EP2991737A1

EP2991737A1 - Control of an exercise machine

Info

Publication number: EP2991737A1
Application number: EP14722267.3A
Authority: EP
Inventors: Philippe CHAZALON; Arnaud VANNICATTE; Aurélien VAUQUELIN
Original assignee: Eracles - Technology
Current assignee: Eracles - Technology
Priority date: 2013-04-29
Filing date: 2014-04-11
Publication date: 2016-03-09
Anticipated expiration: 2034-04-11
Also published as: AU2014261278A1; CA2910923A1; FR3004961A1; WO2014177787A1; FR3004961B1; AU2014261278B2; US20160151675A1; CN105283227B; EP2991737B1; CN105283227A; CA2910923C; US10434368B2

Abstract

A method for controlling an electrical actuator in an exercise device, comprising: providing a first load value (FA, kA) during a movement of the biasing member in a first direction, providing a second load value (FB, kB) during a movement of the biasing member in a second direction opposite the first direction, and detecting an initial position (M) of the mobile portion of the electrical actuator or of the biasing member at the time when the reversal of movement is detected, calculating a transition end position (N) having a variation in the second direction relative to the initial position, providing a transition load value in the form of a monotonic function of the position of the mobile portion of the electrical actuator or of the biasing member, said monotonic function varying from the first load value (FA, kA) to the second load value (FB, kB) between the initial position (M) and the transition end position (N).

Description

Control of an exercise machine

The invention relates to the field of exercise machines. More particularly, the invention relates to the field of electrically powered machines designed to develop or reconstruct the musculature of a user and in particular for sports training or rehabilitation of the muscles of a user.

Muscular exercise machines include weight machines and inertia machines.

The weight machines operate on the principle of cast iron masses or other material that a user moves by providing an effort to counter the weight of the cast iron masses. These machines include presses, free bars, guided load devices etc.

Inertia machines work differently. These consist, for example, in setting a cast iron disk in motion about an axis of rotation. The user must therefore provide an adequate effort to overcome the inertia of the machine. Some machines work with the principle of moving a fluid with a fin system. Although the fluid in motion has an inertia, in these machines the user must overcome mainly the viscous friction induced by the fluids. Other machines use the principle of the eddy current system to generate these viscous friction. These machines producing viscous friction include rowing machines or indoor cycling. There are also dry rubbing machines. Thus, some exercise bikes have a strap rotating on a wheel of inertia with a dry friction.

EP-A1-2255851 discloses a muscle training apparatus adapted to apply a load on a user by means of the motor torque of an electric actuator. It comprises means for detecting the speed and a characteristic curve of the load applied as a function of the speed. In one embodiment shown in FIG. 7, two different isotonic charges are applied, on the one hand in a direction of concentric movement at a speed greater than a first speed threshold and, on the other hand, in an eccentric direction of movement at a speed greater than a second speed threshold. The transition between the two isotonic charges is performed according to an affine function of the speed of displacement. Due to the increase in the applied load proportionally to the speed detected, the user's movement does not necessarily cross the set speed thresholds, so that the programmed isotonic load is not necessarily applied to the course of movement. According to one embodiment, the invention provides a control method for controlling an electric actuator in an exercise device having a biasing member to be moved by the force of a user and coupled to a movable portion of the electric actuator, the control method comprising:

provide a first load setpoint during a displacement of the biasing element in a first direction,

providing a second load setpoint during a displacement of the biasing element in a second opposite direction to the first direction, and

in response to detecting an inversion of the displacement of the biasing element between the first direction and the second direction, providing a transition load setpoint varying progressively from the first load setpoint to the second load setpoint at during a time interval.

According to one embodiment, the method further comprises:

detecting an initial position of the moving part of the electric actuator or the biasing element when the motion reversal is detected,

calculating a transition end position having a deviation in the second direction from the initial position,

supplying the transition charge setpoint in the form of a monotonic function of the position of the moving part of the electric actuator or the biasing element, said monotonic function varying from the first charge setpoint to the second load setpoint between the initial position and the end of transition position.

According to one embodiment, the transition charge setpoint varies with a rate of change per unit of constant displacement from the first charge setpoint to the second charge setpoint, the monotone function being an affine function. According to alternative embodiments, the monotonic function may have other forms, for example a polynomial function, an exponential function, a trigonometric function or the like.

According to one embodiment, the difference between the end of transition position and the initial position is a predetermined constant.

According to one embodiment, the difference between the end of transition position of the biasing element and the initial position of the biasing element is between 2 and 200 mm, preferably between 20 and 100 mm.

In embodiments, the difference between the end of transition position and the initial position is calculated according to one or more parameters, for example in depending on an average speed of the biasing element measured during the movement or as a function of the difference between the first load setpoint and the second load setpoint. According to one embodiment, the difference between the end of transition position and the initial position is an increasing function of the average speed of the biasing element. Thus, even in a very fast exercise, the time interval on which the transition load setpoint develops is not likely to shorten to the point of creating user discomfort, for example a shock sensation.

According to one embodiment, the method further comprises:

detecting an instantaneous speed of the biasing element or the moving part of the electric actuator, and

detecting the inversion of displacement of the biasing element between the first direction and the second direction in response to a change of sign of the detected speed.

According to one embodiment, the method further comprises:

detecting the instantaneous position of the biasing element or the moving part of the electric actuator over time,

detecting an extreme position of the biasing element or the moving part of the electric actuator in the first direction,

determining a deviation in the second direction between the detected instantaneous position and the extreme position, and

detecting the inversion of the displacement of the biasing element between the first direction and the second direction when the difference in the second direction crosses a determined inversion threshold.

According to one embodiment, the inversion threshold is a predetermined constant. The value of the inversion threshold is preferably chosen to satisfy two competing objectives, namely to allow reliable detection without false detection or artifacts and to allow a fast response time that is not or hardly perceptible to the user.

According to one embodiment, the inversion threshold is between 2 and 200 mm, preferably between 5 and 20 mm.

In embodiments, the inversion threshold is calculated based on one or more parameters, for example as a function of an average speed of the biasing element measured during the movement. According to one embodiment, the inversion threshold is a decreasing function of the average speed of the biasing element. Thus, the inversion detection can be performed very responsively and without perceptible delay by the user even in a very fast exercise. According to one embodiment, the method further comprises:

in response to detecting a second inversion of displacement of the biasing element between the second direction and the first direction, providing a second transition load setpoint varying progressively from the second load setpoint to the first set point of charge during a second time interval.

