Przegląd Telekomunikacyjny i Wiadomości Telekomunikacyjne< >, 2012.
[BibTeX] [Abstract]

The aim of the project was to develop an open-source speech recognition system, for controlling industrial and social robots. Particular deal of attention has been paid to the interoperability and speed of calculation in order to maintain realtime performance. Therefore, the source code is minimalistic, elastic and easily adaptable. Mel-frequency Cepstral Coefficents (MFCCs) combined with additional features and multistage semantic classifier have been applied in order to provide appropriate speed of calculation and classification accuracy. Dataset can be created by a user as a set of recorded words, which can be assigned to different classes (also specified by a user). Results of classification can be easily combined with specific robot commands. Those commands are sent to a robot through serial port, Telnet or TCP/IP protocol, accordingly, the system can be used with almost every modern robot. Our experiments have been conducted on Kawasaki FS003N robot.

@Article{kam2012gmerekrobot,
Title = {The open-source speech recognition system for robots control},
Author = {Kami{\'n}ska, D. and Gmerek, A.},
Journal = {Przegląd {T}elekomunikacyjny i {W}iadomości {T}elekomunikacyjne},
Year = {2012},
Abstract = {The aim of the project was to develop an open-source speech recognition system, for controlling industrial and social robots. Particular deal of attention has been paid to the interoperability and speed of calculation in order to maintain realtime performance. Therefore, the source code is minimalistic, elastic and easily adaptable. Mel-frequency Cepstral Coefficents (MFCCs) combined with additional features and multistage semantic classifier have been applied in order to provide appropriate speed of calculation and classification accuracy. Dataset can be created by a user as a set of recorded words, which can be assigned to different classes (also specified by a user). Results of classification can be easily combined with specific robot commands. Those commands are sent to a robot through serial port, Telnet or TCP/IP protocol, accordingly, the system can be used with almost every modern robot. Our experiments have been conducted on Kawasaki FS003N robot.},
Publisher = {Politechnika Warszawska}
}

• D. Kamińska and A. Gmerek, "Automatic identification of bird species: a comparison between kNN and SOM classifiers," in < ">Joint Conference NTAV/SPA 2012 New Trends in Audio and Video Signal Processing< >, 2012.
[BibTeX] [Abstract]

This paper presents a system for automatic bird identification, which uses audio input. The experiments have been conducted on three groups of birds, which were created basing finishing on classification, the system is fully automated. The main problem in automatic bird recognition (ABR) is the choice of proper features and classifiers. Identification has been made using two classifiers � kNN (k Nearest Neighbor) and SOM (Self Organizing Maps). System has been tested using data extracted from natural environment.

@InProceedings{kamgmerekbird2012,
Title = {Automatic Identification of Bird Species: a comparison between k{NN} and {SOM} classifiers},
Author = {Kamińska, D. and Gmerek, A.},
Booktitle = {Joint {C}onference {NTAV/SPA} 2012 {N}ew {T}rends in {A}udio and {V}ideo {S}ignal {P}rocessing},
Year = {2012},
Abstract = {This paper presents a system for automatic bird identification, which uses audio input. The experiments have been conducted on three groups of birds, which were created basing finishing on classification, the system is fully automated. The main problem in automatic bird recognition (ABR) is the choice of proper features and classifiers. Identification has been made using two classifiers � kNN (k Nearest Neighbor) and SOM (Self Organizing Maps). System has been tested using data extracted from natural environment.}
}

• D. Pakulski and A. Gmerek, "The electrooculography control system," < ">Pomiary Automatyka Robotyka (PAR)< >, 2012.
[BibTeX] [Abstract]

The aim of the project described in this paper was to develop the methods of recording and analysis of EOG signals meant for manipulator control. Electrooculography (EOG) is a technique for measuring of the resting potential of an eyeball, indicative of the electrical activity of the retina. This paper presents the complete electrooculographic system which cooperates with the special 2-DOF manipulator. The end-effector of manipulator is a laser pointer. In order to adjust signal to manipulator control, data must be collected and then digitally processed. There has been used the nonparametric model (classifier) based on Artificial Neural Networks (ANN). The task of the classifier was the assignment of an unknown fragment of the signal to one of eight classes of the eyeball movements. Application can be used by handicapped patients, who are able to communicate with others by their eyes only.

