WO2015041618A2

WO2015041618A2 - Upper limb therapeutic exercise robot

Info

Publication number: WO2015041618A2
Application number: PCT/TR2014/000329
Authority: WO
Inventors: Erhan AKDOGAN
Original assignee: Akdogan Erhan
Priority date: 2013-09-20
Filing date: 2014-09-12
Publication date: 2015-03-26
Also published as: WO2015041618A3

Abstract

The therapeutic exercise robot, subject to the invention, is about a therapeutic exercise robot (1) which features a robotic arm (13) to get the upper limbs do therapeutic exercises, a control unit (11) to control the robotic arm, at least one force sensor (1334) making biomechanical measurements, at least one position sensor (1306, 1319, 1329) and an EMG device making biological measurements (120) to measure the reactions of the person undergoing therapy (H), the said force sensor (1334) and/or a position sensor (1306, 1319, 1329) and/or an EMG device (120) sending the biomechanical and biological reactions measured to the control unit (1 1) as feedback, and a control unit (11) which realizes at least one of the following actions to be applied to the person doing the exercise based upon such feedback and makes all the control and/or evaluations by use of the hybrid impedance control method: - change the speed of motion and/or the intensity of motion and/or the position of motion and/or the number of repetition for the motion automatically, i.e. continue operating at a lower value; or - give a warning to change the position of motion and/or the type of motion and/or the number of repetition for the motion and/or the intensity of motion; or - stop the motion, with 3 degree of freedom for upper limbs, and learns and performs passive, active assistive, isotonic, isometric and isokinetic exercises and isotonic-isometric exercises in the same mode of the therapeutic exercise types as well as physiotherapist motions (exercises done manually and active assistive exercises).

Description

SPECIFICATION

UPPER LIMB THERAPEUTIC EXERCISE ROBOT Technical field with which the invention is related:

This invention is related with a therapeutic exercise robot used for physical therapy and rehabilitation of the upper limbs (forearm and wrist) for the purpose of therapeutic exercise. Previous technique:

Home care services has gained importance particularly recently. Physical therapy and rehabilitation is a significant branch where it could be applied. A web-based robotic system with a practical user interface that could be controlled through a therapy center will facilitate the treatment of those persons undergoing therapy with the concept of home care. It will make significant contributions to the individual, their surrounding and, therefore, the society.

Physical therapy and rehabilitation is a medical process for eliminating the problems that occurred in the limbs. For a problematic limb the primary purpose is to ease the existing pain and discomfort within this context. After this procedure the purpose is respectively to earn the limb the full joint movement spread and strengthen the limb. Therapeutic exercises will be used to earn the joint movement spread and strengthen the muscles. A specialist will guide the person undergoing therapy through those exercises, and they could also be made by various mechanical devices.

This treatment process includes various issues. Those are cost, transportation related challenges, time, failure to apply the same therapy conditions in manual exercises one after another, assessments during the manual exercises not being objective, one physiotherapist providing services for only one person undergoing therapy at the same time, limitation of the "number of daily acceptance for persons undergoing therapy" of physiotherapist, limitation of the degree of freedom of the existing mechanical devices used and passive functionality of the same. For this reason, the use of robots in this area has increased in the last 15 years particularly. The recent studies aimed at rehabilitation reveal that the use of robot has several advantages over the traditional techniques. The existing robotic systems have a limited therapeutic exercise capacity and have been particularly developed for the purpose of motor skills training. And measuring, assessing and recording the biomechanical and biological parameters are crucial in physical therapy and rehabilitation.

The benefits of using the robots in the field of physical therapy and rehabilitation are as follows:

• A robot could meet the same treatment conditions with absolute accuracy.

• A robot could measure the values such as force and position accurately through its mechanical sensors.

• A robot could learn the human movements and make those correctly itself.

• A robot could be remote controlled, therefore it will allow for treatment at home for those persons undergoing therapy.

• A physiotherapist could control the operation of several robots. Therefore several persons undergoing therapy could get treatment at the same time.

• A robot could record biomechanical parameters thanks to its force and position sensors. A robot will make more objective measurements compared to a human.

In general the existing robotic systems are controlled with position and force feedback called robot metrics, and measure the biomechanical parameters including joint movement spread, force generated or torque value only.

However movement will first appear in the muscles. Therefore measurements for muscles are also important. Various mechanical and electromechanical devices and tools have been developed for the upper limbs for rehabilitation purposes.

The device called CPM (Continuous Passive Motion) of those devices features a degree of freedom and could make passive exercise motions only.

The systems called CYBEX (US Patent 3465592), BIODEX (Biodex Corporation of Center Moriches, Newyork, USA), Kin-Corn (Chattanooga Group, Ine) could perform passive, active resistant, isotonic and isokinetic exercises for upper limbs. Dynamometer is used for measuring the power. Those systems feature only one motor. The systems feature one degree of freedom and one link. Motion is transmitted to the limbs by moving the link directly by motor stimulation. The patent document, application number US5054774A, in the known status of the technique indicates a data entry method and a computer-controlled muscle exercise device. It is mentioned that the data pertaining to the person undergoing therapy is entered in and recorded on the computer. A computer user interface is used to have the muscles perform isokinetic, isotonic and isometric exercises in muscle exercise machinery. The parameters such as the force applied, speed, number of repetitions are entered with the user interface. Another patent document, application number US5254066A, in the known status of the technique is about a computer-controlled exercise, physical therapy and rehabilitation device. The device makes the joints up to the fingers do concentric, eccentric and isokinetic exercises. The invention features a hydraulic system as a stimulation component.

Another patent document, application number US5583403A, in another known status of the technique mentions about an exercise device which could change the force applied featuring apparatus such as a fixed torque, an motor with a changing speed. It is indicated that isotonic, isokinetic, isotonic/isokinetic, constant, variable, active or passive exercises, etc. could be performed with this exercise device. It is mentioned that those exercises are performed in a computer-controlled manner and the user interface developed is based on programming language C.

Another patent document, application number US5597373, in the known status of the technique indicates that flexion and extension for the hand, wrist and shoulder or wrist motion are done or independent parts of the hand are made to do opposition, flexion or hyper-extension motions. It is said that detectors perceiving the level of muscle contraction are used in the invention. The data received from detectors are evaluated and the level of variable resistance to be applied to the muscle group is determined, and such resistance is used to apply isotonic, isokinetic or isometric types of exercise to patient. It is mentioned that forearm is made to do supination and pronation motions.

Another patent document, application number WO2010108170, in the known status of the technique mentions about a rehabilitation robot system developed as an interactive game based system. The system features a 2 degree of freedom robot and interactive game software realizing the upper limb rehabilitation. The invention is a system where forearm, hand and individual finger exercises are performed. It is indicated that in translational motion of the types of motion in a computer control based system where the data is collected the grasp and release motion of the hand is performed whereas in rotational motion the supination and pronation motion of the forearm is performed.

'

A patent document, application number W09111221A1, in another known status of the technique mentions about a computer-controlled muscle exercise and rehabilitation system. It is indicated that the system gets isokinetic, isotonic and isometric exercises done. In this invention therapists will get the following types of motion performed: flexion - extension, concentric-eccentric and CPM (Continuous Passive Motion).

Additionally the systems in the robotic studies conducted till today are controlled by robot metrics such as force and position. In controlling the rehabilitation robots the impedance force control method and the position control methods such as PD, PID have been used. Moreover the therapeutic exercise robots developed could be used for either the right arm or the left arm. Some mechanical changes are necessary for the systems to perform exercises for both arms.

A brief explanation of the invention:

The purpose of this invention is to put into practice an upper limb therapeutic exercise robot that could learn and perform therapeutic exercises including passive, isometric, isotonic, isokinetic exercises and isotonic-isometric exercises in the same mode for rehabilitating the forearm and wrist of the upper limb as well as physiotherapist motions (exercises performed manually and active assisted) and operates in a single mechanical structure for both arms.