According to one embodiment, the method further comprises:

detecting a second initial position of the moving part of the electric actuator or the biasing element at the moment when the second reversal of motion is detected,

calculating a second end-of-transition position having a deviation in the first direction from the second initial position,

supplying the second transition load setpoint in the form of a monotonic function of the position of the moving part of the electric actuator or the biasing element, said monotonic function varying from the second load setpoint to the first load reference between the second initial position and the second end of transition position. This second transition load setpoint can be calculated in the same manner or differently from the first transition load setpoint, depending on whether a symmetrical or asymmetrical behavior of the electric actuator is desired during the two reversals of direction.

According to one embodiment, the method further comprises:

calculating a force to be exerted by the electric actuator at successive instants during the movements of the biasing element as a function of the charge setpoint supplied at each of said successive instants, and

generating a control signal for controlling the electric actuator with the control signal such that the force exerted by the electric actuator in response to the control signal corresponds to the calculated exercise force.

According to one embodiment, the force to be exerted is calculated as a sum of the charge setpoint supplied to each of said successive instants with at least one additive contribution selected from an inertial force contribution proportional to a measured instantaneous acceleration of the part movable electric actuator or biasing element, an elastic force contribution proportional to the difference between a reference position and a measured instantaneous position of the moving part of the electric actuator or the biasing element , and a contribution of strength viscous proportional to a measured instantaneous velocity of the moving part of the electric actuator or the biasing element.

According to one embodiment, the invention also provides an exercise device comprising:

a biasing member to be moved by the force of a user, an electric actuator having a movable portion, the biasing member being coupled to the movable portion;

a computer configured to calculate a force to be exerted by the electric actuator at successive instants during the movements of the biasing element as a function of a charge setpoint supplied at each of said successive instants and to generate a control signal of the electric actuator according to the calculated force to be exerted, wherein the computer is configured to:

in response to detecting an inversion of the displacement of the biasing element input in the first direction and the second direction, providing a transition load setpoint varying progressively from the first load setpoint to the second load setpoint at during a time interval.

According to one embodiment, the biasing element comprises a handle intended to be held in the hand by the user to exert the force of the user, the handle carrying a control member operable by the user to control a function calculator.

Thanks to these features, the handle simultaneously serves as a grip for exercising the muscular force of the user and as a remote control for certain functions of the exercise device, for example setting the load or the inertia or selecting the work program. According to one embodiment, the handle carries a button or a lever "man-death" performing a positive safety function, for example causing a power failure of the actuator in case of releasing the button or lever. According to one embodiment, the control member on the handle controls a load change function at the reversal of movement, namely that the transition between the two load instructions is triggered only if the button or lever is in a state operated at moment the motion reversal is detected. In the opposite case, the load reference remains unchanged when the movement is inverted.

According to one embodiment, the connection between the biasing element and the mobile part comprises a speed reducer to increase the force of the motor. Generally, such a reducer generates additional real inertia for the user who actuates the biasing element. According to one embodiment, the artificial inertia contribution exerted by the electric actuator can compensate all or part of the additional real inertia generated by the gearbox.

According to one embodiment, the electric actuator is a linear motor. According to one embodiment, the electric actuator is a rotary motor in which the movable portion comprises a rotor of the rotary motor.

According to one embodiment, the acceleration sensor comprises:

a position encoder coupled to the moving part for measuring the position of the moving part, the position encoder generating a position signal, branching elements adapted to derive the position signal to determine the acceleration of the moving part.

According to one embodiment, the exercise device is selected from the group consisting of rowing machines, indoor bicycles, lifting bars and guided load devices.

According to one embodiment, the biasing element is movable in a vertical direction and the computer is able to calculate the force to be exerted in the absence of force exerted by the user so that the force to be exerted by the electric actuator has a default load contribution compensating for a self-weight of the biasing element without causing spontaneous displacement of the biasing element in the absence of force exerted by the user.

An idea underlying the invention is to achieve an asymmetrical bias of the user in an eccentric movement and a concentric movement while preserving a comfortable use of the exercise machine, in particular by avoiding shocks during operation. inversion of the movement.

Some aspects of the invention start from the idea of simulating on an exercise machine, when using the machine by a user, a different inertia from the real inertia of the exercise machine using an electric actuator.

Some aspects of the invention start from the idea of designing a machine that allows to vary the weight and the inertia independently of one another. Some aspects of the invention start from the idea of simulating, on the exercise machine, an additional weight using the electric actuator.

Some aspects of the invention start from the idea of simulating, on the exercise machine, additional friction using the electric actuator.

Some aspects of the invention are based on the observation that combining the "inertia" type exercises characteristic of inertia machines and the "weight" type exercises that are characteristic of weight machines in a single machine makes it possible to save a lot of space and to invest less. expensive.

Certain aspects of the invention start from the idea of generating additional inertial forces during certain phases of a muscle exercise performed by the user and canceling these forces of inertia in the other phases of the exercise. muscular.

Certain aspects of the invention start from the idea of generating inertial forces without fixed load to create specific muscle stresses to the inversion of the movement of a mass launched on a substantially horizontal trajectory, in particular the inversion of the movement. of a runner.

The invention will be better understood, and other objects, details, characteristics and advantages thereof will appear more clearly in the course of the following description of several particular embodiments of the invention, given solely for illustrative and non-limiting purposes. with reference to the accompanying drawings.

On these drawings:

• Figure 1 is a schematic representation of an exercise device comprising a motor.

• Figure 2 is a schematic representation of the engine control system shown in Figure 1.

FIG. 3 is a graph of the position and acceleration as a function of time of the handle described in FIG. 1 corresponding to manipulation by the user.

• Figure 4 is a graph of the force exerted by the motor during a manipulation of the device of Figure 7.

FIG. 5 is a graph of the force exerted by the motor during the manipulation of the device in accordance with FIG. 3 corresponding to a first type of exercise. • Figure 6 is a graph of the force exerted by the motor during the manipulation of the device in accordance with Figure 3 corresponding to a second type of exercise.

• Figure 7 is a schematic representation of a variant of the exercise device.

