"JOINT ACTUATION DEVICE"

FIELD OF INVENTION

The present invention relates to a j oint actuation device to be applied in the field of devices for assisting or rehabilitating various human movements , such as walking, running, j umping, and moving loads .

BACKGROUND TO THE INVENTION

Exoskeletons tend to consist of an anthropomorphic mechanical structure that mimics the movement of the limb to which it is applied, along with force-generating elements and a control system, and their function is to ampli fy, restore , or support the movement made by the user .

Joint actuation systems for exoskeletons can be distinguished as : active systems which use external power supplies to support the user in their movement ; and passive systems which use only the energy contained in the exoskeleton-user system, notably through the use of passive elements such as springs , to support the user in their movement . Thus , in the case of active systems , one of the maj or challenges is the energy supply, and the weight and volume associated with storing this energy, for example through the use of batteries . Passive exoskeletons , on the other hand, although lighter and simpler, have limitations regarding their ef ficiency and ability to adapt to complex movements such as locomotion . A number of devices have been described in the literature that aim to respond to the existing challenges for passive exoskeletons .

Patent US 9492302 B2 relates to an apparatus and clutch for use in the storage and control led release of mechanical energy to aid locomotion . It is an exoskeleton with a passive mechanism compris ing a linear spring external to the quasipassive angular action device and whose operation is based on prediction by means of previously collected data . Power is transmitted using a discontinuous clutch, with no possibility of real-time control .

Patent application US2017246492 Al describes a pass ive exoskeleton that monitors the angular position and angular velocity of the leg, which makes it possible to predict the next movement . The force trans fer system is composed of a discontinuous clutch and an elastic element . The mechanism is operated by actuating the clutch during the entire locking period .

Patent application EP 3820410 Al relates to an autonomous exoskeleton device whose operation is based on the angular position and angular velocity to recognise the locomotion pattern and predict the next movement . The momentum that the exoskeleton provides to the user can be adj usted .

There is therefore a need in the technique for a j oint actuation device that overcomes the limitations of current state-of-the-art devices , in particular there is a need for a device capable of providing improvements in terms of its metabolic costs, while ensuring high efficiency and the ability to adapt to complex movements such as locomotion.

SUMMARY OF THE INVENTION

The present invention relates to a joint actuation device (50) comprising a support structure (20) consisting of a spring chassis (1) connected to a clutch chassis (13) , where the chassis (1, 13) define an interior of a support structure comprising:

• a one-way clutch (30) comprising a cam (4) , at least three rollers (9) , a pin (10) and a lever (11) , wherein the cam (4) is coupled to the clutch chassis (13) , the lever (11) is coupled to the cam (4) and a first end of the pin (10) is connected to the lever (11) ;

• an electronic control unit (ECU) (40) comprising a microprocessor (6) , an actuator (7) , a rod (8) , an inertial measurement unit (IMU) (5) and at least one angular position sensor (2) , in which,

• the inertial measurement unit (5) , the actuator (7) and the angular position sensor (2) are connected in data communication to the microprocessor (6) ;

• the microprocessor (6) , the inertial measurement unit (5) and the actuator (7) are housed in the aforementioned cam (4) of the clutch ( 30 ) ;

• the angular position sensor (2) is housed in the said spring chassis (1) and connected to the said clutch cam (4) (30) ; and • the rod (8) is housed inside the actuator (7) and connected to a second end of said pin (10) of the clutch (30) ; and

• at least one spring (3) housed in the spring chassis (1) and connected to the said clutch cam (4) (30) , in which the rod (8) is movable and can take two alternative positions inside the actuator (7) , a back position or a front position, which corresponds to the actuator (7) being switched off or on, respectively.

In one form of embodiment of the invention, the interior of the support structure (20) further includes an angular position sensor (12) housed in the said clutch chassis (13) and connected to the said clutch cam (4) (30) .

Preferably, the aforementioned angular position sensors (2, 12) are Hall effect sensors.

In another form of embodiment of the device (50) representing the present invention, the said pin (10) of the one-way clutch (30) is connected to the lever (11) of the clutch (30) and to the rod (8) of the ECU (40) by means of a slot contained in the lever (11) .

