Rehabilitation system and method

Abstract

A rehabilitation system for rehabilitation of a subject including at least one end-effector for interacting with the subject, the end-effector having at least two degrees of freedom of motion, at least one actuator for actuating the at least one end-effector, at least one sensor for measuring at least the position and the speed of the at least one end-effector; at least one sensor for measuring the interaction force between the subject and the end-effector; a memory including at least two initial coefficients and a session including at least one exercise including at least one reference trajectory to be carried out by the subject through actuation of the end effector; and an actuator controlling unit. The memory delivers the initial coefficients and the session, the sensors deliver measurement signals to the controlling unit, and the controlling unit provides a force-controlled feedback based on the initial coefficients.

Claims

1. A rehabilitation system for rehabilitation of a subject comprising: at least one end-effector for interacting with the subject, the said end-effector having at least two degrees of freedom of motion; at least one actuator for actuating the at least one end-effector; at least one sensor for measuring at least the position and the speed of the at least one end-effector; at least one sensor for measuring the interaction force between the subject and the end-effector; a memory comprising at least two initial coefficients k.sub.lat, c.sub.lat and a session comprising at least one exercise including at least one reference trajectory to be carried out by the subject through actuation of the end effector; and a controlling unit for controlling the actuator; wherein the memory is connected to the controlling unit for delivering the initial coefficients and the session, the sensors are connected to the controlling unit for delivering measurement signals, and the controlling unit is adapted to provide a force-controlled feedback comprising a lateral interaction force F.sub.lat:
F.sub.lat=k.sub.lat(x.sub.projx)c.sub.lat{dot over (x)}.sub. wherein said lateral interaction force limits deviation of the end effector actuated by the subject along the reference trajectory and wherein (x.sub.projx) is the deviation of the end-effector actuated by the subject along the reference trajectory, {dot over (x)}.sub. is the speed of the end-effector actuated by the subject perpendicular to the reference trajectory, k.sub.lat is a stiffness coefficient and c.sub.lat is a damping coefficient; wherein the memory further comprises a time interval and wherein the controlling unit is adapted to change online the stiffness coefficient k.sub.lat at the end of each time interval by increments based on a statistical analysis of the values of the deviation of the end-effector actuated by the subject along the reference trajectory.

2. The rehabilitation system according to claim 1, wherein the controlling unit is adapted to change online the damping coefficient c.sub.lat at the end of each time interval by increments based on a statistical analysis of the values of the deviation of the end-effector actuated by the subject along the reference trajectory.

3. The rehabilitation system according to claim 1, wherein said stiffness coefficient k.sub.lat and/or said damping coefficient c.sub.lat varies at the end of each time interval by increments which do not exceed a predefined value.

4. The rehabilitation system according to claim 1, wherein said stiffness coefficient k.sub.lat and/or said damping coefficient c.sub.lat varies at the end of each time interval by increments which do not exceed a predefined value and said increments are either positive, negative or zero.

5. The rehabilitation system according to claim 1, wherein said stiffness coefficient k.sub.lat and/or said damping coefficient c.sub.lat varies at the end of each time interval by increments which do not exceed a predefined value; said increments are either positive, negative or zero; and the positive increments and the negative increments are constant during one exercise or one session.

6. The rehabilitation system according to claim 1, wherein said stiffness coefficient k.sub.lat and/or said damping coefficient c.sub.lat varies at the end of each time interval by increments which do not exceed a predefined value and said increments are either positive, negative or zero; and the absolute value of the positive increment differs from the absolute value of the negative increment.

7. The rehabilitation system according to claim 1, wherein said stiffness coefficient k.sub.lat and/or said damping coefficient c.sub.lat varies at the end of each time interval by increments which do not exceed a predefined value; said increments are either positive, negative or zero; and the positive increments and the negative increments are constant during one exercise or one session; and the absolute value of the positive increment differs from the absolute value of the negative increment.

8. The rehabilitation system according to claim 1, wherein said stiffness coefficient k.sub.lat and/or said damping coefficient c.sub.lat varies at the end of each time interval with a positive increment if the mean of the values of the deviation of the end-effector actuated by the subject along the reference trajectory is higher than a predetermined threshold; and wherein k.sub.lat and/or c.sub.lat varies with a negative increment if the mean of the values of the deviation of the end-effector actuated by the subject along the reference trajectory is lower than the said predetermined threshold.