• Figure 8 is a schematic representation partially in section of an exercise device comprising a motor according to another embodiment.

FIG. 9 is a schematic functional representation of a motor control system shown in FIG.

· Figure 10 is a schematic representation of an exercise reversing the movement of a runner.

FIG. 11 is a graphical representation of the operation of a hysteresis comparator that can be used in the control system of FIG. 9.

FIG. 12 is a graphical representation of a load calculation method that can be executed by the control system of FIG. 9.

Figure 13 is a schematic perspective representation of a handle for use in exercise devices.

Figures 1 and 2 illustrate an exercise device in which can be implemented control methods according to the invention. Referring to Figure 1, the exercise device comprises an electric motor 1 which can rotate a shaft 2 and exert a torque on the shaft 2. A pulley 3 is mounted tightly on the shaft 2. A cable 4 is fixed at its first end in the groove of the pulley 3. This cable 4 can be wound in the groove around the pulley 3. At the second end 5 of the cable is fixed a handle 6 through which a user can influence the device with muscle strength when practicing muscle exercises.

The motor 1 comprises a position encoder 10 which measures the position of the motor shaft 2. The position is transmitted to an electronic card 7 in the form of a position signal 9. This electronic card 7 is adapted to receive this signal of position and uses the position signal 9 to generate a control signal. With this control signal, the electronic card 7 controls the torque generated by the motor 1 to control the force exerted by the motor 1, which is transmitted at the handle 6 via the pulley 3 and the cable 4 For this purpose, the electronic card 7 transmits the control signal to the motor 1 via the connection 8. This control signal is received by an organ integrated power supply in the motor 1 which, from this control signal, provides a certain current to the motor 1. The current supplied by the supply member and induces a torque on the movable part 2 and therefore by the intermediate pulley 3 and cable 4 a force on the handle 6. The force exerted by the motor 1 is substantially proportional to the current supplied by the supply member motor 1.

Many control methods can be implemented in such a device to produce different muscular stresses. A first example is to simulate the presence of a predetermined mass suspended from a cable, namely that the motor torque exerts on the handle 6 a constant load as to the direction and intensity.

When a user manipulates the handle 6 during an exercise it opposes the force of the engine 1 with the help of his muscular strength. For example, during a practicable exercise with this device, a user is positioned above the device and pulls the handle 6 from a low position to a high position with his hands. During this upward movement, the user must overcome the downward force exerted by the motor 1 on the handle 6. When the handle 6 arrives in the up position, the user performs the reverse movement and returns the handle 6 to the lower position while still being forced by the same force submitted in the same direction by the motor 1. During the descent, the user accompanies and brakes the movement of the handle down. The exercise device thus simulates a mass to be alternately raised and rested by the user

During this exercise, the position signal is continuously transmitted to the electronic card 7 which calculates and transmits to the motor continuously the corresponding control signal. Thus, the device controls the force generated by the motor 1 throughout the exercise.

However, a slight offset may be present in theory between the moment when the encoder transmits the position and the torque exerted by the motor 1 because of the response time of the motor 1 to the control signal and the response time of the electronic card 7 With good quality electronic components, this offset remains imperceptible and has no effect on the user's sensations of the exercise device.

With reference to FIG. 2, the motor control means will now be described more precisely with reference to a second example.

The electronic card 7 here comprises a microprocessor 20. A position encoder 10 measures the position of the motor shaft 2, this position is encoded into a position signal which is transmitted via the connection 38 to the microprocessor 20. Thus, in a mode of This measurement can be carried out every 30 ms and preferably every 5 ms. In this microprocessor 20, the position signal is transmitted to a branch member 13 via the connection 18. The branch member derives the position signal thereby generating a speed signal which is transmitted to a second branch member 14 via the connection 15. The second branch member derives the speed signal thereby generating an acceleration signal. The acceleration signal is transmitted via the connection 17 to a calculation module 12. Furthermore, the position signal and the speed signal are respectively transmitted to the calculation module 12 via the connections 11 and 16. The calculation module 12 calculates the control signal to be supplied to the motor and transmits it to the motor via the connection 19.

More specifically, the control signal is calculated from the acceleration so that the force exerted by the motor 1 on the handle 6 includes the downward load and a predetermined artificial inertia.

For this, the calculation module 12 takes into account the accumulation of the torque exerted by the motor 1 and the inertia of the rotating parts of the device connected to this motor which are the shaft 2, the pulley 3, the cable 4 and the handle 6 .

Indeed, when a user manipulates the handle 6:

m _r xy = F _m + F _s (1)

Where F _s is the force exerted by the user on the handle 6, F _m is the force exerted by the motor 1 on the handle 6 and controlled by the calculation module 12, m _r is the inertia of the moving parts reduced to the handle 6 and the mass of the handle 6 and y is the acceleration of the handle 6.

Equation (1) corresponds to the fundamental principle of the dynamics applied to a system in translation. However, those skilled in the art will understand that torques exerted on a rotating system can be similarly modeled.

The force exerted by the motor F _m is composed of two components induced by the control signal: a fixed component F _ch representing the load and a component proportional to the acceleration Fj which represents the artificial inertia. So :

Or the force j is defined according to a coefficient of proportionality k:

The coefficient k is a parameter that is programmed in the calculation module 12. The equation (1) can be rewritten:

(m _r + fc) xy = F _ch + F _s (4) In this way, if the proportionality coefficient k used to produce the control signal is negative, namely- m _r <k <0, the device simulates an inertia less than the actual inertia of the device, that is, say the inertia of the rotating parts of the device. If the coefficient of proportionality k is positive, the device simulates greater inertia than the actual inertia of the device.

The user, through a not shown user interface can change the values of the fixed component F _ch and the proportionality factor k and thus determine the type of effort with which it wishes to practice. Thus, it is possible to independently vary the load of inertia. A wide range of types of muscle exercises can be offered to the user.