The invention also relates to the use of the joint actuation device (50) for assisting the joint actuation of a human ankle, human knee, human hip or combinations thereof. BRIEF DESCRIPTION OF THE FIGURES

Fig. 1 is a schematic representation of an exploded view of the basic elements of the joint actuation device of the present invention.

Fig. 2 illustrates an exploded view of the joint actuation device of the present invention.

Fig. 3 is a schematic representation of a detailed exploded view of a preferred embodiment of the joint actuation device of the present invention, featuring an additional angular position sensor (12) .

Fig. 4 schematically shows a clutch in the free movement configuration .

Fig. 5 shows the clutch illustrated in Figure 4, in the energy storage and energy release configuration.

Fig. 6 shows a schematic representation of part of an Electronic Command Unit (ECU) .

Fig. 7 shows the architecture of an electronic control board that could be applied to the embodiment represented in Fig. 3.

Fig. 8 shows the angle vs. % gait cycle graph for the ankle and the respective functional states.

Fig. 9 shows the angle vs. % gait cycle for the knee and the respective functional states. Fig. 10 shows the angle vs. % gait cycle graph for the hip and the respective functional states.

DETAILED DESCRIPTION OF THE INVENTION

The present invention relates to a joint actuation device for supporting human locomotion or rehabilitation.

In the context of this description, the term "comprising" should be understood as "including but not limited to". As such, it should not be interpreted as "consisting only of".

In the context of the present invention, "quasipassive" means a system that has some, but not solely, passive elements, such as a spring, and some, but not solely, active sensing and control elements, such as a microprocessor, and which require an external power supply. The device representing the present invention is therefore understood as a quasi-passive device.

In the invention, "functional state" corresponds to the period of the cycle in which the mechanism is operating in a certain function. The different functional states are:

Energy storage functional state (ES) - When the mechanism is coupled and there is controlled storage of elastic potential energy in the spring (3) , deriving from the kinetic energy of the external element. The mechanism is prepared so that in this functional state it can rotate in the opposite direction to that of the intended load, and in this direction, as a result of its unidirectional behaviour, the mechanism does not offer any resistance. Energy release functional state (ER) - When the mechanism is coupled and the elastic potential energy of the spring (3) is released through a torque applied to the external element.

Free movement functional state (EM) - When the mechanism is uncoupled and there is no storage in or release of energy by the spring (3) .

By "angular action" we mean the controlled storage of elastic potential energy in the spring, derived from the kinetic energy of the adjacent external system, and its subsequent release through a torque applied to that same system. In the specific situation of applying angular action to physiological structures, such as the human body, the concept takes the form of "joint action".

An "adjacent system" means the body or system of bodies to which the joint actuation device is attached, and which benefit from its use. For applications involving physiological structures, the device is generally attached to the bodies upstream and downstream of a given joint, including, among others, the ankle, knee and hip joints.

The term "data-linked" means an electronic connection, wired or wireless, between different elements, which allows the transmission of data between the elements thus linked.

Referring to Figs. 1 and 2, the present invention relates to a joint actuation device (50) comprising a support structure (20) that consists of a spring chassis (1) connected to a clutch chassis (13) . The connection between the chassis (1, 13) defines an interior of the support structure (20) which in turn comprises: • a one-way clutch (30) comprising a cam (4) , at least three rollers (9) , a pin (10) and a lever (11) , wherein the cam (4) is coupled to the clutch chassis (13) , the lever (11) is coupled to the cam (4) , and one end of the pin (10) is connected to the lever (11) ;

• an electronic control unit (40) comprising a microprocessor (6) , an actuator (7) , a rod (8) , an inertial measurement unit (5) , and at least one angular position sensor (2) , in which

• the inertial measurement unit (5) , the actuator (7) and the angular position sensor (2) are connected in data communication to the microprocessor (6) ;

• the microprocessor (6) , the inertial measurement unit (5) and the actuator (7) are housed in the aforementioned cam (4) of the clutch ( 30 ) ;

• the angular position sensor (2) is housed in said spring chassis (1) and connected to the said clutch cam (4) (30) ; and