@Article{pakulskigmerek,
Title = {The electrooculography control system},
Author = {Pakulski, D. and Gmerek, A.},
Journal = {Pomiary {A}utomatyka {R}obotyka ({PAR})},
Year = {2012},
Abstract = {The aim of the project described in this paper was to develop the methods of recording and analysis of EOG signals meant for manipulator control. Electrooculography (EOG) is a technique for measuring of the resting potential of an eyeball, indicative of the electrical activity of the retina. This paper presents the complete electrooculographic system which cooperates with the special 2-DOF manipulator. The end-effector of manipulator is a laser pointer. In order to adjust signal to manipulator control, data must be collected and then digitally processed. There has been used the nonparametric model (classifier) based on Artificial Neural Networks (ANN). The task of the classifier was the assignment of an unknown fragment of the signal to one of eight classes of the eyeball movements. Application can be used by handicapped patients, who are able to communicate with others by their eyes only.}
}

### 2011

• A. Gmerek, "High-level controller for an arm rehabilitation robot-positioning algorithms with respect to EMG data," in < ">16th International Conference on Methods and Models in Automation and Robotics (MMAR)< >, 2011, pp. 182-187.
[BibTeX] [Abstract]

This paper touches upon the issue of dexterous control of rehabilitation robots. There have been proposed a high-level control system and algorithms used for training patients. Emphasis was made especially on the gravitation compensation algorithm, because of the low velocity and acceleration of movement parameters during rehabilitation. The direction of movements was estimated based on the model of surface electromyographic (sEMG) signal, which renders muscle activity very well. The model of manipulator is stable, behaves in a predictable way and, according to the EMG interpretation block, satisfyingly deduces the intention of motion and controls the model.

@InProceedings{gmerek2011high,
Title = {High-level controller for an arm rehabilitation robot-positioning algorithms with respect to {EMG} data},
Author = {Gmerek, A.},
Booktitle = {16th {I}nternational {C}onference on {M}ethods and {M}odels in {A}utomation and {R}obotics ({MMAR})},
Year = {2011},
Organization = {IEEE},
Pages = {182--187},
Abstract = {This paper touches upon the issue of dexterous control of rehabilitation robots. There have been proposed a high-level control system and algorithms used for training patients. Emphasis was made especially on the gravitation compensation algorithm, because of the low velocity and acceleration of movement parameters during rehabilitation. The direction of movements was estimated based on the model of surface electromyographic (sEMG) signal, which renders muscle activity very well. The model of manipulator is stable, behaves in a predictable way and, according to the EMG interpretation block, satisfyingly deduces the intention of motion and controls the model.}
}

• A. Gmerek and E. Jezierski, "Robot do rehabilitacji kończyny górnej - patent," , iss. P-394910, 2011.
[BibTeX]
@Patent{mojpatent,
Title = {Robot do rehabilitacji kończyny górnej - patent},
Nationality = {PL},
Number = {P-394910},
Year = {2011},
Yearfiled = {2011},
Author = {Gmerek, A. and Jezierski, E.},
Owner = {Artur},
Timestamp = {2013.10.31}
}

### 2010

• A. Gmerek, "Adaptacyjny system sterowania protezą ręki wykorzystujący elektromiografię powierzchniową," in < ">Problemy robotyki 2010 Prace Naukowe Politechniki Warszawskiej - Elektronika< >, K. Tchoń and C. Zielińskiego, Eds., Politechnika Warszawska, 2010.
[BibTeX] [Abstract]

An effective control of multi-DOF limb prosthesis is a difficult task, which require innovative approach to the problem. In high-tech limb prosthesis the electrical-nature biological signals are used for this task. Author experimented with EMG (electromyographic) signals, registered with a surface electromyography. In this paper for control a constructed artificial limb an algorithm based on Artificial Neural Networks (ANN) and Fuzzy Logic was proposed. In order to extract unique features of analyzed signals (needed for input of the net), they were expose to multi-level procedure of signal processing.