The user interface is connected to the calculation module 12 and is able to receive data on position, speed, acceleration or information calculated from these data, for example, the effort provided or the power expended. These data and information are calculated by the calculation module 12 from the acceleration, speed and position signals transmitted to the calculation module 12 respectively by the connections 17, 16 and 11. With these data and these information, the user interface may sensually solicit the user by displaying this information. The user can in this way follow the level of his effort during his physical exercises. However, these solicitations can be of different natures, solicitations are for example possible. Furthermore, the user interface comprises control members allowing the user to vary the values of the fixed component F _ch and the proportionality factor k, preferably independently of one another. These control members are for example buttons on the user interface corresponding to fixed component pairs F _ch and proportionality factor k predetermined. These couples define several types of exercises. A storage device, for example a memory in the calculation module 12, stores this information and data. Thanks to this storage, the user can follow the evolution of his performances over time. With reference to FIGS. 3, 5 and 6, several particular examples of exercises that can be produced by the device presented above will be described.

FIG. 3 represents the position of the handle 6 along the z axis of FIG. 1 and the acceleration of the handle 6 as a function of time during the tensile stresses of the handle presented with reference to FIG. dashed curve 21 represents the position of the handle which is measured by the position encoder 10. The continuous curve 22 represents the acceleration corresponding to the position curve 21. By convention, the z-axis has been oriented downwards in FIG. 24 of the position curve 21 corresponds to the moment when the handle 6 is in the low position and the point 23 corresponds to the upper position of the handle.

For purposes of illustration between point 23 and point 25, the position curve 21 is substantially sinusoidal. Thus the acceleration also forms, along this period, a sinusoidal curve. Subsequently, the position curve is no longer sinusoidal and therefore the acceleration is no longer sinusoidal.

FIG. 5 represents the force opposed by the motor 1 to the user as a function of time for the same time interval as FIG. 3. The curve 28 is constant at a threshold 26. Indeed, FIG. at a first exercise where the calculation module provides a control signal to the motor so that the force opposite the user is constant over time. For this, the calculation module produces a control signal inducing a force having a load component equal to the threshold 26 and a zero inertia component. In this exercise, the user therefore only opposes a fixed load and the actual inertia of the system.

FIG. 6 represents a second exercise which partially uses the principle of the first exercise presented with reference to FIG. 5. The curve 40 represents the force generated by the engine 1 during this exercise. It comprises two phases: a high phase 31 during which the curve is constant at the threshold 27 and a low phase during which the curve adopts the shape of the acceleration curve at the threshold 27. Indeed, the user is subjected to a load force corresponding to the threshold 27 when the measured acceleration is positive, that is to say here during high phases 31 of the handling of the handle where the handle is close to its high position 23. The user is however subjected to an additional inertial force oriented in the same direction as the load force when the measured acceleration is negative, that is to say during a low phase 29 when the handle arrives in low position 24 and the user decelerates the descent and then accelerates to pull the handle to the high position 23. This low phase corresponds to the phase 30 during which the acceleration is negative. In this way, the user is subjected to additional artificial inertia when he arrives in the low position and wishes to raise the handle to the high position, that is to say at the moment when his muscle solicitation is the most intense. Thus, the exercise device makes it possible produce an additional solicitation that opposes the user when reversing the direction of movement of that user.

For the implementation of the second exercise, the calculation module 12 applies a coefficient of proportionality k determined as follows:

If (gamma) γ> 0, k = 0 (5)

If y <0, k = + k ₀ , ie k> 0 (6)

Where k ₀ is a predetermined positive constant.

The exercises described above are given for illustrative purposes. In particular, the calculation module can control the coefficient of proportionality k in multiple ways. By way of example, the calculation module can vary the coefficient of proportionality as a function of the position or the speed of the handle. Thus, in one variant, the exercise device produces an additional inertia component when the handle reaches a certain position. In a variant of the exercise device, this additional inertia component is added when the speed is in a particular direction. In this way a multitude of interesting exercises for muscle development can be produced. This allows in particular to solicit the muscles of the user more intensely when they are in a particular position.

In a variant of the device shown in FIG. 1, the driving shaft 2 is connected to a speed reducer having a reduction ratio r. The presence of such a gearbox makes it possible to generate relatively large forces while reducing the size of the motor, in order to miniaturize the device. The pulley 3 is fixed on an output shaft of the gearbox. In this variant, the presence of a reducer greatly increases the real inertia of the moving parts of the motor 1 brought back to the handle 6. The real inertia of the device is also increased by the reduced inertia of the rotating parts of the gearbox. The inertia of the motor and the gearbox brought back to the output of the gearbox J _tot can be written:

with the inertia of the reducer J _rec j and the real inertia of the motor J _word . Thus, if the reduction ratio r is large, the actual inertia of the system is greatly increased. Thus, the use of a proportional factor k negative in this variant to compensate all or part of the inertia induced by the reducer. This compensation is all the more precise as the acceleration which is measured to generate the artificial inertia force is the acceleration of the motor shaft 2, so that this measurement takes into account the effect of the gearbox, which effect consists in increasing by the ratio r the acceleration at the level of the motor shaft 2 with respect to the acceleration exerted on the handle 6. The very simple exercise device described with reference to Figures 1 and 2 is given for illustrative purposes, the invention is therefore in no way limited to this type of exercise device. In particular, the invention can be adapted to any type of exercise machine that solicits any part of the body. By way of example, the invention may be adapted to constitute a rowing-type device, an indoor bicycle or a lifting bar.

Referring to Figure 7 there is shown an exercise device 50 for exercising the muscles of the arms in traction and thrust in which control methods according to the invention can be implemented.

The device 50 comprises two levers 53 which can be moved alternately forwards and backwards by a user. The levers 53 are each coupled to an electric motor 54 which is controlled by the control device 55. According to one embodiment, the motors 54 are controlled so as to generate a force represented by the curve 33 of FIG. For purposes of simplification, the rotary motion of the levers is approximated in a linear motion along the x-axis.

Thus, FIG. 4 represents the effort opposed to a user in the context of the exercise device represented in FIG. 7. Curve 33 represents the force generated by the motor and has a value proportional to the acceleration curve 30. A user is assumed to bias the lever 53 so that the measured position and the acceleration are the same as in FIG. 3, the x-axis here replacing the z-axis. In this type of exercise, the control device 55 subjects a control signal to the motors 54 which does not induce a charge component. Only a component of artificial inertia is produced by the motors 54. Thus, the effort experienced by the user is proportional to the acceleration and therefore corresponds to a simulated inertia without load that is greater than the actual inertia of the device.