The invention also intends to implement an upper limb therapeutic exercise robot where the position, force and muscle signals could be controlled with feedback and could also be used for the purpose of measurement by evaluating the biological and biomechanical parameters.

The robot, subject to the invention, is controlled by hybrid impedance control technique. The robot, subject to the invention, could be used for both right and left arm without any change in the mechanism thanks to the solution offered in a single mechanical structure for both arms. The therapeutic exercise robot, subject to the invention, features a robotic arm to get the upper limbs do therapeutic exercises, a control unit to control the robotic arm, at least one force sensor (1334) making biomechanical measurements, at least one position sensor (1306, 1319, 1329) and an EMG device making biological measurements (120) to measure the reactions of the person undergoing therapy, the said force sensor (1334) and/or a position sensor (1306, 1319, 1329) and/or an EMG device (120) sending the biomechanical and biological reactions measured to the control unit as feedback, and a control unit which realizes at least one of the following actions to be applied to the person doing the exercise based upon such feedback and makes all the control and/or evaluations by use of the hybrid impedance control method:

- change the speed of motion and/or the intensity of motion and/or the position of motion and/or the number of repetition for the motion automatically, i.e. continue operating at a lower value; or

- give a warning to change the position of motion and/or the type of motion and/or the number of repetition for the motion and/or the intensity of motion; or

- stop the motion.

EMG device that the robot, subject to the invention, includes is used for the purpose of measurement and evaluation. All the muscles included in such motion made by EMG device that the robot, subject to the invention, includes are continuously controlled. The control unit features a main control unit. The main control unit will stop the motion if it finds that the wrong muscles work or the correct motion is performed with the wrong muscles as a result of checking.

The robot also features a control unit including a man-machine interface, data acquisition cards, at least one emergency stop button and motor drivers.

Additionally the robot, subject to the invention, includes a robotic arm featuring a main body, a pronation - supination group, an adduction - abduction group and a flexion - extension group. The main body supports the robotic arm to ensure that it stands upright, not swing or roll over as the motion is performed by the robotic arm and includes moving parts carriers.

The robotic arm features a pronation-supination group with an motor ensuring pronation- supination motions are performed, a gear box, a position sensor, a swing gear, an intermediate swing, a main symmetric swing, a limb connecting apparatus, motion and support rollers, a forearm support component and a pinion; an adduction-abduction group with a position sensor ensuring adduction-abduction motions are performed, a motor, a gear box, a L type interconnection component, a circular connecting component and an interconnection component, and a flexion-extension group with an adjusted duct main part ensuring flexion-extension motions are performed, a tightening apparatus, a link main component, a semi circular part, a position sensor, a motor, a gear box, a grip, a grip - sensor connecting piece and a force sensor. The therapeutic exercise robot, subject to the invention, also includes a control unit featuring a man-machine interface with a user interface, a central controller, a database, a rule base and a hybrid impedance controller.

The man-machine interface features a database where the following data are kept: the personal details of the person undergoing therapy, types of exercise and position and force data relating to the types of exercise, joint motion spread pertaining to the person undergoing therapy following the rehabilitation session, force value generated, position- force relevance charts, muscle activation level parameters, etc. The man-machine interface also includes a control unit which displays the parameters such as the joint motion spread pertaining to the person undergoing therapy during and after the rehabilitation session, force value generated, position-force relevance charts, level of muscle activation on the user interface by means of a screen. The control unit has a rule base where the impedance parameter values to be used on the hybrid impedance controller based on the type of exercise and the responses to the reactions of the person undergoing therapy are stored.

The man-machine interface which the robot, subject to the invention, includes is remote web-based controlled through the user interface. The robot also features a central controller processing EMG signals with a rectification unit where the raw muscle signals received from EMG device are rectified, an amplification unit where such signals are amplified and a filtering unit which filters the signals elevated.

The man-machine interface also includes a hybrid impedance controller generating the torque values to control the robotic arm.

The robot has a force sensor measuring the force with 3 positive (x,y,z) and 3 negative (- x,-y,-z) axes. The upper limb therapeutic exercise robot has 3 degree of freedom (3DOF) which represents the number of motors in the system, thus the number of moving joints with these motors. The said robot could be used at treatment centers and is also suitable for home use thanks to its web-based control feature. Additionally the robot features a database where the test and/or exercise outcomes of the person undergoing therapy are entered ad stored, and a practical user interface where the data could be viewed.

And the forearm of the person undergoing therapy could be made to do pronation- supination motions and the wrist could be made to do flexion-extension and abduction- adduction motions for the purpose of the invention. The invention could perform passive, active assistive, isotonic, isometric, isokinetic motions. And it could learn and perform on the patient the exercises manually performed by physiotherapist. It could also perform isotonic and isometric exercises in the same mode.

The force and position data are recorded in the invention. The robot, subject to the invention, is light so it could be carried effortlessly and adjusted to the arm length of the person undergoing therapy. And the safety of the person undergoing therapy could be checked in terms of software and hardware against any potential issues. The invention has current restriction and an emergency stop button for the motor drivers in terms of hardware. Additionally the limits of mechanical motion of the system are based upon the limits of motion of the human limb. In terms of software the current values transmitted to the motor drivers are limited in order to provide safety with the software developed.

The types of exercise are implemented with the hybrid impedance control rule. This rule will generate for the robotic arm the torque command necessary for the robot to perform the types of exercise. With the hybrid impedance control method the mechanical impedance of the end point of the robotic arm could be adjusted and the desired force and speed trajectory could be followed. The hybrid impedance control rule operates in two modes: the force and position modes.

Robot in therapy to make people passive exercise; motion trajectory information is determined by central controller according to exercise data entered by therapist. Torque value which will be sent to the motor is produced by the hybrid impedance controller. For passive exercises the hybrid impedance controller operates in position impedance control mode. In this mode the purpose is to make the robotic arm optimally follow the motion trajectory created within the limb motion limits defined in the control unit with the user interface path by therapist. Proper impedance parameter values are received from the rule base.

Active assistive exercise is one of those exercises which the person undergoing manual therapy applied by therapist is made to perform in real practices. Therefore an exercise performed by therapist will be modelled by the robot. This type of exercise is performed by the robot as follows. The robotic arm in the force based control mode is moved by the person undergoing therapy and the hybrid impedance parameter values are received from the rule base. The position where the person undergoing therapy could not move the robotic arm will be perceived by means of EMG signals with position feedback and the position based control mode is activated through the software. The robotic arm will make the person undergoing therapy complete the motion.

Isotonic exercises are performed against a fixed force. With the robot, subject to the patent, the exercise is performed as follows. Therapist will select on the user interface the isotonic exercise mode and the force to be applied and the level of flexibility as exercise mode input data. Flexibility will be obtained by changing the hybrid impedance parameter values. Those values are determined as a result of the test studies conducted previously. Accordingly there are three levels of flexibility which are low, medium and high. In the hybrid impedance force based control mode the person undergoing therapy will move the robotic arm. Therefore the predefined force value will be applied by the robotic arm in the opposite direction against the motion of the person undergoing therapy with such flexibility desired.

Isometric exercise will be performed by applying force to the limb in the opposite direction on the joint motion spread. For this purpose elastic band or manually performed exercises are used in current practices for this exercise. The robot, subject to the invention, will perform this exercise as follows. Therapist will enter the force value to be applied in the opposite direction on the joint motion spread by means of the user interface. The person undergoing therapy will move the robotic arm in the position based hybrid impedance control mode. In the meantime the hybrid impedance parameter values will be received from the rule base in such manner to allow the person undergoing therapy to move their limb the most naturally. That the person undergoing therapy reached the joint motion spread will be perceived by position feedback and E G signals by the control unit. At this point the hybrid impedance controller will switch to the force based control mode and start applying force in the opposite direction on the person undergoing therapy.