@InCollection{gmerekadaptacyjny2010,
Title = {Adaptacyjny system sterowania protezą ręki wykorzystujący elektromiografię powierzchniową},
Author = {Gmerek, A.},
Booktitle = {Problemy robotyki 2010 {P}race {N}aukowe {P}olitechniki {W}arszawskiej - {E}lektronika},
Publisher = {Politechnika Warszawska},
Year = {2010},
Editor = {Tchoń, K. and Zielińskiego, C.},
Abstract = {An effective control of multi-DOF limb prosthesis is a difficult task, which require innovative approach to the problem. In high-tech limb prosthesis the electrical-nature biological signals are used for this task. Author experimented with EMG (electromyographic) signals, registered with a surface electromyography. In this paper for control a constructed artificial limb an algorithm based on Artificial Neural Networks (ANN) and Fuzzy Logic was proposed. In order to extract unique features of analyzed signals (needed for input of the net), they were expose to multi-level procedure of signal processing.}
}

• A. Gmerek, "Trudna decyzja Pana Procesora czyli jak sterować protezami przy użyciu sygnałów biologicznych," < ">Forum Akademickie< >, iss. 6, pp. 56-58, 2010.
[BibTeX]
@Article{gmerektrudna,
Title = {Trudna decyzja {P}ana {P}rocesora czyli jak sterować protezami przy użyciu sygnałów biologicznych},
Author = {Gmerek, A.},
Year = {2010},
Number = {6},
Pages = {56--58}
}

• A. Gmerek, "Neuro-EMG-driven model for upper extremity powered orthoses," < ">Forum Innowacji M\lodych Badaczy< >, vol. 1, 2010.
[BibTeX] [Abstract]

The aim of this paper is to develop a special driver based on neuro-EMG model, which is a part of the bigger system - Human Robot Interface dedicated for controlling a powered orthoses for rehabilitation of upper extremities. The system uses myopotentials (EMG signals) in order to estimate the motion intention of the patients. After signal processing and analysis, it is possible not only to improve the rehabilitation process by adding information about fatigue and force, but also fully control the mechanical orthoses based only on EMG signals. The algorithm is based on Artificial Neural Networks and statistical methods.

@Article{neuroemggmerek,
Title = {Neuro-{EMG}-driven model for upper extremity powered orthoses},
Author = {Gmerek, A.},
Journal = {Forum {I}nnowacji {M}{\l}odych {B}adaczy},
Year = {2010},
Volume = {1},
Abstract = {The aim of this paper is to develop a special driver based on neuro-EMG model, which is a part of the bigger system - Human Robot Interface dedicated for controlling a powered orthoses for rehabilitation of upper extremities. The system uses myopotentials (EMG signals) in order to estimate the motion intention of the patients. After signal processing and analysis, it is possible not only to improve the rehabilitation process by adding information about fatigue and force, but also fully control the mechanical orthoses based only on EMG signals. The algorithm is based on Artificial Neural Networks and statistical methods.}
}

### 2009

• A. Gmerek, "Sterowanie złączem o jednym stopniu swobody za pomocą miopotencjałów," < ">Innowacyjne rozwiązania w obszarze automatyki, robotyki i pomiarów< >, pp. 177-185, 2009.
[BibTeX]
@Article{Gmerek2009,
Title = {Sterowanie złączem o jednym stopniu swobody za pomocą miopotencjałów},
Author = {Gmerek, A.},
Journal = {Innowacyjne rozwiązania w obszarze automatyki, robotyki i pomiarów},
Year = {2009},
Pages = {177--185},
Publisher = {Janusz Kacprzyk}
}

• A. Gmerek, "Sterowanie złączem o jednym stopniu swobody przy pomocy miopotencjałów," , vol. 1, pp. 30-31, 2009.
[BibTeX]

• A. Gmerek, "Strategie sterowania modelem protezy ręki z wykorzystaniem miopotencjałów," < ">Pomiary Automatyka Robotyka (PAR)< >, pp. 45-47, 2009.
@Article{gmerek2009strategie,
}