9. A rehabilitation system for rehabilitation of a subject comprising: at least one end-effector for interacting with the subject, the said end-effector having at least two degrees of freedom of motion; at least one actuator for actuating the at least one end-effector; at least one sensor for measuring at least the position and the speed of the at least one end-effector; at least one sensor for measuring the interaction force between the subject and the end-effector; a memory comprising at least two initial coefficients k.sub.lat, c.sub.lat and a session comprising at least one exercise including at least one reference trajectory to be carried out by the subject through actuation of the end effector; and a controlling unit for controlling the actuator; wherein the memory is connected to the controlling unit for delivering the initial coefficients and the session, the sensors are connected to the controlling unit for delivering measurement signals, and the controlling unit is adapted to provide a force-controlled feedback comprising a lateral interaction force F.sub.lat:
F.sub.lat=k.sub.lat(x.sub.projx)c.sub.lat{dot over (x)}.sub. wherein said lateral interaction force limits deviation of the end effector actuated by the subject along the reference trajectory and wherein (x.sub.projx) is the deviation of the end-effector actuated by the subject along the reference trajectory, {dot over (x)}.sub. is the speed of the end-effector actuated by the subject perpendicular to the reference trajectory, k.sub.lat is a stiffness coefficient and c.sub.lat is a damping coefficient; wherein the memory further comprises a time interval and wherein the controlling unit is adapted to change online the stiffness coefficient k.sub.lat at the end of each time interval by increments based on a statistical analysis of the values of the deviation of the end-effector actuated by the subject along the reference trajectory and wherein the memory further comprises two initial coefficients c.sub.long, v.sub.ref and a second time interval; and wherein the force-controlled feedback provided by the controlling unit further comprises a longitudinal interaction force F.sub.long: $F_{\mathrm{long}}=c_{\mathrm{long}}\left(v_{\mathrm{ref}}-.Math.{\stackrel{.}{x}}_{.Math.}.Math.\right)\frac{{\stackrel{.}{x}}_{.Math.}}{.Math.{\stackrel{.}{x}}_{.Math.}.Math.}$ wherein said longitudinal interaction force helps the subject to move along the reference trajectory at a reference speed and wherein (v.sub.ref|{dot over (x)}.sub.|) is the difference between the reference speed (v.sub.ref) and the speed of the end-effector actuated by the subject parallel to the reference trajectory, {dot over (x)}.sub. is the speed of the end-effector actuated by the subject parallel to the reference trajectory, c.sub.long is a damping coefficient and v.sub.ref is a reference speed; and wherein the controlling unit is adapted to change online the damping coefficient c.sub.long at the end of each second time interval by increments based on the difference between the longitudinal speed of the end-effector actuated by the subject parallel to the reference trajectory and the reference speed.

10. The rehabilitation system according to claim 9, wherein the controlling unit is adapted to change online the reference speed v.sub.ref at the end of each second time interval by increments based on the difference between the longitudinal speed of the end-effector actuated by the subject parallel to the reference trajectory and the reference speed.

11. The rehabilitation system according to claim 9, wherein said damping coefficient c.sub.long and/or the said reference speed v.sub.ref varies at the end of each such second time interval by increments which do not exceed a predefined value.

12. The rehabilitation system according to claim 9, wherein said increments of c.sub.long and/or v.sub.ref are either positive, negative or zero.

13. The rehabilitation system according to claim 9, wherein said increments of c.sub.long and/or v.sub.ref are either positive, negative or zero and the positive increments and the negative increments are constant during one exercise or one session.

14. The rehabilitation system according to claim 9, wherein said increments of c.sub.long and/or v.sub.ref are either positive, negative or zero and the positive increments and the absolute value of the positive increment differs from the absolute value of the negative increment.

15. The rehabilitation system according to claim 9, wherein c.sub.long and/or v.sub.ref varies with a positive increment if the mean of the differences between the reference speed v.sub.ref and the speed of the end-effector actuated by the subject is higher than a predetermined threshold; and wherein c.sub.long and/or v.sub.ref varies with a negative increment if the mean of the differences between the reference speed v.sub.ref and the speed of the end-effector actuated by the subject is lower or equal than said predetermined threshold.

16. A rehabilitation method comprising the following steps: providing a robotic device comprising: at least one end-effector for interacting with a subject, the said end-effector having at least two degrees of freedom of motion; at least one actuator for moving the at least one end-effector; at least one sensor for recording at least the position and the speed of the at least one end-effector; at least one sensor for recording the interaction force between the subject and the end-effector; and a controlling unit for controlling the actuator; initiating a session comprising at least one exercise, including at least one reference trajectory to be carried out by the subject through actuation of the end effector; said initiation comprising initializing at least two coefficients k.sub.lat, c.sub.lat; the controlling unit providing a force-controlled feedback comprising a lateral interaction force F.sub.lat:
F.sub.lat=k.sub.lat()(x.sub.projx)c.sub.lat(){dot over (x)}.sub. wherein (x.sub.projx) is the deviation of the end-effector actuated by the subject along the reference trajectory, {dot over (x)}.sub. is the speed of the end-effector actuated by the subject perpendicular to the reference trajectory, k.sub.lat is a stiffness coefficient and c.sub.lat is a damping coefficient; said lateral interaction force limiting deviation of the end effector actuated by the subject along the reference trajectory; wherein k.sub.lat is a variable stiffness coefficient and c.sub.lat is a damping coefficient depending on , wherein is selected from a statistical analysis of the values of the deviation of the end-effector actuated by the subject along the reference trajectory; the initiation comprises initializing a time interval; and wherein said variable stiffness coefficient k.sub.lat varies online at the end of each time interval by increments.