Isokinetic exercises are performed at a fixed speed. This type of exercise is performed by a device in the previous technique, and it is not possible to do it manually. CYBEX, BIODEX and Kin-Corn are isokinetic therapy devices developed for this purpose. Therapeutic device will not allow the speed of the limb of the person undergoing therapy to go beyond the predefined speed value. Therefore the maximum torque value is generated by the limb. This exercise will be performed as follows in this invention. The person undergoing therapy will move the robotic arm with the lowest parameter values entered in the control unit through the interface in the hybrid impedance force based control mode. The speed of the robotic arm will be obtained by means of the software with the position values measured by the position sensors connected to the joint. The force value generated by the person undergoing therapy will be continuously perceived by force feedback, and the torque value produced by the arm will be calculated by use of such value. The torque value calculated as the speed value of the robotic arm goes beyond the predefined speed value will be directly transmitted to the motor connected to the joint. It will be ensured that the motor generates a torque value which is in the opposite direction of the motion and equal to the torque value produced by the limb of the person undergoing therapy. Thus the person undergoing therapy could not go beyond the speed of the robotic arm over the predefined speed value. A detailed explanation of the invention:

*

Explanation of the figures:

"The upper limb therapeutic exercise robot", subject to this invention, is displayed in the figures enclosed herewith and the explanation of those figures is as follow. Figure 1. A general block diagram pertaining to the operation of the upper limb therapeutic exercise robot subject to the invention

Figure 2. A perspective view of the robotic arm of the upper limb therapeutic exercise robot subject to the invention.

Figure 3. A right-side view of the robotic arm of the upper limb therapeutic exercise robot subject to the invention.

Figure 4. A left-side view of the robotic arm of the upper limb therapeutic exercise robot subject to the invention.

Figure 5. A block diagram of the control unit of the upper limb therapeutic exercise robot subject to the invention.

Figure 6. A block diagram for processing the signals received from EMG device on the upper limb therapeutic robot subject to the invention.

Figure 7. A block diagram of the electronic hardware of the upper limb therapeutic robot subject to the invention.

Figure 8. A schematic view of the passive exercise algorithm of the upper limb therapeutic robot subject to the invention.

Figure 9. A schematic view of the active assistive exercise algorithm of the upper limb therapeutic exercise robot subject to the invention.

Figure 10. A schematic view of the isotonic exercise algorithm of the upper limb therapeutic exercise robot subject to the invention.

Figure 11. A schematic view of the isometric exercise algorithm of the upper limb therapeutic robot subject to the invention.

Figure 12. A schematic view of the isotonic-isometric exercise algorithm of the upper limb therapeutic robot subject to the invention.

Figure 13. A schematic view of the isokinetic exercise algorithm of the upper limb therapeutic exercise robot subject to the invention.

Explanation of the references in the figures: The parts in the figures enclosed herewith are numbered one by one for the purpose of this invention and what those numbers correspond to is mentioned below.

1 Robot

11 Control unit

110. Man-machine interface

111. User interface 112. Central controller

1121. Rectification unit

1122. Amplification unit

1123. Filtering unit

113. Database

114. Rule base

115. Hybrid impedance controller

120. EMG (Electromyograph) device

130. Main control unit

140. Data acquisition cards

150. Emergency stop button

160. Motor drivers

13 Robotic arm

131. Main Body

1301. Moving parts carrier

1302. Moving parts carrier

1303. Moving parts carrier

132 Pronation-Supination Group

1304. Motor

1305. Gear box

1306. Position sensor

1307. Swing gear

308. Intermediate swing

1309. Main symmetric swing

1310. Limb connecting apparatus

1311. 1312., 1313., 1314., 1315., 1316. Motion and support rollers

1317. Forearm support component

1318. Pinion

133 Adduction-Abduction Group

1319. Position sensor

1320. Motor

1321. Gear box

1322. L type interconnection component

1323. Circular connection component

1324. Interconnection component 134. Flexion-Extension Group

1325. Adjusted duct main part

1326. Tightening apparatus

1327. Main link component

1328. Semi circular part

1329. Position sensor

1330 Motor

1331 Gear box

1332 Grip

1333 Grip-sensor connecting piece

1334 Force sensor

V. Treatment center database

T. Therapist

H. Person undergoing therapy

The upper limb therapeutic exercise robot, subject to the invention (1), includes a control unit (11) and a robotic arm (13) (Figure 1).

The control unit (11) features a man-machine interface (110), an EMG device (120), a main control unit (130), data acquisition cards (140), at least one emergency stop button (150) and motor drivers (160).

The robotic arm (13) features four components: a main body (131), a pronation - supination group (132), an adduction - abduction group (133) and a flexion - extension group (134) (Figure 2, Figure 3, Figure 4).

The main body (131) supports the robotic arm to ensure that the robotic arm (13) stands upright, not swing or roll over as the motion is performed by the robotic arm (13). The main body (131) also includes moving parts carriers (1301 , 1302, 1303).

Pronation-supination group (132) features a motor ensuring pronation-supination motions (1304) are performed by the robot (1), a gear box (1305), a position sensor (1306), a swing gear (1307), an intermediate swing (1308), a main symmetric swing (1309), a limb connecting apparatus (1310), motion and support rollers (1311 , 1312, 1313, 1314, 1315, 1316 ), a forearm support component (1317) and a pinion (1318). Adduction-abduction group (133) features a position sensor (1319) ensuring adduction- abduction motions are performed by the robot (1), a motor (1320), a gear box (1321), an L type interconnection component (1322), a circular connecting component (1323) and an interconnection component (1324).

Flexion-extension group (134) features an adjusted duct main part (1325) ensuring flexion-extension motions are performed by the robot (1), a tightening apparatus (1326), a main link component ( 327), a semi circular part (1328), a position sensor ( 329), a motor (1330), a gear box (1331), a grip (1332), a grip - sensor connecting piece (1333) and a force sensor (1334).

The robotic arm (13) will perform the pronation - supination motions for forearm with the following components: a motor (1304), a gear box (1305), a position sensor (1306), a swing gear (1307), an intermediate swing (1308), a main symmetric swing (1309), a limb connection apparatus (1310), motion and support rollers (1311 , 1312, 1313, 1314, 1315, 1316 ), a forearm support component (1317), a pinion (1318) and a force sensor (1334). The patient's arm will be connected to the robotic arm (13) with the limb connection apparatus (1310). The torque command received from the motor (1304) will be enhanced by the gear box (1305) and the motion will be transmitted to the pinion (1318). Thus the swing gear (1307) connected to the pinion (1318) will move on the intermediate swing (1308), main symmetric swing (1309) parts, movement and support rollers (1311 , 1312, 1313, 1314, 1315, 1316 ). The position details will be continuously measured by the position sensor (1306) and used as feedback data. The robotic arm (13) will also perform abduction-adduction motions for the wrist with a position sensor (1319), a motor (1320), a gear box (1321), an L type interconnection component (1322), a circular connection component (1323) and an interconnection component (1324). The patient's arm will be connected to the robotic arm (13) with a limb connection apparatus (1310). The torque command received from the motor (1320) will be enhanced by the gear box (1321) and the motion will be provided with a circular connection component (1323) and an interconnection component (1324).

The robotic arm (13) will perform the flexion-extension motions for the wrist with the following components: an adjusted duct main part (1325), a tightening apparatus (1326), a main link component (1327), a semi circular part (1328), a position sensor (1329), a motor (1330), a gear box (1331), a grip (1332), a grip-sensor connecting piece (1333) and a force sensor (1334).

The man-machine interface (110) which is the software controlling the robot (1) features a user interface (111), a central controller (112), a database (113), a rule base (114) and a hybrid impedance controller ( 5) (Figure 5).