• the rod (8) is housed inside the actuator (7) and connected to the other end of the said pin (10) of the clutch (30) ; and

• at least one spring (3) housed in the spring chassis (1) and connected to the said clutch cam (4) (30) , in which this rod (8) is movable and can take two alternative positions inside the actuator (7) , a back position or a front position, which corresponds to the actuator (7) being switched off or on, respectively. In one embodiment of the invention, the pin (10) is connected to the lever (11) and the rod (8) by means of a slot contained in the lever (11) which controls the position of the at least three rollers (9) , these being preferably housed in a cavity between the outer diameter of the cam (4) , a cavity of the outer diameter of the lever (11) and the interior surface of the clutch chassis (13) .

Depending on the position of the rod (8) , the lever (11) and the at least three rollers (9) have two alternative positions they can take in relation to the cam (4) , consisting respectively of an unlocking position associated with the back position of the rod (8) or a locking position associated with a front position of the rod (8) .

According to the constructive arrangement of the device (50) representing the invention, the active one-way clutch (30) is connected to the actuator (7) which, in turn, is connected to the microprocessor (6) . In operation, when the actuator (7) is switched off, the rod (8) is retracted inside it, so that the cam (4) and lever (11) assembly, and the clutch chassis (13) , rotate freely and independently. When the actuator (7) is actuated, the rod (8) advances and moves the pin (10) through the lever slot (11) , imposing a rotational movement on the latter which moves the rollers to their locking position at the wedge end of the cam cavities (4) .

Surprisingly, it was found that combining an actuator controlled by a microprocessor with a one-way continuous clutch makes it possible to control the clutch's locking start parameters, even in real time, turning it into an active one-way continuous clutch. This minimises the time it takes for the actuator to engage in order to lock, thereby minimising energy consumption. A one-way continuous clutch allows immediate engagement without interrupting torque transmission, unlike discrete clutches .

A one-way continuous clutch with passive control makes it possible to pre-select one position for locking and another for unlocking using a roller positioning lever, which defines the locking and unlocking position .

When the clutch ' s actuation needs depend on the state of the system, clutches with active control are usually used, which allow dynamic control of locking and unlocking, but which require permanent actuation during the locking phase , which implies high energy consumption .

A one-way continuous clutch has the particularity of locking movement in one direction of rotation and, on the other hand, allowing free movement in the opposite direction . In both cases , the one-way continuous clutch works without any need for actuation . Examples of one-way clutches include bicycle transmission systems and ratchet spanners , which, although already known, are not known to be associated with this type of electronically controlled j oint actuation device . As these mechanisms do not have electronic controls for the actuating lever, they are unable to adapt to the dynamics of the movement .

Furthermore , the association of an angular position sensor with the control system makes it possible to identi fy the moment when the clutch is locked; at which point , due to the natural sel f-locking of this type of system, the actuator is no longer needed and can therefore be switched o f f . This procedure is very important for the autonomy of the j oint actuation device as it makes energy consumption residual during the locking phase, unlike what happens in the active systems mentioned above.

In the present invention, the external power supply is used exclusively to power the sensing and control elements and does not supply energy directly to the user. Locking, achieved through the actuation of a low-power actuator, has a short working cycle, lasting only until the start of transmission of energy to the spring. This is highly advantageous as it combines the efficiency of active exoskeletons with the autonomy of passive exoskeletons.

The structure (20) supporting the joint actuation device (50) of the invention makes it possible to establish the connection of the device (50) to the adjacent system, e.g. ankle, knee, or hip, so that the adjacent system benefits from the action of the joint actuation device (50) . The spring chassis (1) and the clutch chassis (13) are concentric to ensure the alignment of the entire joint actuation device and the revolution between them.

The number of springs (3) and their geometry are customised to the type of joint action, the desired stiffness and the expected stiffness changes. Whenever applicable, the use of just one spring (3) is preferred. However, it may be necessary or desirable to use more than one spring (3) depending on the aforementioned factors of typology, stiffness, and expected stiffness changes.