17. The rehabilitation method according to claim 16, wherein: said initiation further comprises initializing one coefficients (c.sub.long), a reference speed (v.sub.ref) and providing a second time interval; the force-controlled feedback provided by the controlling unit further comprises a longitudinal interaction force F.sub.long: $F_{\mathrm{long}}=c_{\mathrm{long}}\left(\right)\left(v_{\mathrm{ref}}\left(\right)-.Math.{\stackrel{.}{x}}_{.Math.}.Math.\right)\frac{{\stackrel{.}{x}}_{.Math.}}{.Math.{\stackrel{.}{x}}_{.Math.}.Math.}$ wherein (v.sub.ref|{dot over (x)}.sub.|) is the difference between the reference speed (v.sub.ref) and the speed of the end-effector actuated by the subject parallel to the reference trajectory; {dot over (x)}.sub. is the speed of the end-effector actuated by the subject parallel to the reference trajectory and c.sub.long is a damping coefficient; said longitudinal interaction force helps the subject to move along the reference trajectory at a reference speed; wherein c.sub.long is a variable damping coefficient and v.sub.ref is a reference speed depending on , wherein is selected from the difference between the longitudinal speed of the end-effector actuated by the subject parallel to the reference trajectory and the reference speed; and wherein said variable damping coefficient c.sub.long varies online at the end of each second time interval by increments.

Description

DETAILED DESCRIPTION

(1) This invention relates to a rehabilitation system providing an as-needed assistance to a subject in need thereof.

(2) The said rehabilitation system for rehabilitation of a subject comprises: at least one end-effector for interacting with the subject, the said end-effector having at least two degrees of freedom of motion; at least one actuator for actuating the at least one end-effector; at least one sensor for measuring at least the position and the speed of the at least one end-effector; at least one sensor for measuring the interaction force between the subject and the end-effector; a memory comprising at least two initial coefficients and a session comprising at least one exercise including at least one reference trajectory to be carried out by the subject through actuation of the end effector; and a controlling unit for controlling the actuator;
wherein the memory is connected to the controlling unit for delivering the initial coefficients and the session; wherein the sensors are connected to the controlling unit for delivering measurement signals and wherein the controlling unit is adapted to provide a force-controlled feedback based on the said initial coefficients.

(3) In order to ensure rehabilitation, the subject has to perform a movement along the reference trajectory corresponding to the ideal path that the patient must follow to perform the exercise. In order to improve the rehabilitation, the controlling unit of the present invention generates forces smoothly assisting the movement of a limb of a subject.

(4) According to one embodiment, the reference trajectory is at least a two dimensional trajectory. According to one exemplary embodiment, as depicted in FIG. 1, the reference trajectory is a two dimensional trajectory.

(5) According to the embodiment depicted in FIG. 1, the at least one sensor of the rehabilitation system acquires at least the position of the end effector x, the speed of the end-effector actuated by the subject {dot over (x)}, especially the speed of the end-effector actuated by the subject parallel to the reference trajectory {dot over (x)}.sub. and the speed of the end-effector actuated by the subject perpendicular to the reference trajectory {dot over (x)}.sub..

(6) According to one embodiment, the at least one sensor of the rehabilitation system acquires at least the lateral interaction force F.sub.lat and longitudinal interaction force F.sub.long between the patient and the end-effector.

(7) According to one embodiment, in order to provide a force-controlled feedback, the controlling unit is adapted to modify at least a lateral interaction force F.sub.lat that helps the patient to stay on the reference trajectory. Especially, the lateral interaction force limits deviation of the end effector actuated by the subject along the reference trajectory. If the lateral interaction force is strictly positive, an assisting attractive force towards the reference trajectory is applied on the end-effector (see FIG. 1). On the contrary, if the lateral interaction force is strictly negative, a resistive force away from the reference trajectory is applied on the end-effector, thereby challenging the subject. According to one embodiment, the lateral interaction force is strictly positive. According to one embodiment, the lateral interaction force is positive and negative.

(8) According to one embodiment, the force-controlled feedback comprises a lateral interaction force F.sub.lat correlated to the deviation of the end-effector actuated by the subject along the reference trajectory according to the equation:
F.sub.lat=k.sub.lat(x.sub.projx)c.sub.lat{dot over (x)}.sub.
wherein (x.sub.projx) is the deviation of the end-effector actuated by the subject along the reference trajectory, {dot over (x)}.sub. is the speed of the end-effector actuated by the subject perpendicular to the reference trajectory, k.sub.lat is a stiffness coefficient and c.sub.lat is a damping coefficient.

(9) According to one embodiment, the controlling unit comprises a processor adapted to compute, from the measurement of the sensors, at least the deviation of the end-effector actuated by the subject along the reference trajectory (x.sub.projx).