All the data recorded during the rehabilitation session will be sent to the treatment centre database (V) by the main control unit (130). This database is hosted by a computer outside connected to the main control unit (130) with a network. The entire rehabilitation team could access the data pertaining to the person undergoing therapy (H) kept there in the database (V). Therefore the condition of the person undergoing therapy (H) could be closely monitored. It enables to develop treatment methods for those persons undergoing therapy (H) in a similar condition.

A central controller (112) will provide the communication between all the units on the man- machine interface (110).

The personal details pertaining to the person undergoing therapy (H), the types of exercise and the position and force data relating to the types of exercise, etc. are kept in the database (113). The joint motion spread pertaining to the person undergoing therapy (H) following the rehabilitation process, force value generated, position-force relevant charts, muscle activation pattern parameters are also stored on that unit. In the upper limb therapeutic exercise robot (1), subject to the invention, the therapist (T) will enter the details of the person undergoing therapy (H) in the control unit (1 ) by means of a user interface (111) and select the type of exercise. The joint motion spread pertaining to the person undergoing therapy (H) during and after the rehabilitation session, force value generated, position-force relevance charts, muscle activation pattern parameters are also viewed on the user interface (111). Those images could be printed out with a printer. The user interface ( ) will be presented to the user by means of a screen.

The rule base (114) will keep the impedance parameter values to be used for the hybrid impedance controller (115) based on the type of exercise. It will also store the responses of the robot (1) to the reactions of the person undergoing therapy (H). Those reactions could be predefined or could be learned by the robotic arm (13) itself. That is to say the robotic arm (13) will store each reaction and create a new response to the same. For instance; the robot (1) is capable of making the person undergoing therapy perform a predefined new type of motion and/or number of repetitions following the reaction of the person undergoing therapy to the robot (1) based on the intensity of the reaction during a practice for the invention.

The reactions of the person undergoing therapy (H) will be evaluated by a central controller (1 12). For this evaluation the data received from EMG device (120) and/or force sensor (1334) and/or position sensor (1306, 1319, 1329) are used. The robot (1) could be controlled remotely as web-based through a user interface (111).

Central controller (112) features a rectification unit (1121 ), an amplification unit (1122) and a filtering unit (1 123) in order to process EMG signals. The raw muscle signals received from EMG device (120) will be rectified, amplified and filtered. The raw EMG signal will first enter the rectification unit (1121), then the amplification unit (1122) and finally the filtering unit (1123) to be processed. This process is called the processing of EMG signal (Figure 6). In a practice of the invention Butterworth, a low pass filter circuit with a certain corner frequency value is used for filtering. EMG signals are perceived and processed with a predefined sampling time. In the robot (1) the processing of EMG signal is performed by means of a central controller (1 12). Rectification unit (1 121), amplification unit (1122) and filtering unit (1123) are implemented in terms of software on the system but could also be implemented with electronic components in terms of hardware.

Hybrid impedance controller (115) is a conventional controller generating the torque values to control the robotic arm (13). The rule of hybrid impedance control is a robot control method developed by Anderson & Spong (1988) (IEEE Journal of Robotics and Automation, (Volume:4, Issue^" 5). The rule of hybrid impedance control allows controlling the position and force within a single control structure unlike other impedance rules. This method is used on the robot (1) to ensure that the types of exercise are performed by the robotic arm (13). Hybrid impedance control features two sub-modes: position- and force- based. Force measurements are conducted with 3 positive and 3 negative axes. That is to say force measurements could be made in 3 directions: x,y,z and -x,-y,-z. For the robot, subject to the patent, (1) the control made by the hybrid impedance controller (1 15) is performed by a man-machine interface (110). The robotic arm (13) features a structure with 3 degree of freedom. In a practice of the invention the robotic arm (13) is produced from aluminium because it is a lightweight material. And it is also quite easy to carry the robotic arm (13) as it is a lightweight material. The production cost of the robotic arm (13) is quite budget friendly because the aluminium material used for the invention is cheap and easy to find. Additionally the robotic arm ( 3) will operate smoothly as it is made of aluminium.

The mechanic structure of the robotic arm (13) is suitable for doing exercises for the wrist and forearm. The robotic arm (13) performs flexion-extension, abduction-adduction motions for the wrist and pronation-supination motions for the forearm. Rotation of the forearm to have the palm face upwards is called supination and the rotation to have the palm face downwards is called pronation. Approximation of the palm to the body as it faces downwards is called radial deviation (abduction) and deviation from the body is called ulnar deviation (adduction). Bending of the wrist is called flexion and stretching means extension.

The electronic hardware of the robotic arm (13) is illustrated in Figure 7. The main control unit (130) includes man-machine interface ( 10) software and ensures that the software is run. Man-machine interface (110) is downloaded to the control unit (11). This interface (110) is operated by the control unit (11). The main control unit (130) may be a computer or a micro-processor or a micro-controller or a programmable logic controller (PLC) or an embedded controlling system. Data acquisition cards (140) are used to process all analogue and digital data on the robot (1). Data acquisition cards (140) will generate an analogue output to drive the motors (1304, 1320, 1330), and receive data from the position sensors (1306, 1319, 1329) to process the position feedback details, from the force sensor (1334) to process the force feedback details and from EMG device (120) to process the biological feedback signals. Data acquisition cards (140) may be of such type that is placed in computer PCI data path or connected through USB. Instead of data acquisition cards (140) the electronic circuits that make analogue - digital and digital - analogue data transformation could also be used. The robot (1) also features an emergency stop button (150) on it to be used in case of an emergency. The motors (1304, 1320, 1330) are driven with motor drivers (160). The robotic arm (13) will be activated with such torque received from the motors (1304, 1320, 1330). Feedback data will be received on the position sensors (1306, 1319, 1329), force sensor (1334) and EMG device connected to the patient's (H) limb on it and transmitted to data acquisition card (140). The robot (1) could do passive, active assistive, isotonic, isometric, isokinetic exercises and isotonic-isometric exercises on the same mode. Personal details pertaining to the person undergoing therapy (H) (name, surname, sex, age, etc. and height, weight, wrist length, forearm length values) will be defined by the therapist (T) by means of a user interface (111) or the person undergoing therapy (H) in case of home use before the exercises start. User interface (111) is viewed by the user by means of a screen.

At every step of the algorithm the following control steps will be continuously performed during the operation of the robotic arm (13) as long as any exercise is done for all types of exercise:

- Checking the force and muscle activation level data received from the person undergoing therapy (H) continuously

- Reverting the robotic arm (13) to the starting position if at least one of the force and muscle activation values goes beyond the predefined level, otherwise resuming the algorithm.

For passive exercise a man-machine interface (110) user interface (111) will be used and the exercise data will be entered in the main control unit (130) by the therapist (T). The therapist (T) may be a nursemaid, a physician or another informed person or the person undergoing therapy based on the practices for the invention. Such exercise data will be converted to motion trajectory data by a central controller (112). Hybrid impedance controller (115) will operate in position control mode in this type of exercise and generate such torque data to be sent to the motors (1304, 1320, 1330) by use of such motion trajectory data. The robotic arm (13) will move with such torque data.

Passive exercise will be performed by a main control unit (130) according to the following motion algorithm (100) for the upper limb therapeutic exercise robot (1) subject to the invention: - Personal details pertaining to the person undergoing therapy (H) (i.e. name, surname, sex, age, etc. and height, weight, wrist length, forearm length values) will be defined in the control unit (11) through user interface (11 ) (101)

- The arm of the person undergoing therapy (H) will be placed in robotic arm (13) (102)

- Therapist (T) will select the passive exercise mode through user interface (111) (103) - Therapist (T) will select any types of motion (flexion-extension or abduction- adduction or pronation-supination) through user interface (111 ) (104)

- Therapist (T) will enter the speed, number of repetitions for the motion, joint motion spread limit values relating to the exercise (105)

- Therapist (T) will enter the data relating to the reaction force and muscle activation level of the person undergoing therapy (H) representing the pain symptom of the person undergoing therapy (H) (106)

- Therapist (T) will push the exercise start button (107)

- Based on the type of motion selected by the therapist (T) robotic arm (13) will move between the angle values of that motion with such torque command to be generated by hybrid impedance controller (115) for motors in position control mode (1304, 1320, 1330) (108)

- The number of motions will be checked and if it is completed the step number (108) of the algorithm will resume (109)

- If the number of repetitions for the motion is completed the exercise will be finalized (112)

- Robotic arm (13) will stop (110) (Figure-8).