The actuation of the spring (3) is based on the concept of functional states, according to the following principles:

(i) In the functional state of energy storage (ES) , the torsion spring only stores energy in one direction of angular movement. The direction of movement is determined by the orientation of the clutch (30) .

(ii) In the functional state of energy release (ER) , the torsion spring (3) releases energy only in the opposite and contiguous direction to that of loading.

(iii) In the functional state of free movement (EM) , the spring (3) remains undeformed, i.e. with zero elastic potential energy.

The spring (3) reduces the joint torque required by the user to fulfil the same function. To this end, loading occurs in functional states characterised by belonging to the support phase of the respective limb, and in which there is joint torque produced physiologically by muscular action, by the nature of the movement, and by the inertia of the joint, and prior to the functional state in which there is a maximum joint torque required of the joint.

The spring (3) and the hinge operate in parallel: where, Qy is the natural quasi-rigidity of joint j without a spring, kj _S is the torsional spring stiffness of joint j and Qy' is the complementary quasi-rigidity of joint j with spring .

The operation of the joint actuation device (50) in its specific application to the ankle during the gait cycle is shown in Fig. 8. In the controlled dorsal flexion phase, which represents maximum energy absorption, the Energy Storage (ES) state occurs. The next functional state results from the reversal of the movement, characterised by an absolute maximum of stored energy, and the start of the Energy Release (ER) state.

The joint actuation device (50) , in its specific application to the knee during the gait cycle, is shown in Fig. 9. In the controlled flexion phase, which represents a maximum of absorbed energy, there is a state of Energy Storage (ES) . The next functional state is the result of the reversal of the movement and a maximum of stored energy, and the start of the Energy Release (ER) state.

In its specific application to the hip, during the gait cycle, the joint actuation device (50) is shown in Fig. 10. In the controlled extension phase, which represents a maximum of absorbed energy, there is the state of Energy Storage (ES) . The next functional state is the result of the reversal of the movement and presents a maximum of stored energy, and the beginning of the energy release state (ER) .

The clutch chassis (13) fulfils a dual function, since it is an integral element of the support structure (20) , but it is also essential for the operation of the clutch (30) . The clutch (30) is responsible for transmitting the torque produced by the adjacent system, T, to the spring (3) , and vice versa, i.e. when the energy is released, the clutch (30) is responsible for transmitting the spring's elastic potential energy back to the adjacent system, thus producing the desired joint action in the form of torque.

The clutch parameters (30) are scaled on the basis of the external torque T with which it has to operate. Specifically, these parameters are the radius R of the ring (13) , the radius r of the rollers (9) and the height h of the cam cavity surface (4) . These parameters are determined taking into account the value of the angle a of contact between the rollers (9) and the ring (13) , which allows correct locking and unlocking.

In order to ensure it functions correctly, the clutch (30) must have at least 3 rollers (9) in its composition, and the calculation of the effective number of rollers N _max is given by the following equation: where s is the arc of the circle between the angle of intersection of the cam surface (4) and the radius R — r, represented by and which embodies the safety required for the roller not to leave the operating cavity, and the angle with the perpendicular distance 1.5r from the cam to the ring (13) , represented by ? ₂ , which embodies the total safety required for the clutch (30) to release.

The rollers (9) must be arranged on the clutch (30) with equal angular spacing between them, the number of rollers (9) being determined so as to ensure the functional and structural stability of the clutch.

Immediately before the start of the energy storage (ES) function state, the clutch (30) is locked, i.e. the system is coupled and ready to transmit the external torque T to the spring (3) in the specified direction. When the spring (3) is loaded, the system remains locked, i.e. it does not need to be actuated during the storage and release of energy. When the spring (3) returns to zero energy, the clutch (30) unlocks and the system decouples autonomously. In the free movement state (ML) , the clutch (30) is unlocked, so the system is decoupled and there is no transfer of energy to the spring (3) .