(10) According to one embodiment, the memory comprises at least the two initial coefficients c.sub.lat and k.sub.lat. According to one embodiment c.sub.lat, is positive, negative or zero. According to one embodiment, k.sub.lat is nonzero.

(11) According to one embodiment, the controlling unit is adapted to change online the stiffness coefficient k.sub.lat and/or the damping coefficient c.sub.lat during each reference trajectory, at the end of each reference trajectory or at the end of each exercise. Especially, according to one embodiment, the memory comprises a first time interval and the controlling unit is adapted to change the stiffness coefficient k.sub.lat and or the damping coefficient c.sub.lat at the end of each such first time interval by increments.

(12) According to one embodiment, by online it is meant that the first time interval is ranging from 1 ms to 10 s, from 5 ms to 5 s, from 100 ms to 2 s, preferably about 1 second.

(13) According to one embodiment, the first time interval is lower than the duration required by a healthy subject (not impaired) to perform a reference trajectory during one exercise. Preferably, the first time interval is lower than 1/10, or lower than 1/100, of the duration required by a healthy subject (not impaired) to perform a reference trajectory during one exercise.

(14) According to one embodiment, the controlling unit is adapted to change the stiffness coefficient k.sub.lat and/or the damping coefficient c.sub.lat based on a statistical analysis of the values of the deviation of the end-effector actuated by the subject along the reference trajectory. According to one embodiment, said statistical analysis is based on a statistical descriptor such as the means, median, mode, variance or standard deviation.

(15) According to an embodiment, the controlling unit is adapted to change the stiffness coefficient k.sub.lat and/or the damping coefficient c.sub.lat based on the ratio between the length covered by the end-effector and the length along the reference trajectory.

(16) Therefore, according to one embodiment, the lateral interaction force F.sub.lat of the force-controlled feedback is correlated to the deviation of the end-effector actuated by the subject along the reference trajectory according to the equation:
F.sub.lat=k.sub.lat()(x.sub.projx)c.sub.lat(){dot over (x)}.sub.
wherein (x.sub.projx) is the deviation of the end-effector actuated by the subject along the reference trajectory, {dot over (x)}.sub. is the speed of the end-effector actuated by the subject perpendicular to the reference trajectory, k.sub.lat is a stiffness coefficient, c.sub.lat is a damping coefficient and a is a statistical descriptor of the values of the deviation of the end-effector actuated by the subject along the reference trajectory.

(17) According to one embodiment, the increments of k.sub.lat and/or c.sub.lat do not exceed a predefined value. According to one embodiment, the increments of k.sub.lat and/or c.sub.lat are positive, negative or zero. According to one embodiment, the increments of k.sub.lat and/or c.sub.lat are constant during each movement along a reference trajectory. According to one embodiment, the increments of k.sub.lat and/or c.sub.lat are constant during one reference trajectory of an exercise. According to one embodiment, the increments of k.sub.lat and/or c.sub.lat are constant during one exercise. According to one embodiment, the increments of k.sub.lat and/or c.sub.lat are constant during one session.

(18) According to one embodiment, the absolute value of the positive increments of k.sub.lat and/or c.sub.lat differs from the absolute value of the negative increment of respectively k.sub.lat and/or c.sub.lat. According to one embodiment, the increments of k.sub.lat and/or c.sub.lat are constant during each exercise or may vary between each reference trajectory of an exercise or between each exercise of a session.

(19) According to one embodiment, the positive increments and the negative increments of k.sub.lat and/or c.sub.lat are constant during one reference trajectory of an exercise. According to one embodiment, the positive increments and the negative increments of k.sub.lat and/or c.sub.lat are constant during one exercise. According to one embodiment, the positive increments and the negative increments of k.sub.lat and/or c.sub.lat are constant during one session.

(20) It has been found by the Applicant, that constant increments ensure an improved degree of smoothness of the movement and thus an optimal rehabilitation. According to one embodiment, whatever the deviation, the increments is of k.sub.lat and/or c.sub.lat remain constant during a reference trajectory, an exercise and/or a session.

(21) According to one embodiment, k.sub.lat and/or c.sub.lat varies with a positive increment if the mean of the values of the deviation of the end-effector actuated by the subject along the reference trajectory is higher than a threshold, such as for instance 0.01 m for k.sub.lat. According to one embodiment, k.sub.lat and/or c.sub.lat varies with a negative increment if the mean of the values of the deviation of the end-effector actuated by the subject along the reference trajectory is lower than a threshold, such as for instance 0.01 m for k.sub.lat. According to one embodiment, the memory further comprises the said thresholds. According to one embodiment, the said threshold may be adjusted automatically by the controlling unit or manually by the physiotherapist.

(22) Thus according to an exemplary embodiment, the deviation of the end-effector actuated by the subject along the reference trajectory (x.sub.projx) is acquired at a frequency of 125 Hz and each second (the time interval) means of said 125 values is computed. Based on said means, k.sub.lat and/or c.sub.lat is incremented either positively or negatively.