For the passive exercise the said "resuming" process represents stopping the exercise and going back to the ordinary posture of the leg.

Active assistive exercise is performed by therapists (T) in practice. The robot subject to the invention will perform this exercise round as follows: The person undergoing therapy (H) will place their arm to the limb connecting apparatus (1310) and hold the grip (1332) and robotic arm (13) will move in the direction of the force as long as it is applied. The force value will be perceived by force sensor (1334). Hybrid impedance controller (115) will run in the force control mode at the time. That the person undergoing therapy could not apply any force or move their limb will be perceived by EMG device (120), force sensor (1334) and position sensors (1306, 1319, 1329) and hybrid impedance controller (115) will switch to position control mode and make the person undergoing therapy (H) complete the motion.

Active assistive exercise algorithm will be performed by a main control unit (130) according to the following steps (200) for the upper limb therapeutic exercise robot (1) subject to the invention: - Personal details pertaining to the person undergoing therapy (H) (i.e. name, surname, sex, age, etc. and height, weight, wrist length, forearm length values) will be defined in the control unit (11) through user interface (111) (201)

- The arm of the person undergoing therapy (H) will be placed in robotic arm (13) (202)

- Therapist (T) will select active assistive exercise mode through user interface (111) (203)

- Therapist (T) will select the type of motion (flexion-extension or abduction- adduction or pronation-supination) through user interface (111) (204)

- Therapist (T) will enter the limit value for the joint motion spread relating to the exercise (205)

- Therapist (T) will enter the data relating to the reaction force and muscle activation level of the person undergoing therapy representing the pain symptom of the person undergoing therapy (H) (206)

- Therapist (T) will push the exercise start button (207)

- Patient (H) will start moving robotic arm (13) with the lowest parameter values of hybrid impedance force control mode (208)

- The person undergoing therapy (H) will take their arm to such spread together with robotic arm (13) as far as they could move it (209)

- Events where the person undergoing therapy (H) could not move their arm will be perceived by EMG device ( 20) by means of position, force and muscle activation feedback signals (210)

- Where the person undergoing therapy (H) could not move their arm, hybrid impedance position control mode will be operated by control unit (11) (211) - Robotic arm (13) will take the arm of the person undergoing therapy (H) to the limit value for the joint motion spread defined through user interface (111) before therapy from the moment the position control mode is activated (212) (Figure-9).

Hybrid impedance controller (115) will work in force control mode during isotonic exercise. Isotonic resistance levels will be acquired by adjusting the forcer and control parameters desired. Data relating to such resistance levels will be stored in rule base (114). And the force and control parameters are the input signs of hybrid impedance controller (115). Therapist (T) will enter the force values to be applied by the robotic arm (13) against the motion of the person undergoing therapy (H) through user interface ( 1). Therapist (T) will also enter in the control unit (11) the level of severity of such force value to be applied to the person undergoing therapy (H) through user interface (111). The level of severity of such force value to be applied to the person undergoing therapy (H) could be defined by a physician previously based upon the practices relating to the invention or by the therapist (T) at the time of application. In a practice relating to the invention those levels are adjusted as low, medium, severe. Force value to be applied by the person undergoing therapy (H) to robot (1) will be measured by force sensor (1334).

In the upper limb therapeutic exercise robot (1), subject to the invention, isotonic motion algorithm (300) will be performed by main control unit (130) based on the following steps:

- Personal details pertaining to the person undergoing therapy (H) (i.e. name, surname, sex, age, etc. and height, weight, wrist length, forearm length values) will be defined in the control unit (11) through user interface (111) (301)

- The arm of the person undergoing therapy (H) will be placed in robotic arm (13) (302)

- Therapist (T) will select the isotonic exercise mode through user interface (111) (303)

- Therapist (T) will select the type of motion (flexion-extension or abduction- adduction or pronation-supination) through user interface (111) (304)

- Therapist (T) will enter in control unit (11) the resistance force value, level of resistance and number of repetitions for the motion relating to the exercise through user interface (111) (305)

- Therapist (T) will enter in control unit (11) the details of reaction force and level of muscle activation relating to the person undergoing therapy (H) through user interface (111) (306)

- Therapist (T) will push the exercise start button (307)

- Hybrid impedance force control mode will be activated (308)

- The person undergoing therapy (H) will move their arm (309)

- The number of motions will be checked and if it is completed the step number (308) of the algorithm will resume (310)

- Number of repetitions will be checked and if it is completed the exercise will be over (311)

- Robotic arm (13) will stop (312) (Figure-10).

Isometric exercise will be performed by applying a fixed torque value to the limb of the person undergoing therapy (H) in the joint motion spread. For this exercise the therapist (11) will enter such force value to be applied to the limb of the person undergoing therapy (H) by robotic arm (13) through user interface (121). Such force will be applied to the person (H) by robotic arm (13) as the joint of the person undergoing therapy (H) reaches the desired spread of motion. The person undergoing therapy (H) will hold the grip (1332) and move their arm. At the time the hybrid impedance controller (115) will have such input parameters that will allow the person undergoing therapy (H) to move their arm the most easily. Those parameter values will be stored in a rule base (114). Whether the limb of the person undergoing therapy (H) has reached the joint motion spread will be checked with instant values measured by force sensor (1334), a position sensor (1306, 1319, 1329) and an EMG device (120). Those values will be evaluated by a man-machine interface (110) included on the main control unit (130). As the limb of the person undergoing therapy (H) reaches the joint motion spread, the motors (1304, 1320, 1330) connected to robotic arm (13) will generate a torque and apply force to the person undergoing therapy (H) in the opposite direction.

In the upper limb therapeutic exercise robot (1), subject to the invention, isometric motion algorithm (400) will be performed by main control unit (130) based on the following steps:

- Personal details pertaining to the person undergoing therapy (H) (i.e. name, surname, sex, age, etc. and height, weight, wrist length, forearm length values) will be defined in the control unit (11) through user interface (111) (401)

- The arm of the person undergoing therapy (H) will be placed in robotic arm (13) (402)

- Therapist (T) will select the isometric exercise mode through user interface (111) (403)

- Therapist (T) will select the type of motion (flexion-extension or abduction- adduction or pronation-supination) through user interface (111) (404)

- Therapist (T) will enter in control unit (11) the resistance force value to be applied in the joint motion spread and the number of repetitions for the motion relating to the exercise through user interface (111) (405)

- Therapist (T) will enter the data relating to the reaction force and muscle activation threshold level of the person undergoing therapy (H) representing the pain symptom of the person undergoing therapy (H) (406)

- Therapist (T) will push the exercise start button (407)

- Hybrid impedance force control mode will be activated by control unit (11) with the lowest parameter and force value (408)

- The person undergoing therapy (H) will move their arm to the joint motion spread limit (409) - That the arm of the person undergoing therapy (H) has reached the joint motion spread limit will be perceived by control unit (11) by means of the data received from position sensors (1334) (410)

- A torque value equal to the predefined resistance force value will be applied to the person undergoing therapy (H) by robotic arm (13) (411)

- The person undergoing therapy (H) will resist the robotic arm (13) in such manner not to allow robotic arm (13) to move their arm (412)

- If the person undergoing therapy (H) does not resist, it will be perceived by force sensors (1334) and position sensors (1306, 1319, 1329) (413)

- If the person undergoing therapy (H) does not resist, the main control unit (130) will operate in hybrid impedance position control mode and robotic arm (13) will be brought to the beginning point by hybrid impedance controller (115) (414)

- The number of motions will be checked and if it is completed the step number (408) of the algorithm will be resumed (415)

- If the number of repetitions is completed, the exercise will be over (416)

- Robotic arm (13) will stop (417) (Figure-11).