The clutch (30) is locked by activating the actuator

(7) and thus advancing the rod (8) , which applies a force to the connecting pin (10) and moves the lever (11) . The rod

(8) transfers the force to the lever (11) via the connecting pin (10) which is housed in the lever slot (11) . The connecting pin (10) moves in the slot of the lever (11) , which allows the former (10) to move to offset and cancel out the difference between the linear path of the rod (8) and the angular path of the lever (11) . The angular displacement of the lever (11) presses on the rollers (9) between the cam (4) and the clutch chassis (14) . When the clutch chassis (13) moves in the same direction as the lever (11) , the clutch (30) locks by crushing the rollers (9) , increasing the friction between the clutch chassis (13) and the cam (4) , thus deforming the spring (3) and consequently storing elastic potential energy in this component.

Once the energy storage in the spring (3) has started, the actuator (7) is no longer activated, but the clutch (30) remains locked autonomously until the spring (3) returns to zero energy. In the free movement state (FM) , the actuator (7) remains unactuated, and the clutch (30) is not locked.

In a preferred embodiment of the invention, illustrated in Fig. 3, the ECU (40) comprises an inertial measurement unit (IMU) (5) , a microprocessor (6) , and two angular position sensors (2, 12) , the microprocessor (6) and the IMU (5) being housed in the cam (4) of the clutch (30) . The ECU (40) is responsible for managing the operation of the entire system. The ECU (40) has a microprocessor (6) that identifies the movement pattern (e.g. the user's walking pattern) and, based on this identification, controls the functional states of the mechanism. The control is based on monitoring the linear acceleration and angular velocity via the IMU (5) and monitoring the spring deformation angle (3) and the total joint angle via two Hall-effect angular position sensors (2, 12) , respectively. These sensors (2, 12) are calibrated at the start of operation, and the angular position sensor (2) measures the deformation of the spring (3) through the difference between the inside diameter of the spring (3) and the outside diameter of the spring (3) , while the angular position sensor (12) measures the complementary angle measured between the clutch chassis (13) and the cam (4) . The angular amplitude is given by the sum of the values of the two angular position sensors (2, 12) .

The microprocessor (6) is responsible for communicating between the IMU (5) and the angular position sensors (2, 12) , from which it receives information on acceleration and angular velocity. It is also responsible for activating and deactivating the actuator (7) , based on the information collected by the IMU (5) . The connection between these components is described in Figure 7.

The active clutch (30) , in the particular situation of use to increase human capacity, only performs one-way locking, which prevents energy from being stored in the spring in the opposite direction to that intended.

The architecture of the ECU (40) , in the particular situation of being made from separate components, can be seen in Fig. 6, which is one example of multiple possible configurations. As an example, the architecture could consist of a printed circuit board, for instance an Arduino Pro mini board with an ATMega328 microprocessor, or equivalent. Likewise, an angular position measurement board, for example the AS5600 sensor. It should also incorporate a UMI, such as the MPU-9250 board. The angular position sensors (2, 12) are Hall effect sensors, so as to preserve precision, and consist of a board with a potentiometer circuit and a magnet, centred with each other and separated by an air gap that can vary between 0.5 mm and 2.5 mm.

The IMU (5) monitors acceleration, angular velocity and operating temperature.

The angle position sensors (2, 12) measure the torsion angle of the spring and the angle that the joint embodies, respectively, without the spring being loaded (3) . The amplitude of the joint is obtained by adding the two angles recorded by the angle position sensors (2, 12) .

The actuator (7) is controlled by the microprocessor (6) on the basis of the monitoring carried out by the angular position sensors (2, 12) and the algorithm defined for the microprocessor (6) . These monitors make it possible to identify the different functional states. Before entering the energy storage functional state, the microprocessor (6) activates the actuator (7) , locking the clutch (30) . When energy storage is identified in the spring (3) , i.e. when deformation of the spring (3) is detected, the microprocessor (6) promotes the end of activation of the actuator (7) . The process is repeated cyclically when a new storage functional state is identified.

The series interconnection of the spring (3) and the clutch (30) , controlled by the ECU (40) , provides added- value functionalities for the joint actuation device (50) , since its assembly allows the actuation work cycle to be significantly reduced. Its duration is strictly necessary for the clutch (30) to lock and, consequently, for the elastic potential energy to be stored in the spring (3) . At this point, the actuator (7) is no longer actuated, and the mechanism works autonomously for the rest of the cycle, thanks to the characteristics of the spring (3) and the clutch ( 30 ) .