(23) According to one embodiment, k.sub.lat increases with an increment ranging from 1 N/m to 500 N/m, preferably 157.5 N/m.

(24) According to one embodiment, k.sub.lat decreases with an increment ranging from 1 N/m to 500 N/m, preferably 157.5 N/m.

(25) According to one embodiment, k.sub.lat has an absolute value ranging from ON/m to 5000 N/m, preferably from ON/m to 3150 N/m.

(26) According to one embodiment, k.sub.lat has an absolute value ranging from 1 N/m to 5000 N/m, preferably from 1 N/m to 3150 N/m.

(27) According to one embodiment, c.sub.lat increases with an increment ranging from 1 N.s/m to 250 N.s/m, preferably 78.75 N.s/m.

(28) According to one embodiment, c.sub.lat decreases with an increment ranging from 1 N.s/m to 250 N.s/m, preferably 78.75 N.s/m.

(29) According to one embodiment, c.sub.lat has an absolute value ranging from 0 N.s/m to 2500 N.s/m, preferably from 0 N.s/m to 1575 N.s/m.

(30) According to one embodiment, k.sub.lat is incremented at a frequency which can be adjusted, such as for instance 0.1 Hz, 0.5 Hz, 1 Hz, 10 Hz, 100 Hz or 1000 Hz.

(31) According to one embodiment, c.sub.lat is incremented at a frequency which can be adjusted, such as for instance 0.1 Hz, 0.5 Hz, 1 Hz, 10 Hz, 100 Hz or 1000 Hz.

(32) According to one embodiment, k.sub.lat is not inferior to ON/m. According to an alternative embodiment, if a resistive lateral interaction force is sought, k.sub.lat may decrease below 0 N/m.

(33) According to one embodiment, in order to provide a force-controlled feedback, the controlling unit is also adapted to modify at least a longitudinal interaction force F.sub.long that helps the patient to move along the trajectory at a reference speed. If the longitudinal interaction force is strictly positive, an assisting force helps the subject to move along the reference trajectory at a reference speed (see FIG. 1). On the contrary, if the longitudinal interaction force is strictly negative, a resistive force slows down movement of the subject along the reference trajectory, thereby challenging the subject. According to one embodiment, the longitudinal interaction force is strictly positive. According to one embodiment, the longitudinal interaction force is positive and negative.

(34) According to one embodiment, the processor is also adapted to compute, from the measurement of the sensors, the difference between the reference speed v.sub.ref and the speed of the end-effector actuated by the subject parallel to the reference trajectory {dot over (x)}.sub..

(35) According to one embodiment, the force-controlled feedback comprises a longitudinal interaction force F.sub.long correlated to difference between the reference speed and the speed of the end-effector actuated by the subject parallel to the reference trajectory according to the equation:

(36) $F_{\mathrm{long}}=c_{\mathrm{long}}\left(v_{\mathrm{ref}}-.Math.{\stackrel{.}{x}}_{.Math.}.Math.\right)\frac{{\stackrel{.}{x}}_{.Math.}}{.Math.{\stackrel{.}{x}}_{.Math.}.Math.}$
wherein (v.sub.ref|{dot over (x)}.sub.|) is the difference between the reference speed (v.sub.ref) and the speed of the end-effector actuated by the subject parallel to the reference trajectory; {dot over (x)}.sub. is the speed of the end-effector actuated by the subject parallel to the reference trajectory and c.sub.long is a damping coefficient.

(37) According to one embodiment, the memory comprises at least the reference speed v.sub.ref and the damping coefficient c.sub.long According to one embodiment, the damping coefficient c.sub.long is positive, negative or zero.

(38) According to one embodiment, the controlling unit is adapted to change online the damping coefficient c.sub.long and the reference speed v.sub.ref during each reference trajectory. Especially, according to one embodiment, the memory comprises a second time interval and the controlling unit is adapted to change the damping coefficient c.sub.long and the reference speed v.sub.ref at the end of each such second time interval by increments. According to one embodiment, the first time interval and the second time interval are the same or not.

(39) According to one embodiment, by online it is meant that the second time interval is ranging from 1 ms to 10 s, from 5 ms to 5 s, from 100 ms to 2 s, preferably about 1 second.

(40) According to one embodiment, the second time interval is lower than the duration required by a healthy subject (not impaired) to perform a reference trajectory during an exercise. Preferably, the second time interval is lower than 1/10, or lower than 1/100 of the duration required by a healthy subject (not impaired) to perform a reference trajectory during an exercise.

(41) According to one embodiment, the controlling unit is adapted to change the damping coefficient c.sub.long and the reference speed v.sub.ref based on the difference between the longitudinal speed of the end-effector actuated by the subject parallel to the reference trajectory and the reference speed.

(42) According to an alternative embodiment, the controlling unit is adapted to change the damping coefficient c.sub.long and the reference speed v.sub.ref based on the ratio between the reference speed and the speed of the end-effector actuated by the subject parallel to the reference trajectory.