Isometric and isotonic exercises could be done together with the robot (1) subject to the invention.

Hybrid exercise motion algorithm (500) where isotonic and isometric exercises are performed at the same time with the upper limb therapeutic exercise robot (1), subject to the invention, includes the following steps:

- Personal details pertaining to the person undergoing therapy (H) (i.e. name, surname, sex, age, etc. and height, weight, wrist length, forearm length values) will be defined in the control unit (1 ) through user interface (111) (501)

- The arm of the person undergoing therapy (H) will be placed in robotic arm (13) (502)

- Therapist (T) will select the hybrid (isotonic plus isometric) exercise mode through user interface (111) (503)

- Therapist (T) will select the type of motion (flexion-extension or abduction- adduction or pronation-supination) through user interface (11 ) (504)

- Therapist (T) will enter the resistance force value, level of resistance, number of repetitions for the motion and force values to be applied in the joint motion spread relating to the exercise through user interface (111) (505) - Therapist (T) will enter the data relating to the reaction force and muscle activation level of the person undergoing therapy representing the pain symptom of the person undergoing therapy (H) through user interface (111) (506)

- Therapist (T) will push the exercise start button (507)

- Hybrid impedance force control mode will be run by control unit (11) (508)

- The person undergoing therapy (H) will move their arm (509)

- The person undergoing therapy (H) will beat the predefined resistance level and take their arm up to the joint motion spread limit (5 0)

- That the arm of the person undergoing therapy (H) has reached the joint motion spread limit will be perceived by control unit (11) by means of the data received from position sensors (1306, 1319, 1329) (511)

- A torque value equal to the predefined resistance force value will be applied to the person undergoing therapy (H) by robotic arm (13) (512)

- The arm of person undergoing therapy (H) will resist the robotic arm (13) in such manner that robotic arm (13) will not allow the person undergoing therapy (H) to move their arm (513)

- If the person undergoing therapy (H) does not resist, it will be perceived by force sensors (1334) and position sensors (1306, 1319, 1329) (514)

- Robotic arm (13) will be brought back to the starting point by hybrid impedance controller (115) with position control by main control unit (130) (515)

- The number of motions will be checked and if it is completed the step number (508) of the algorithm will be resumed (516)

- If the number of repetitions is completed, the exercise will be over (517)

- Robotic arm (13) will stop (518) (Figure -12).

Isokinetic exercise is performed by those devices preventing the limb motion speed from going beyond the predefined value. Thus maximum contraction will occur on the limb of the person undergoing therapy (H). For this exercise the therapist (T) will enter the speed value which the robotic arm (13) may not exceed through user interface (111). The person undergoing therapy (H) may not take their limb beyond such speed value during the exercise. As the person undergoing therapy (H) reaches the predefined speed value, a torque value equal to the torque value generated by the person undergoing therapy (H) will be applied by robotic arm (13) against the person undergoing therapy (H). The speed of the limb of the person undergoing therapy (H) will be obtained once the derivative of the position details received from position sensors (1306, 1319, 1329) is calculated by a central controller (112). Torque value generated by the limb of the person undergoing therapy will be calculated by central controller (112) by use of the force data received from force sensor (1334).

Isokinetic motion algorithm (600) includes the following steps in the upper limb therapeutic exercise robot ( ), subject to the invention:

- Personal details pertaining to the person undergoing therapy (H) (i.e. name, surname, sex, age, etc. and height, weight, wrist length, forearm length values) will be defined in the control unit (11) through user interface (111) (601)

- The arm of the person undergoing therapy (H) will be placed in robotic arm (13) (602)

- Therapist (T) will select the isokinetic exercise mode through user interface ( 1 ) (603)

- Therapist (T) will select the type of motion (flexion-extension or abduction- adduction or pronation-supination) through user interface (111) (604)

- Therapist (T) will enter the speed value and the number of repetitions for the motion relating to the exercise through user interface (111) (605)

- Therapist (T) will push the exercise start button (606)

- Control unit (11) will start operating hybrid impedance force control mode with the lowest parameter and force value (607)

- The person undergoing therapy (H) will move the arm (608)

- The speed of motion will be controlled by control unit (11) (609)

- If the speed of motion reaches the predefined level, a torque value equal to the torque value generated by the person undergoing therapy (H) will be produced by motor (1304, 1320, 1330) and robotic arm (13) will resist to the motion of the person undergoing therapy (610)

- If the speed of motion is below the predefined level, hybrid impedance force control will be enabled (611 )

- The number of motions will be checked and if it is completed the step number (607) of the algorithm will be resumed (612)

- Upon completion of the number of repetitions the exercise will be over (613)

- Robotic arm (13) will stop (614) (Figure -13).

The robot (1), subject to the patent, could be used as both a therapeutic exercise robot and a biomechanical and biological parameter measurement robot. Biomechanical measurements are made by means of a force sensor ( 334) and position sensors (1306, 1319, 1329). And biological parameter measurements are made by EMG device (120). The results of measurement are stored in a database (113).

Muscle activation will be evaluated by robotic control unit (11) and bio-feedback will be measured by EMG device (120) and recorded to database (113) so as to collect data relating to the person undergoing therapy (H). Such data could be used to determine the type of exercise which is suitable for the treatment of those patients under similar conditions. The robot (1), subject to the invention, could learn the therapist's (T) motions and apply those motions learned to the person undergoing therapy (H). Control unit (11) will evaluate the data received from position sensors (1306, 1319, 1329) and force sensor (1334) and EMG device (120). The data will be stored in a database (113). Therapist (T) will select the type of practice (111) on user interface and the data in database (113) will be sent from main control unit (130) to hybrid impedance controller ( 5). Hybrid impedance controller (115) will convert the data to motor torque data and activate the motors (1304, 1320, 1330). With the motors (1304, 1320, 1330) activated, robotic arm (13) will get the person undergoing therapy (H) perform the motions. With respect to the robot (1), subject to the invention, EMG device (120) is used for the purpose of measurement and evaluation as well as control during exercises.

EMG device will measure the level of muscle activation and transmit muscle contractions to control unit (11) as feedback. The force to be applied to the person undergoing therapy (H) could be changed transiently when necessary based on the results of measurement with EMG device (120) at the time of getting the person undergoing therapy (H) do any exercise. Muscle contraction level or level of force applied in the opposite direction of the motion will be continuously measured by EMG device (120) and/or force sensor (1334) and/or position sensor (1306, 1319, 1329) when the person undergoing therapy (H) is doing a motion. If the predefined threshold levels are exceeded it will represent the pain. In that case robot (1) will stop the exercise and go back to the starting joint motion spread. If this happens, main control unit (130) will take an action to stop the motion automatically. All muscles included in the motion will be controlled by EMG device (120) in the robot (1), subject to the invention. If it is found that the wrong muscles work or the correct motion is done with the wrong muscles working as a result of control, main control unit (130) will stop the exercise. In case of an emergency the motion could be stopped by pushing the emergency stop button (150). The invention is not limited to the abovementioned practices and a person specialized in the technique could effortlessly bring about various practices relating to the invention. Those should be considered within the scope of the protection required for the invention with the requests.