This type of stress with artificial inertia without additional load is also interesting in an exercise machine soliciting the leg muscles. Indeed, the muscular stress produced by the motor when it is controlled in this way corresponds substantially to the muscular stress required to reverse the movement of a runner on a horizontal ground. Such an exercise is illustrated in Figure 10.

In FIG. 10, the runner 34 is initially running at high speed in the x-axis direction, as schematically represented by the speed vector 35. At the end of the exercise, the runner 34 is in the process of to run at high speed in the opposite direction to the x-axis, as schematically represented by the speed vector 36. During the exercise, the rider 34 therefore had to slow down his movement until the stoppage, by example at the point χθ, then re-accelerate in the other direction. The muscles of the runner 34 were therefore solicited during this exercise essentially to overcome the inertia of the runner itself, oriented along the x axis. The force of gravity being perpendicular to the movement, it does not create any particular muscular solicitation in this exercise, that is to say that the muscular solicitation specific to the exercise is a solicitation of pure inertia. The exercise machine programmed to produce this type of solicitation is all the more advantageous as this race reversal situation is very common in ball sports, for example rugby or football.

Similarly, a control program associating the artificial inertia force with a constant load makes it possible to produce a muscular solicitation similar to the accomplishment of the same exercise on a sloping ground.

A device for simulating an additional viscous friction force will now be presented. The device is similar to the device described in FIG. 7 and comprises a microprocessor having the same structure as the microprocessor 20 of the control system described in FIG. 2. The force exerted by the motor here comprises three components. The first two components correspond to the load component and the inertia component described above. The third component is a viscous friction component. So :

Where the force Ff _V , corresponding to the viscous friction component, is defined according to a coefficient of proportionality k ₂ and as a function of the speed v of the handle:

F _fv = k ₂ xv (9)

The speed v is determined by the calculation module 12 by means of a speed signal which is transmitted to the calculation module 12 via the connection 16.

Thus, when the user moves the levers in one direction, the motor generates a torque on the lever comprising the viscous friction component proportional to the speed of movement of the lever in addition to a component of inertia. This viscous friction component causes an additional stress which opposes the direction of movement of the user. In this way, the device simulates a viscous friction that can be produced by a machine comprising a finned system.

The coefficient k ₂ can be a constant stored in the memory of the microprocessor 20. In the same way as the inertia component, the calculation module 12 can control the coefficient of proportionality k ₂ in multiple ways. For exemple, the calculation module can vary the coefficient of proportionality k ₂ depending on the position of the handle.

Referring to Figures 8 and 9, there will now be described another exercise machine 60 using an electric motor. The machine 60 has a relatively similar shape to a weight machine known as a squat machine. But it can provide a much wider range of muscular solicitations.

The structure of the machine comprises a metal base 61 placed on the ground, shown in section in Figure 8, and a guide column 62 vertically fixed to the base 61. The upper surface of the base 61 is a platform 68 for receiving an athlete, for example in standing position as illustrated in ghost line. A carriage 63 is slidably mounted on the column 62 by guide means not shown, so as to translate vertically along the column 62. According to one embodiment, the carriage 63 is a four-sided structure which completely surrounds column 62, one and the other having a square section. The carriage 63 carries gripping rods 69 which extend above the platform 68 and are intended to be engaged with the athlete, for example at the level of his shoulders or his arms or legs according to the invention. desired exercise.

A transmission belt 64 is mounted in the column 62 and extends between a idler pulley 65 pivotally mounted at the top of the column 65 and a driving pulley 66 pivotally mounted in the base directly above the column 62. belt 64 is a toothed belt that makes a round-trip closed loop between the pulleys 65 and 66 so as to be coupled without slipping to the drive pulley 66. The carriage 63 is secured to one of the two branches of the belt 64, by example by means of rivets 67 or other fastening means, so that it is also coupled without slipping to the drive pulley 66, any rotation of the pulley 66 resulting in a vertical translation of the carriage 63. Preferably, the belt 64 is formed of an AT-type toothed belt whose two ends are fixed to the carriage 63, so as to close the loop at the carriage 63.

A motor unit 70 is housed in the base 61 and coupled to the drive pulley 66 by means of a speed reducer 71. More specifically, the speed reducer 71 comprises an input shaft 72 coupled without sliding to the motor shaft 70 of the motor group, which is shown in more detail in Figure 9, and an output shaft 73 which carries the drive pulley 66. The speed reducer 71 imposes a reduction ratio r between the rotational speed wl of the 72 and the rotation speed w2 of the shaft 73, namely wl / w2 = r. According to embodiments, the reduction ratio r is chosen between 3 and 100, and preferably between 5 and 30.

The machine 60 also comprises a control console 74 which can be integral with the base 61 or independent thereof. In addition, a power cable 75 out of the base 61 to be connected to the power grid. The machine 60 does not require exceptional electrical power and can therefore be powered by a common home network.

FIG. 9 represents more precisely the motor group 70 and its control unit 80, which is also housed in the base 61. The motor unit 70 comprises an electric motor 76, for example an autopilot synchronous motor, and a current variator 77 which controls the supply current 78 of the motor 76.

It is recalled that the synchronous motor autopilot has a constant rotor flow. This flux is created by permanent magnets or coils mounted in the rotor, while the variable stator flux is created by a three-phase winding to orient it in all directions. The electronic control of this motor is to control the phase of the current waves so as to create a rotating field, always ahead of 90 ° on the field of the magnets, so that the torque is maximum. Under these conditions, the engine torque on the drive shaft 2 is proportional to the stator current. This current is precisely controlled in real time by the control unit 80 via the current controller 77.

For this purpose, the control unit 80 comprises a low-level controller 81, for example of the FPGA type, which receives the position signal 83 from the position encoder 84 of the motor shaft 2 and performs real-time calculations. from the position signal 83 to determine the instantaneous values of the position, the speed and the acceleration of the motor shaft 2. The position encoder 84 is for example an optical device which provides two squared signals in quadrature according to the technique known.

The high-level controller 82 includes a memory and a processor and executes complex control programs from the information provided in real time by the low-level controller 81. Possible control programs have been described above with reference to the Figures 3 to 6.

The control console 74 is connected to the high-level controller 82 via a wired or wireless TCP / IP link 85, and has an interface allowing the athlete or his trainer to select pre-recorded exercise programs or to adjust them. precisely and in a personalized way the parameters of such a program. In the example shown, the interface is a touch screen 86 which includes a slider 87 for setting the value of the load F _ch along a predetermined scale, for example 0 to 3000 N, and a slider 88 to set the value of the coefficient k along a predetermined scale, that is to say the artificial inertia force F; .