(43) Therefore, according to one embodiment, the longitudinal interaction force F.sub.long of the force-controlled feedback is correlated to difference between the reference speed and the speed of the end-effector actuated by the subject parallel to the reference trajectory according to the equation:

(44) $F_{\mathrm{long}}=c_{\mathrm{long}}\left(\right)\left(v_{\mathrm{ref}}\left(\right)-.Math.{\stackrel{.}{x}}_{.Math.}.Math.\right)\frac{{\stackrel{.}{x}}_{.Math.}}{.Math.{\stackrel{.}{x}}_{.Math.}.Math.}$
wherein (v.sub.ref|{dot over (x)}.sub.|) is the difference between the reference speed (v.sub.ref) and the speed of the end-effector actuated by the subject parallel to the reference trajectory; {dot over (x)}.sub. is the speed of the end-effector actuated by the subject parallel to the reference trajectory, c.sub.long is a damping coefficient and is a parameter such as the difference between the longitudinal speed of the end-effector actuated by the subject parallel to the reference trajectory and the reference speed.

(45) According to one embodiment, the increments of c.sub.long and/or v.sub.ref do not exceed a predefined value. According to one embodiment, the increments of c.sub.long and/or v.sub.ref are positive, negative or zero. According to one embodiment, the increments of c.sub.long and/or v.sub.ref are constant during one movement along a reference trajectory. According to one embodiment, the increments of c.sub.long and/or v.sub.ref are constant during one reference trajectory of an exercise. According to one embodiment, the increments of c.sub.long and/or v.sub.ref are constant during one exercise. According to one embodiment, the increments of c.sub.long and/or v.sub.ref are constant during one session.

(46) According to one embodiment, the absolute value of the positive increments of c.sub.long and/or v.sub.ref differs from the absolute value of the negative increment of c.sub.long and/or v.sub.ref.

(47) According to one embodiment, the positive increments and the negative increments of c.sub.long and/or v.sub.ref are constant during one reference trajectory of an exercise. According to one embodiment, the positive increments and the negative increments of c.sub.long and/or v.sub.ref are constant during one exercise. According to one embodiment, the positive increments and the negative increments of c.sub.long and/or v.sub.ref are constant during one session.

(48) It has been found by the Applicant, that constant increments ensure an improved degree of smoothness of the movement and thus an optimal rehabilitation. According to one embodiment, whatever the speed of the end-effector actuated by the subject parallel to the reference trajectory, the increments of c.sub.long and/or v.sub.ref remain constant during a reference trajectory, an exercise and/or a session.

(49) According to one embodiment, c.sub.long and/or v.sub.ref varies with a positive increment if the differences between the reference speed v.sub.ref and the speed of the end-effector actuated by the subject is higher than a threshold, such as for instance 0 m/s. According to one embodiment, c.sub.long and/or v.sub.ref varies with a negative increment if the differences between the reference speed v.sub.ref and the speed of the end-effector actuated by the subject is lower or equal than a threshold, such as for instance 0 m/s. According to one embodiment, the memory further comprises the said thresholds. According to one embodiment, the said threshold may be adjusted automatically by the controlling unit or manually by the physiotherapist.

(50) According to one embodiment, c.sub.long increases with an increment ranging from 1 N.s/m to 250 N.s/m, preferably 78.75 N.s/m.

(51) According to one embodiment, c.sub.long decreases with an increment ranging from 1 N.s/m to 250 N.s/m, preferably 78.75 N.s/m.

(52) According to one embodiment, c.sub.long has an absolute value ranging from 0 N.s/m to 2500 N.s/m, preferably from 0 N.s/m to 1575 N.s/m.

(53) According to one embodiment, c.sub.long is incremented at a frequency which can be adjusted, such as for instance 0.1 Hz, 0.5 Hz, 1 Hz, 10 Hz, 100 Hzz or 1000 Hz.

(54) According to one embodiment, c.sub.long is not inferior to 0 N.s/m. According to an alternative embodiment, if a resistive longitudinal interaction force is sought, then c.sub.long varies with a positive increment if the differences between the reference speed v.sub.ref and the speed of the end-effector actuated by the subject lower and equal to 0; and c.sub.long varies with a negative increment if the differences between the reference speed v.sub.ref and the speed of the end-effector actuated by the subject is above 0.

(55) According to one embodiment, v.sub.ref increases with an increment ranging from 0.001 m/s to 0.05 m/s, preferably 0.01 m/s.

(56) According to one embodiment, v.sub.ref decreases with an increment ranging from 0.001 m/s to 0.05 m/s, preferably 0.01 m/s.

(57) According to one embodiment, v.sub.ref has an absolute value ranging from 0 m/s to 1 m/s, from 0 m/s to 0.5 m/s, preferably 0.1 m/s.

(58) According to one embodiment, v.sub.ref is incremented at a frequency which can be adjusted, such as for instance 0.1 Hz, 0.5 Hz, 1 Hz, 10 Hz, 100 Hz or 1000 Hz.