Claims

1. A therapeutic exercise robot (1) comprising: a robotic arm (13) to get the upper limbs do therapeutic exercises, a control unit (11) to control the robotic arm and characterized in that it also comprises at least one force sensor (1334) making biomechanical measurements, at least one position sensor (1306, 1319, 1329) and an EMG device making biological measurements (120) to measure the reactions of the person undergoing therapy (H), the said force sensor (1334) and/or a position sensor (1306, 1319, 1329) and/or an EMG device (120) sending the biomechanical and biological reactions measured to the control unit (11) as feedback, and a control unit (11) which realizes at least one of the following actions to be applied to the person doing the exercise based upon such feedback and makes all the control and/or evaluations by use of the hybrid impedance control method:

- stop the motion,

with 3 degree of freedom for upper limbs, and learns and performs passive, active assistive, isotonic, isometric and isokinetic exercises and isotonic-isometric exercises in the same mode of the therapeutic exercise types as well as therapist motions (exercises done manually and active assistive exercises). 2. A robot (1) according to claim 1 characterized by an EMG device (120) continuously controlling all the muscles included in the motion performed and a control unit (11) featuring a main control unit (130) which would stop the motion if it finds that the wrong muscles work or the correct motion is done but with the wrong muscles working as a result of control.

3. A robot (1) according to in claim 2 which is characterized by a control unit (11) featuring a man-machine interface (110), data acquisition cards (140), at least one emergency stop button ( 50) and motor drivers (160)

4. A robot (1) according to claim 3 which is characterized by a robotic arm (13) featuring a main body (131), a pronation-supination group (132), an adduction-abduction group (133) and a flexion-extension group (134). 5. A robot (1) according to claim 4 which is characterized by a main body (131) that supports the robotic arm ( 3) to ensure that it stands upright, not swing or roll over as the motion is performed by the robotic arm (13) and includes moving parts carriers (1301 , 1302, 1303) 6. A robot (1) according to claim 4 or 5 characterized by a robotic arm (13) which features a pronation-supination group (132) with a motor ensuring pronation-supination motions (1304) are performed, a gear box (1305), a position sensor (1306), a swing gear (1307), an intermediate swing (1308), a main symmetric swing (1309), a limb connecting apparatus (1310), motion and support rollers (1311 , 1312, 1313, 1314, 1315, 1316), a forearm support component (1317) and a pinion (1318)

7. A robot (1) according to any of the claims 4-6 characterized by a robotic arm (13) featuring an adduction-abduction group (133) with a position sensor (1319) ensuring adduction-abduction motions are performed, a motor (1320), a gear box (1321), a L type interconnection component (1322), a circular connecting component (1323) and an interconnection component (1324).

8. A robot (1) according to any of the claims 4-7 characterized by a robotic arm (13) featuring a flexion-extension group (134) with an adjusted channel main part (1325) ensuring flexion-extension motions are performed, a tightening apparatus (1326), a main link component (1327), a semi circular part (1328), a position sensor (1329), a motor (1330), a gear box (1331), a grip (1332), a grip - sensor connecting piece (1333) and a force sensor (1334). 9. A robot (1) according to any of the claims 3-8 characterized by a control unit (1 ) featuring a man-machine interface (110) with a user interface (111), a central controller (112), a database (113), a rule base (114) and a hybrid impedance controller (115)

10. A robot (1) according to claim 9 characterized by a database (1 3) where the following data are kept: the personal details of the person undergoing therapy (H), types of exercise and position and force data relating to the types of exercise, joint motion spread pertaining to the person undergoing therapy following the rehabilitation session, force value generated, position-force relevance charts, muscle activation level parameters, etc.

11. A robot (1) according to claim 10 characterized by a control unit (11) which displays the parameters such as the joint motion spread pertaining to the person undergoing therapy (H) during and after the rehabilitation session, force value generated, position- force relevance charts, level of muscle activation on user interface (111) by means of a screen. 12. A robot (1) according to claim 11 characterized by a control unit (11) featuring a rule base (114) where the impedance parameter values to be used on the hybrid impedance controller (115) based on the type of exercise and the responses to the reactions of the person undergoing therapy (H) are stored. 13. A robot (1) according to claim 12 characterized by being remote web-based controlled through user interface (111).

14. A robot (1) according to claim 13 characterized by a central controller (112) processing EMG signals with a rectification unit (1121) where the raw muscle signals received from EMG device (120) are rectified, an amplification unit (1122) where such signals are amplified and a filtering unit ( 123) which filters the signals elevated.

15. A robot (1) according to claim 14 characterized by a hybrid impedance controller ( 15) and a man-machine interface (110) including a hybrid impedance controller ( 5) generating the torque values to control the robotic arm ( 3).

16. A robot (1) according to claim 14 characterized by a force sensor ( 334) measuring the force with 3 positive (x,y,z) and 3 negative (-x,-y,-z) axles. 17. A robot (1) according to any of the claims 9-16 characterized by a central controller (112) converting the exercise data entered in main control unit (130) by therapist into motion trajectory data to get the person undergoing therapy (H) do passive exercise, a hybrid impedance controller (115) generating such torque data to be sent to motors (1304, 1320, 1330) by use of the said motion trajectory data. 18 A robot (1) according to claim 17 characterized by a main control unit (130) which will take the following steps at every step of the algorithm as long as any exercise is done for all types of exercise:

- Checking the force and muscle activation level data received from the person undergoing therapy (H) continuously and

19. A robot (1) according to claim 18 characterized by a main control unit (130) which performs passive exercises to be applied to the person undergoing therapy (H) according to the following steps:

Personal details pertaining to the person undergoing therapy (H) (i.e. name, surname, sex, age, etc. and height, weight, wrist length, forearm length values) will be defined in the control unit (1 1) through user interface (11 1 ) (101 )

The arm of the person undergoing therapy (H) will be placed in robotic arm (13)

(102)

Therapist (T) will select the passive exercise mode through user interface (111) (103)

Therapist (T) will select any types of motion (flexion-extension or abduction- adduction or pronation-supination) through user interface (1 11 ) (104)

Therapist (T) will enter the speed, number of repetitions for the motion, joint motion aperture limit values relating to the exercise (105)

Therapist (T) will enter the data relating to the reaction force and muscle activation level of the person undergoing therapy (H) representing the pain symptom of the person undergoing therapy (H) (106)

Therapist (T) will push the exercise start button (107)

Based on the type of motion selected by the therapist (T) robotic arm (13) will move between the angle values of that motion with such torque command to be generated by hybrid impedance controller ( 15) for motors in position control mode (1304, 1320, 1330) (108)

The number of motions will be checked and if it is completed the step number (108) of the algorithm will be resumed (109)

If the number of repetitions for the motion is completed the exercise will be finalized (110) - Robotic arm (13) will stop (111)

20. A robot (1) according to any of the claims 9-19 characterized by a robotic arm (13) that will move in the direction of the force as long as the person undergoing therapy (H) places their arm to the limb connecting apparatus and hold the grip (1332) applying force, a force sensor (1334) that will perceive the force value, a hybrid impedance controller (115) that will run in the force control mode as long as the person applies force, an EMG device (120), a force sensor (1334) and position sensors (1306, 1319, 1329) that will perceive any situation where the person undergoing therapy (H) could not apply force, and a hybrid impedance controller (115) that will switch to position control mode when the person undergoing therapy (H) could not apply force and make them complete the motion.

21. A robot (1) according to claim 20 characterized by a main control unit (130) that performs active assistive exercises to be applied to the person undergoing therapy according to the following steps:

- Personal details pertaining to the person undergoing therapy (H) (i.e. name, surname, sex, age, etc. and height, weight, wrist length, forearm length values) will be defined in the control unit (11) through user interface ( 1 ) (201)

- Therapist (T) will push the exercise start button (207)

- The person undergoing therapy (H) will take their arm to such spread together with robotic arm (13) as far as they could move it (209) - Events where the person undergoing therapy (H) could not their arm will be perceived by EMG device (120) by means of position, force and muscle activation feedback signals (210)

- Where the person undergoing therapy (H) could not move their arm, hybrid impedance position control mode will be operated by control unit (11 ) (211 )

- Robotic arm (13) will take the arm of the person undergoing therapy (H) to the limit value for the joint motion spread defined through user interface ( 11) before therapy from the moment the position control mode is activated (212) 22. A robot (1) according to any of the claims 9-21 that is characterized by a hybrid impedance controller (115) running in force control mode during isotonic exercise to be applied to the person undergoing therapy (H), a rule base ( 14) where data relating to isotonic resistance levels acquired by adjusting the desired force and control parameters are stored, and a control unit (11) in which the force values to be applied by the robotic arm (13) against the motion of the person undergoing therapy (H) and the level of severity of such value to be applied are entered through a user interface (111).