Depending on the exercise program executed, the high-level controller 82 processes the information provided in real time by the low-level controller 81 and calculates the instantaneous torque to be exerted by the motor group 70. The base controller Level 81 generates a control signal 90 corresponding to this instantaneous torque and transmits the signal 90 to the current controller 77, for example in the form of an analog control voltage varying between 0 and 10V. Alternatively, a CAN digital interface may also be used.

The control programs for simulating different exercises can be very numerous. Preferably, regardless of the detail of the program, it is always the athlete who controls the machine 60 and the machine 60 that reacts to the solicitation exerted by the athlete on the gripping bars 69. For this, it is preferable that the machine 60 can react quickly to changes of direction imposed by the athlete, despite the friction that inevitably exist in such a mechanical system.

For this, according to one embodiment, the high-level controller 82 implements a friction compensation algorithm that will now be explained.

Note the mass of the carriage 63. Note Fc = (mc.g), the force that must impose the motor 76 on the belt 64 to compensate for the weight of the carriage 63 without the user does not support any load. The algorithm uses parameters Fa and Fb defined by the fact that if the motor 76 applies (Fc + Fa) the carriage 63 is at the limit of the movement in the positive direction, upwards, and if the motor 76 applies (Fc -Fb) the carriage 63 is at the limit of the setting in motion in the negative direction, downwards. These parameters Fa and Fb can be measured experimentally. The algorithm governs the transition from the force (Fc + Fa) to the force (Fc - Fb) in the event of a change in the direction of the user's request. The algorithm applies laws that use the linear velocity v of the carriage 63 and a coefficient kf, namely:

FchO = Fc + kf. v (10)

(Fc - Fb) <FchO <(Fc + Fa) (1 1)

Where FchO denotes the force imposed by default on the belt 64 by the motor 76, namely the value that is applied when the cursor 87 is placed on the graduation 0. In other words, if the cursor 37 is placed on the graduation 3000N for a program exercise planing to exercise this load alternately in both directions, and the carriage 63 weighs 60kg, the electric motor will actually exert a force of about 3600 N in climb and 2400 N downhill.

Thus, the higher the coefficient kf, the faster the machine reacts to changes in direction imposed by the user. Beyond a certain limit, a very strong reactivity could require a frequency filtering of the speed measurement, for example of the low-pass type of the first order.

According to the selected program, for example when an artificial inertia force proportional to the acceleration and / or a viscous force proportional to the speed is applied by the engine, or when the program provides different reactions in the concentric direction and in the the eccentric direction, the force to apply calculated can undergo a discontinuity at the time of reversal of the direction, which is necessarily detrimental to the comfort of use of the machine.

According to one embodiment, the high-level controller 82 implements an algorithm to avoid these discontinuities. To do this, the controller 82 detects a change of direction by the passage of the speed signal in a hysteresis comparator shown schematically in FIG. 11.

At the start of the concentric phase, if the speed v> ε, the controller 82 triggers the passage from F2 to Fl. This variation is made at constant rate of variation per unit of time, for example of the order of 200N / s .

Similarly, during the passage from the concentric phase to the eccentric phase, when the speed becomes negative and passes under a threshold v <-ε, the controller 82 triggers the passage from Fl to F2. The threshold value ε is chosen so as to ensure sufficient stability, namely that the engine does not go from F1 to F2 untimely when the athlete decides to stop during his movement.

In FIG. 11, it should be noted that the curves of variation of the force as a function of the speed between the values F1 and F2 are not imposed by the system and depend in fact on the behavior of the user, namely how he does vary the speed as a function of time, since the system imposes a force variation rate as a function of time.

In addition, the control program can prevent the motor from making more than two consecutive changes if the difference in position of the moving part between the two changes does not exceed a certain limit, for example of 10 cm. In other embodiments, the exercise program may also comprise an elastic force contribution F _e defined as a function of a coefficient of proportionality k ₃ and as a function of the position z of the carriage 63:

F _e = k ₃ x (z - z0) (12)

Where zO is a parameterizable reference height and the position z is determined by the low-level controller 81.

It is therefore understandable that many exercise programs can be designed by combining additive contributions selected from the group comprising an artificial inertia contribution proportional to the measured acceleration, a viscous friction contribution proportional to the measured speed, a elastic contribution proportional to the measured position, and a predetermined load contribution. According to one embodiment, the human-machine interface allows the user to independently adjust the parameters of each of these contributions, in particular the coefficients k, c _{2 3} .

When the exercise program is asymmetrical, namely that it provides for different reactions in the concentric and eccentric directions, for example a first load value F _Ch = F _A in the uplink direction and a second load value F _Ch = F _D <F _A in the downward direction of the carriage, the force applied by the actuator can undergo a discontinuity at the time of reversal of the direction. The use of a force ramp having a constant rate of change per unit of time to eliminate this discontinuity at the time of the reversal of direction, however, has a disadvantage in the case of exercise performed at a high speed. Indeed, this ramp force is spread over a period of time determined by the difference between the load values F _D and F _A. At a high speed, the user can perform a significant portion of the stroke of the carriage during the time interval of the transition, so that the charges theoretically provided for the exercise are only applied to a small portion of the exercise and that a goal of the exercise program in athletic and physiological terms is not really achieved.

Referring to FIG. 12, an alternative method of calculating the load component at the time of reversal in an asymmetric exercise program will now be described. In Fig. 12, the abscissa represents the position of the carriage 63 along an upward z axis, and the ordinate axis represents the load component applied by the electric actuator during an exercise. .

The principle of this method is explained with reference to a cyclic up-down movement carried out by a user and shown schematically in FIG. comprises a rising phase symbolized by the arrows directed in the positive direction of the z axis and a descent phase symbolized by the arrows directed in the negative direction of the z axis. The points M (abscissa z ₂ -a ₂ ) and P (abscissa Z 1 + ai) are the points where the two changes of direction of movement made by the user are respectively detected. The exercise program provides a load component F _C h = F _A in the upstream direction and a load component F _ch = F _D <F _A in the downward direction of the carriage. This charge component is optionally combined with other additive components not shown, as previously described.