(59) According to one embodiment, the reference trajectory comprises simple discrete trajectories consisting in reaching a target in the most precise and direct manner, complex discrete trajectories consisting in reaching target through a curved reference trajectory or cyclic simple or complex trajectories consisting in performing a series of simple discrete exercises and/or complex discrete exercises (e.g. point-to-point moves).

(60) According to one embodiment, at the beginning of each exercise along a reference trajectory, the control unit is temporarily disabled during a timet.sub.init.

(61) According to one embodiment, t.sub.init is ranging from 0.5 s to 5 s, from 0.5 s to 2 s, preferably about 1 s.

(62) According to one embodiment, if the subject initiates the movement before t.sub.init, t.sub.init decreases for challenging the subject. According to one embodiment, if the subject initiates the movement after t.sub.init, t.sub.init increases.

(63) According to one embodiment, the at least one end-effector comprises a handle that can be grasp by the subject or to which the subject may be attached by an orthosis.

(64) According to one embodiment, the acquisition frequency of the sensors is ranging from 50 to 1000 HZ, preferably about 250 Hz. According to one embodiment, the at least one sensor is connected to the controlling unit.

(65) According to one embodiment, as depicted in FIG. 2, the controlling unit comprises a virtual system and a force controller.

(66) According to one embodiment, the controlling unit controls the at least one actuator and thereby the end-effector. According to one embodiment, the controlling unit controls the motion of the end-effector and is connected to the at least one actuator. According to one embodiment, said controlling unit is parameterized by a computer intended to be used by the therapist. According to one embodiment, at it is well known from those skilled in the art, the controlling unit comprises memory for storing data and a processor for calculating parameters changes depending on the subject-device interaction and the therapist instructions.

(67) According to one embodiment, the rehabilitation system enables to assist the movement of a limb of a subject manipulating the end-effector, especially an upper limb of a subject.

(68) According to one embodiment, the rehabilitation system according to the present invention also comprises a stimulation generator, such as an auditory generator or a visual generator delivering sensory stimulation to the subject based on at least one predetermined condition.

(69) According to one embodiment, the predetermined condition is selected from or .

(70) According to one embodiment, the predetermined condition is selected from a statistical analysis of the values of the deviation of the end-effector actuated by the subject along the reference trajectory. According to one embodiment, said statistical analysis is based on a statistical descriptor such as the means, median, mode, variance or standard deviation. According to one embodiment, the predetermined condition is selected from the difference between the longitudinal speed of the end-effector actuated by the subject parallel to the reference trajectory and the reference speed.

(71) According to one embodiment, the rehabilitation system according to the present invention further comprises a display, such as a screen or a head-mounted display. According to one embodiment, the rehabilitation system comprises a first display for controlling the controlling unit by the physiotherapist (Therapist GUI), and a second display for immersion of the subject in the therapy (Patient GUI). According to one embodiment, thanks to the second display, a graphical subject interface (Patient GUI) creates a user-friendly environment contextualized as task or game. According to one embodiment, the controlling unit uses input data from the sensors to drive an icon, cursor or other figure graphically on the display. Interactive games, such as maze games, controlled by the controlling may be used to ensure immersion of the subject in the therapy.

(72) According to one embodiment, the reference trajectory and the movement of the end-effector actuated by the subject are displayed.

(73) This invention also relates to a robotic device-assisted rehabilitation method providing an as-needed assistance to a subject in need thereof.

(74) In particular, the robotic device-assisted rehabilitation method comprises the following steps: providing a robotic device comprising: at least one end-effector for interacting with a subject; at least one actuator for moving the at least one end-effector; at least one sensor for measuring the position and the speed of the at least one end-effector; at least one sensor for measuring the interaction force between the subject and the end-effector; and a controlling unit for controlling the actuator; initiating a session comprising at least one exercise, including at least one reference trajectory, to be carried out by the subject through actuation of the end effector; said initiation comprising initializing at least two coefficients; providing by the controlling unit a force-controlled feedback based on the said coefficients.

(75) According to one embodiment, the force-controlled feedback comprises a lateral interaction force F.sub.lat correlated to the deviation of the end-effector actuated by the subject along the reference trajectory according to the equation:
F.sub.lat=k.sub.lat()(x.sub.projx)c.sub.lat(){dot over (x)}.sub.
wherein (x.sub.projx) is the deviation of the end-effector actuated by the subject along the reference trajectory, {dot over (x)}.sub. is the speed of the end-effector actuated by the subject perpendicular to the reference trajectory, k.sub.lat is a stiffness coefficient and c.sub.lat is a damping coefficient.

(76) According to one embodiment, k.sub.lat and c.sub.lat are initialized at the beginning of the session.

(77) According to the present invention, k.sub.lat is a variable stiffness coefficient depending on , wherein is selected from a statistical analysis of the values of the deviation of the end-effector actuated by the subject along the reference trajectory. According to one embodiment, said statistical analysis is based on a statistical descriptor such as the means, median, mode, variance or standard deviation.