23. A robot (1) according to claim 22 characterized by a main control unit (130) which performs isotonic exercises to be applied to the person undergoing therapy (H) according to the following steps:

- Personal details pertaining to the person undergoing therapy (H) (i.e. name, surname, sex, age, etc. and height, weight, wrist length, forearm length values) will be defined in the control unit (11) through user interface ( 1 ) (301)

- Therapist (T) will select isotonic exercise mode through user interface (111) (303)

- Therapist (T) will select the type of motion (flexion-extension or abduction- adduction or pronation-supination) through user interface ( 11) (304)

- Therapist (T) will enter in control unit (11) the resistance force value, level of resistance and number of repetitions for the motion relating to the exercise through user interface (11 ) (305)

- Therapist (T) will enter in control unit (11) the details of reaction force and level of muscle activation relating to the person undergoing therapy (H) through user interface ( 11) (306)

- Therapist (T) will push the exercise start button (307)

- Hybrid impedance force control mode will be activated (308)

- The person undergoing therapy (H) will move their arm (309) - Number of repetitions will be checked and if it is completed the exercise will be over (310)

- Robotic arm (13) will stop (311) 24. A robot (1) according to any of the claims 9-23 characterized by a robotic arm (13) which will apply force to the person undergoing therapy (H) as the joint of the person reaches the desired spread of motion once the therapist enters the force value to be applied to the limb of the person undergoing therapy (H) by robotic arm (13) through user interface (111) for the isometric exercise to be applied to the person undergoing therapy (H).

25. A robot (1) as in request 24 characterized by the following for isometric exercise:

- A hybrid impedance controller (115) featuring such input parameters that will allow the person undergoing therapy (H) to get their arm to the joint motion spread the most easily while holding the grip (1332) and move their arm,

- A rule base where the parameter values are stored (114)

- A force sensor ( 334), a position sensor (1306, 319, 1329) and an EMG device (120) that will instantly measure whether the limb of the person undergoing therapy (H) has reached the joint motion spread,

- A main control unit (130) that will check and evaluate the instantaneous values received from a force sensor (1334), a position sensor (1306, 1319, 1329) and an EMG device (120)

- Motors (1304, 1320, 1330) that are connected to a robotic arm ( 3) and generate a torque and apply force to the person undergoing therapy (H) in the opposite direction once the limb of the person undergoing therapy (H) reaches the joint motion spread.

26. A robot (1) according to claim 25 characterized by a main control unit (130) to perform isometric exercises to be applied to the person undergoing therapy (H) according to the following steps:

- Personal details pertaining to the person undergoing therapy (H) (i.e. name, surname, sex, age, etc. and height, weight, wrist length, forearm length values) will be defined in the control unit ( 1) through user interface (1 1) (401)

- Therapist (T) will select the isometric exercise mode through user interface ( 11) (403) - Therapist (T) will select the type of motion (flexion-extension or abduction- adduction or pronation-supination) through user interface (111) (404)

- Therapist (T) will push the exercise start button (407)

- Hybrid impedance force control mode will be activated by control unit (1 ) with the lowest parameter and force value (408)

- The person undergoing therapy (H) will move their arm to the joint motion spread limit (409)

- That the arm of the person undergoing therapy (H) has reached the joint motion spread limit will be perceived by control unit (11) by means of the data received from position sensors (1334) (410)

- A torque value equal to the predefined resistance force value will be applied to the person undergoing therapy (H) by robotic arm (13) (41 )

- The person undergoing therapy (H) will resist the robotic arm (13) in such manner not to allow robotic arm (13) to move their arm (4 2)

- If the person undergoing therapy (H) does not resist, it will be perceived by force sensors (1334) and position sensors (1306, 1319, 1329) (4 3)

- If the number of repetitions is completed, the exercise will be over (415)

- Robotic arm (13) will stop (416)

27. A robot (1) according to claim 26 characterized by a main control unit (130) to perform hybrid exercise algorithm (isotonic plus isometric) (500) where isotonic and isometric exercises are done in the same mode according to the following steps:

- Personal details pertaining to the person undergoing therapy (H) (i.e. name, surname, sex, age, etc. and height, weight, wrist length, forearm length values) will be defined in the control unit (11) through user interface (111) (501)

- The arm of the person undergoing therapy (H) will be placed in robotic arm ( 3) (502) - Therapist (T) will select the hybrid (isotonic plus isometric) exercise mode through user interface (111) (503)

- Therapist (T) will select the type of motion (flexion-extension or abduction- adduction or pronation-supination) through user interface (111) (504)

- Therapist (T) will enter the resistance force value, level of resistance, number of repetitions for the motion and force values to be applied in the joint motion spread relating to the exercise through user interface (111) (505)

- Therapist (T) will enter the data relating to the reaction force and muscle activation level of the person undergoing therapy representing the pain symptom of the person undergoing therapy (H) through user interface (111) (506)

- Therapist (T) will push the exercise start button (507)

- Hybrid impedance force control mode will be run by control unit (11) (508)

- The person undergoing therapy (H) will move their arm (509)

- The person undergoing therapy (H) will beat the predefined resistance level and take their arm up to the joint motion spread limit (510)

- That the arm of the person undergoing therapy (H) has reached the joint motion spread limit will be perceived by control unit (11) by means of the data received from position sensors (1306, 1319, 1329) (51 )

- If the number of repetitions is completed, the exercise will be over (517)

- Robotic arm (13) will stop (518)

28. A robot (1) according to any of the claims 9-27 characterized by the following once the speed value that the robotic arm (13) could not go beyond is entered through user interface (111) for isokinetic exercise to be applied to the person undergoing therapy (H): - A robotic arm (13) applying a torque value equal to the torque value generated by the person undergoing therapy (H) as the person undergoing therapy (H) reaches the predefined speed value,

- A central controller ( 12) calculating the derivative of the position details received from position sensors (1306, 1319, 1329) to obtain the speed of the limb of the person undergoing therapy (H),

- A central controller (112) calculating the torque value generated by the limb of the person undergoing therapy by use of the force data received from force sensor (1334)

29. A robot (1) according to claim 28 characterized by a main control unit (130) which performs isokinetic exercises to be applied to the person undergoing therapy (H) according to the following steps:

- Therapist (T) will select isokinetic exercise mode through user interface (111) (603)

- Therapist (T) will select the type of motion (flexion-extension or abduction- adduction or pronation-supination) through user interface (1 1) (604)

- Therapist (T) will enter the speed value and the number of repetitions for the motion relating to the exercise through user interface (1 1) (605)

- Therapist (T) will push the exercise start button (606)

- The person undergoing therapy (H) will move their arm (608)

- The speed of motion will be controlled by control unit (11) (609)

- If the speed of motion is below the predefined level, hybrid impedance force control will be enabled (61 )

- The number of motions will be checked and if it is completed the step number (607) of the algorithm will be resumed (612) - Upon completion of the number of repetitions the exercise will be over (613)

- Robotic arm (13) will stop (614)

30. A robot (1) according to claim 29 characterized by an EMG device (120) used for the purpose of measurement and evaluation

31. A robot (1) according to claim 30 characterized by an emergency stop button (150) to be used to stop the robot (1) and end the exercise in case of an emergency.