In the case of the up / down inversion, from the current position of the carriage at the moment when the reversal of direction is detected (point M, abscissa z ₂ -a ₂ ), a transition end position is computed at distance b ₂ , namely the point N (abscissa z ₂ -a ₂ -b ₂ ). Then the load component is calculated as a decreasing monotonic function, for example linear, of the position of the carriage between the points M and N to go from F _A to F _D.

In the case of the descent to descent inversion, from the current position of the trolley at the moment when the reversal of direction is detected (point P, abscissa Zi + a, a position of end of transition is calculated at a distance b ie the point Q (abscissa Zi + ai + bi) Then the load component is calculated as an increasing monotonic function, for example linear, of the position of the carriage between the points P and Q to go from F _A to F _D.

The distances b ₁ and b ₂ are for example constant parameters, possibly equal, stored in the memory of the control unit 80. Preferably, the distances b ₁ and b ₂ are between 20 and 100 mm. In FIG. 12, the distances b ₁ and b ₂ have been exaggerated for the sake of readability, but in practice the distances b ₁ and b ₂ may represent a very small proportion of the stroke of the carriage.

The above method may be employed with different methods of detecting motion reversal such as a method based on detecting a sign reversal of the detected velocity or any other suitable method. We will now describe a particular detection method which is also illustrated in Figure 12.

In the movement shown schematically in FIG. 12, the extreme points actually reached by the carriage 63 are at the top the point T (abscissa z ₂ ) and at the bottom the point S (abscissa Zi). The detection of the reversal of the upward movement is here based on a hysteresis threshold of position a ₂ : the method consists in detecting the extreme position T and detecting the distance traveled in opposite direction from the extreme point. When this distance reaches the position hysteresis threshold a ₂ (point M, abscissa z ₂ -a ₂ ), the inversion detection takes place. In the same way, the detection of the reversal of the descent to uphill movement is based on a position hysteresis threshold ai: the method consists in detecting the extreme position S and detecting the distance traveled in opposite direction from the extreme point. When this distance reaches the position hysteresis threshold ai (point P, abscissa Zi + a,), the inversion detection takes place.

The thresholds ai and a2 are for example constant parameters, possibly equal, stored in the memory of the control unit 80. Preferably, the thresholds ai and a ₂ are between 5 and 20 mm. In Figure 12, the distances ai and a ₂ have been exaggerated for readability, but in practice the distances ai and a ₂ may represent a very small proportion of the stroke of the carriage.

In the methods described above, it will be appreciated that abscissas Z 1 and z ₂ are set by the user and not by the control unit. Nothing imposes that the movement of the user is perfectly repetitive. The points S and T may therefore be different at each cycle and the other points are calculated each time as a result of the actual movement made by the user.

The methods described with reference to FIG. 12 for effecting a transition between two values of the load component F _ch are similarly applicable to other parameters of an asymmetric exercise program. In a second example of asymmetric exercise, the coefficient k used to generate the artificial inertia component takes different values in the concentric direction and in the eccentric direction, for example a first value k = k _A in the uplink direction and a second value k = k _D <k _A in the downward direction of the carriage. In this exercise, the force applied by the actuator may also be discontinuous because the instantaneous acceleration is typically high at the time of reversal of direction. In the same way, it can be provided a method of calculating the coefficient k which performs a smoother transition at the time of the reversal of direction. The principle of this method will be understood immediately according to the indications in parentheses of the parameters (k), (k _D ) and (k _A ) in FIG.

Such a change in the value of the coefficient k used to generate the artificial inertia component can also be implemented at the moment when the acceleration changes sign canceling, in which case no progressive transition is necessary since the component d artificial inertia is substantially zero at the moment of the change in value.

In an alternative embodiment, the coefficient k used to generate the artificial inertia component varies as a function of one or more parameters of the movement, for example according to an increasing linear function of the measured acceleration. By way of illustration, reference has been made to the carriage 63 of FIG. 8 in the above description, but any exercise machine whatever the shape of its mobile biasing element can exploit the calculation methods indicated below. above.

Referring to Fig. 13, there is shown a biasing member in the form of a handle 91 having control buttons 92 and 93 operable to control, trigger, or inhibit various functions of the exercise machine in the manner of 'a remote control. In the example shown, the handle 91 intended to be held with one or both hands is attached to the end of a rope 94 used for example in the machine of Figure 1. In addition to being used to exercise the force of traction on the rope 94, the handle 91 thus allows the machine to be controlled during the exercise. For this, the button 92 located at the end of the bar is actuable by a thumb pressure, while the elongated button 93 is operable by pressing the fingers of the hand by tightening the bar 95. These positions of the buttons 92 and 93 are purely illustrative .

The functions of the buttons 92 and 93 can be various. In one example, the button 93 performs a "deadman" function, namely that the motor power supply is deactivated as soon as the button 93 is released, which fulfills a security purpose. In one example, the button 92 performs a function for triggering the load value change, namely that the transition between two load values F _A and F _{D takes} place only if the button 92 is depressed when the motion reversal is detected. If not, the exercise continues with a constant load value before and after the reversal of motion.

In another example, user actuation of the button 92 or 93 immediately triggers a gradual transition of the load component from a first programmed value FA to a second programmed value FB, larger or smaller, regardless of the phase. of the movement during which this operation is performed.

Other types of control elements may be arranged similarly on the handle 91 or on the gripping bar 69, for example buttons, levers, potentiometers and the like to facilitate the control of the machine by the user during the operation. 'exercise.

Although the embodiments described above include rotary motors, the control methods described above may be employed with any other type of electric actuator. In particular, a linear motor can be used to generate a force on the handling element. Moreover, the calculation of the control signal can be performed in different forms, unitarily or distributed, by means of hardware and / or software components. Useful hardware components are ASIC specific integrated circuits, FPGA programmable logic networks or microprocessors. Software components can be written in different programming languages, for example C, C ++, Java or VFfDL. This list is not exhaustive.

Although the invention has been described in connection with several particular embodiments, it is obvious that it is not limited thereto and that it comprises all the technical equivalents of the means described and their combinations if they are within the scope of the invention.