(78) According to one embodiment, c.sub.lat is a variable damping coefficient depending on , wherein is selected from a statistical analysis of the values of the deviation of the end-effector actuated by the subject along the reference trajectory. According to one embodiment, said statistical analysis is based on a statistical descriptor such as the means, median, mode, variance or standard deviation.

(79) According to the present invention, the step of initiating a session further comprises the step of providing a first time interval. According to one embodiment, the variable coefficient k.sub.lat and/or and c.sub.lat varies online at the end of each first time interval by increments.

(80) According to one embodiment, the increments of k.sub.lat and/or c.sub.lat are constant during one movement along a reference trajectory. According to one embodiment, the increments of k.sub.lat and/or c.sub.lat are constant during one reference trajectory of an exercise. According to one embodiment, the increments of k.sub.lat and/or c.sub.lat are constant during one exercise. According to one embodiment, the increments of k.sub.lat and/or c.sub.lat are constant during one session.

(81) According to one embodiment, the force-controlled feedback comprises a longitudinal interaction force F.sub.long correlated to difference between the reference speed and the speed of the end-effector actuated by the subject parallel to the reference trajectory according to the equation:

(82) $F_{\mathrm{long}}=c_{\mathrm{long}}\left(\right)\left(v_{\mathrm{ref}}\left(\right)-.Math.{\stackrel{.}{x}}_{.Math.}.Math.\right)\frac{{\stackrel{.}{x}}_{.Math.}}{.Math.{\stackrel{.}{x}}_{.Math.}.Math.}$
wherein (v.sub.ref|{dot over (x)}.sub.|) is the difference between the reference speed (v.sub.ref) and the speed of the end-effector actuated by the subject parallel to the reference trajectory; {dot over (x)}.sub. is the speed of the end-effector actuated by the subject parallel to the reference trajectory and c.sub.long is a damping coefficient.

(83) According to one embodiment, c.sub.long and v.sub.ref are initialized at the beginning of the session.

(84) According to one embodiment, c.sub.long is a variable damping coefficient depending on , wherein is selected from the difference between the longitudinal speed of the end-effector actuated by the subject parallel to the reference trajectory and the reference speed, the ratio between the reference speed and the speed of the end-effector actuated by the subject parallel to the reference trajectory or combination thereof.

(85) According to one embodiment, v.sub.ref is a variable damping coefficient depending on , wherein is selected from the difference between the longitudinal speed of the end-effector actuated by the subject parallel to the reference trajectory and the reference speed, the ratio between the reference speed and the speed of the end-effector actuated by the subject parallel to the reference trajectory or combination thereof.

(86) According to one embodiment, a second time interval is provided at the beginning of the session. The variable damping coefficient c.sub.long and/or v.sub.ref varies online at the end of each second time interval by increments. According to one embodiment, the increments of c.sub.long and/or v.sub.ref are constant during one movement along a reference trajectory. According to one embodiment, the increments of c.sub.long and/or v.sub.ref are constant during one reference trajectory of an exercise. According to one embodiment, the increments of c.sub.long and/or v.sub.ref are constant during one exercise. According to one embodiment, the increments of c.sub.long and/or v.sub.ref are constant during one session.

(87) According to one embodiment, the rehabilitation method according to the present invention also comprises the step of providing a sensory signal to the subject based on at least one predetermined condition. In particular, the sensory signal may be a visual signal, an auditory signal or an haptic signal.

(88) According to one embodiment, the predetermined condition is selected from or .

(89) According to one embodiment, the rehabilitation method according to the present invention also comprises the step of providing a display, such as a screen or a head-mounted display. According to one embodiment, the rehabilitation method comprises the step of providing a first display for controlling the controlling unit by the physiotherapist (Therapist GUI), and a second display for immersion of the subject in the therapy (Patient GUI). According to one embodiment, thanks to the second display, a graphical subject interface (Patient GUI) creates a user-friendly environment contextualized as task or game. According to one embodiment, the controlling unit uses input data from the sensors to drive an icon, cursor or other figure graphically on the display. Interactive games, such as maze games, controlled by the controlling may be used to ensure immersion of the subject in the therapy.

(90) According to one embodiment, the rehabilitation method comprises the step of displaying the movement of the subject together with a serious game corresponding to the exercise.

BRIEF DESCRIPTION OF THE DRAWINGS

(91) FIG. 1 illustrates a reference trajectory, the position of the end effector x, the speed of the end-effector {dot over (x)}, the deviation of the end-effector actuated by the subject along the reference trajectory (x.sub.projx), the speed of the end-effector actuated by the subject parallel to the reference trajectory {dot over (x)}.sub., an assisting attractive lateral interaction force F.sub.lat and an assisting longitudinal interaction force F.sub.long according to one embodiment of the invention.

(92) FIG. 2 illustrates a flow chart of the rehabilitation method according to one embodiment of the invention. The controlling unit according to the invention comprises a virtual system determining how the device reacts to the interaction with the subject and a force controller controlling the device.