In its particular application to the ankle, during the gait cycle, the IMU (5) has the x-axis parallel to the tibia, the y-axis perpendicular to the tibia in the sagittal plane, and the z-axis perpendicular to the sagittal plane, in a medial direction. Thus, the IMU (5) reads the following variables : a_x - centripetal acceleration; a_y - tangential acceleration to the tibia in the sagittal plane; a_z - medial acceleration; w_x - angular velocity of the abduction/adduction; w_y - angular velocity of eversion/ inversion; w_z - angular velocity of dorsif lexion/plantar flexion; T - temperature.

Also in the ankle-specific application, during the gait cycle, the microprocessor (6) collects information from the measurement group made up of the IMU (5) and the angular position sensors (2, 12) . When the monitoring values reach a certain limit, the actuator (7) is activated. The actuator (7) is deactivated when spring loading (3) is detected. For safety reasons, the maximum actuation time has a preset time depending on the joint and movement. When activated, the actuator (7) allows the rod (8) to move forward, thus pushing the connecting pin (10) which transfers the force needed to move the lever (11) (see Figs. 4 and 5) . When the lever (11) is moved, the rollers (9) are pressed between the clutch chassis (13) and the cam (4) . When the clutch frame (4) moves in the opposite direction to the lever (11) , there is no power transmission. When the direction of movement of the clutch frame (13) coincides with that of the lever (11) , there is an increase in the friction forces between the rollers (9) , the clutch frame (13) and the cam (4) . In this way, the assembly (4, 9, 13) moves as a block, allowing kinetic energy to be transformed into elastic potential energy, stored in the spring (3) . When the direction of movement of the clutch frame (13) is reversed, the elastic potential energy is transformed into kinetic energy, which is transmitted to the clutch frame (13) until it is exhausted.

Fig. 4 shows when the mechanism is in the free movement position, i.e. unlocked. In this function, the actuator (7) is not activated, and the rod (8) is retracted. The lever (11) is placed in the free movement position by the force exerted on the connecting pin (10) by the rod (8) . At this stage, the rollers (9) are in the free movement position and there is no friction between the rollers (9) , the clutch chassis (13) , and the cam (4) , thus preventing the transmission of force between the user and the spring (3) .

In the specific application to the ankle, during the gait cycle, the joint actuation device (50) has an axis of rotation in the cam (4) , under which the spring chassis (1) , the clutch chassis (13) , and the lever (11) rotate. The spring (3) is fixed in its internal diameter on the same axis as the cam (4) . The spring chassis (1) and the clutch chassis (13) can be supported by bearings, and they keep the joint actuation device (50) closed by an appropriate fit.

The stiffness of the spring (3) is customisable and is applied to the mechanism with a quick change. To do this, open the spring chassis (1) , remove the existing spring (3) and insert the new spring (3) with the desired stiffness.

In the specific application to the ankle, the joint actuation device (50) can be anchored to the foot and leg in a non-invasive way. As an example, anchoring to the foot is done through a support attached to the sole of the shoe in the heel area, making it possible to absorb the variations resulting from ankle rotation. The leg anchor takes the form of an ergonomic shin guard that goes around the leg. The ergonomic shin guard is rigid at the front, ensuring a suitable interface between the joint actuation device and the biological structure, and it can be adjusted by means of an elasticated strap at the back.

The joint actuation device can be applied to any joint independently. The joint actuation device of the invention allows the natural degrees of freedom of each joint, and can be applied to a single joint or several simultaneously, thereby achieving increased capacity in various segments of the limb. The stiffness is adjusted to the joint and/or the user by changing the torsional spring.

The device (50) of the present invention is very versatile, since its locking/unlocking mechanism can be applied to different types of joints that benefit from articular action. The device (50) of the present invention can be an ankle, knee, or hip joint actuation device.

It should be noted that although the present invention has been described with reference to its preferred forms of embodiment, modifications and alternatives can be made by a person skilled in the art without departing from the scope of the invention, which is defined by the claims.