The use of the verb "to include", "to understand" or "to include" and its conjugated forms does not exclude the presence of other elements or steps other than those set out in a claim. The use of the indefinite article "a" or "an" for an element or a step does not exclude, unless otherwise stated, the presence of a plurality of such elements or steps. Several means or modules can be represented by the same hardware element.

In the claims, any reference sign in parentheses can not be interpreted as a limitation of the claim.

Claims

A control method for controlling an electric actuator (1, 76) in an exercise device having a biasing member (6, 69, 91) to be moved by the force of a user and coupled to a moving part (2) electrical actuator, a control unit (12, 80) and a position encoder (84) arranged to detect an instantaneous position of the movable portion (2) and to generate a representative position signal (83). the instantaneous position of the moving part (2), the control method comprising the execution by the control unit (12, 80) of the steps of:

receive the position signal generated by the position encoder,

detecting a direction of displacement of the biasing element from the position signal generated by the position encoder,

supplying a first load reference (F _A , k _A ) during a displacement of the biasing element in a first direction, the first load setting being a control signal (90) generated by the control unit ( 12, 80),

supplying a second load reference (FB, k _B ) during a displacement of the biasing element in a second direction opposite to the first direction, the second load setting being a control signal (90) generated by the unit control (12, 80), and

in response to detecting an inversion of the displacement of the biasing element between the first direction and the second direction, providing a transition load setpoint varying progressively from the first load setpoint to the second load setpoint at during a time interval,

the method further comprising:

detecting an initial position (M) of the moving part of the electric actuator or the biasing element at the moment when the reversal of motion is detected from the position signals (83) generated by the position encoder (84) ,

calculating an end of transition position (N) having a deviation in the second direction from the initial position,

supplying the transition charge setpoint in the form of a control signal (90) generated by the control unit (80), said control signal being representative of a monotonic function of the position of the moving part of the actuator or biasing element, said monotonic function varying from the first load setting (FA _> k _A ) to the second load setting (F _B , k _B ) between the initial position (M) and the position end of transition (N).

2. Method according to claim 1, wherein the transition charge setpoint varies with a rate of change per unit of constant displacement from the first load setpoint to the second load setpoint, the monotone function being an affine function.

3. Method according to claim 2, wherein the difference between the end of transition position (N) and the initial position (M) is a predetermined constant (0 ₂ ).

4. Method according to one of claims 1 to 3, wherein the difference between the end of transition position of the biasing element and the initial position of the biasing element is between 2 and 200 mm, preferably between 20 and 100 mm.

5. Method according to one of claims 1 to 4, further comprising:

6. Method according to one of claims 1 to 5, further comprising:

detecting an extreme position (T) of the biasing element or the moving part of the electric actuator in the first direction,

detecting the inversion of the displacement of the biasing element between the first direction and the second direction when the difference in the second direction crosses a determined inversion threshold (a ₂ ).

The method of claim 6, wherein the inversion threshold (a2) is a predetermined constant.

8. Method according to one of claims 6 to 7, wherein the inversion threshold is between 2 and 200 mm, preferably between 5 and 20 mm.

9. Method according to one of claims 1 to 8, further comprising:

in response to detecting a second inversion of displacement of the biasing element between the second direction and the first direction, providing a second transition load setpoint varying progressively from the second load setpoint to the first set point of load during a second time interval,

detecting a second initial position (P) of the movable portion of the electric actuator or the biasing element when the second motion reversal is detected,

calculating a second end-of-transition position (Q) having a deviation in the first direction from the second initial position,

supplying the second transition load setpoint in the form of a monotonic function of the position of the moving part of the electric actuator or the biasing element, said monotonic function varying from the second charging setpoint to the first load reference between the second initial position (P) and the second end transition position (Q).

10. Method according to one of claims 1 to 9, further comprising:

generating a control signal for controlling the electric actuator with the control signal such that the force exerted by the electric actuator (1,76) in response to the control signal corresponds to the calculated exercise force.

11. The method of claim 10, wherein the force to be exerted is calculated as a sum of the load reference (F _C h) supplied to each of said successive instants with at least one additive contribution selected from a contribution of inertial force. proportional to a measured instantaneous acceleration of the moving part of the electric actuator or the biasing element, an elastic force contribution proportional to the difference between a reference position and a measured instantaneous position of the moving part of the electric actuator or biasing element, and a viscous force contribution proportional to a measured instantaneous velocity of the movable portion of the electric actuator or the biasing member, the viscous force having a predetermined viscosity coefficient stored in a memory.

12. Exercise device comprising:

a biasing member (6, 69, 91) to be moved by the force of a user, an electric actuator (1, 76) having a movable portion (2), the biasing member (6, 69, 91); ) being coupled to the moving part,

a position encoder arranged to detect an instantaneous position of the moving part and to generate a position signal representative of the instantaneous position of the moving part

a calculator (12, 80) configured to calculate a force to be exerted by the actuator at successive instants during the displacements of the biasing element as a function of a charge setpoint supplied to each of said successive instants and to generate a control signal of the electric actuator as a function of the calculated force to be exerted, wherein the computer (12, 80) is configured to:

receiving a position signal generated by the position encoder and detecting a direction of displacement of the biasing element from the position signal generated by the position encoder,

supplying a first load setpoint during a displacement of the biasing element in a first direction, the first load setting being a control signal (90) generated by the computer (12, 80),

providing a second load setpoint during a displacement of the biasing element in a second opposite direction to the first direction, the second load setting being a control signal (90) generated by the computer (12, 80), and

in response to the detection by the computer (12, 80) of an inversion of the displacement of the biasing element between the first direction and the second direction, providing a transition load setpoint varying progressively from the first load setpoint up to at the second load setting during a time interval,

detecting an initial position of the movable portion of the electric actuator or the biasing member at the time the motion reversal is detected from the position signal (83) generated by the position encoder (84),

calculating in the calculator (12, 80) a transition end position having a deviation in the second direction from the initial position, and

supplying the transition charge setpoint in the form of a control signal (90) generated by the computer (12, 80), said control signal being representative of a monotonic function of the position of the moving part of the electric actuator or biasing element, said monotonic function varying from the first load setting to the second load setting between the initial position and the transition fm position.

The device according to claim 12, wherein the biasing element comprises a handle (91) to be manually held by the user to exert the force of the user, the handle carrying a control member (92). , 93) operable by the user to control a function of the computer (12, 80).