PROSTHESIS

Abstract

The invention relates to a prosthesis to replace a missing extremity of a living being, which has: one or more prosthetic links driven by actuators, first sensors, which sense a current state ZUS(t) of the prosthesis; an interface to second sensors, which sense biosignals SIGBIO(t) of the living being to control the missing extremity; third sensors for sensing data DUMG(t), which describe a current environment of the prosthesis; a prediction unit, which determines, based on the biosignals SIGBIO(t) and on the state ZUS(t) of the prosthesis and of the data DUMG(t), a model MA(t) of an action A to be executed with the prosthesis and predicts motions Beweg(MA(t)), dependent on the model MA(t), of the prosthetic links for a period of time [t, t+Δt]; an evaluating unit, by which the discrete decision E to replace the action A with another action A′(E) can be determined on the basis of an evaluation of the biosignals SIGBIO(t), of the state ZUS(t), of the data DUMG(t), and of the predicted motions Beweg(MA(t)) in accordance with a specified evaluation scheme, wherein the action A′(E) can define a reflexive motion and/or protective motion of the prosthesis that is autonomously controlled in an open-loop/closed-loop manner, and wherein, if the action A′(E) does not define such a reflexive motion and/or protective motion of the prosthesis that is autonomously controlled in an open-loop/closed-loop manner, the prediction unit determines a model MA′(t) of the action A′ to be performed by the prosthesis and predicts motions Beweg(MA′(t)), dependent on the model MA′(t), of the prosthetic links for a time period [t, t+Δt]; and a control unit, which derives control signals Sig(t), based on the currently valid, predicted motions Beweg(MA(t)) or Beweg(MA′(t)) or based on the reflexive and/or protective motion autonomously controlled in an open-loop/closed-loop manner, for controlling the actuators and controls/regulates the actuators based on the control signals Sig(t).

Description

BACKGROUND

Field

The invention relates to a prosthesis for replacing a missing extremity of a living being, particularly a human being. The prosthesis has prosthetic links driven by actuators.

Related Art

The prostheses driven by actuators available today, for example hand/arm prostheses, are associated with high procurement costs. Moreover, these prostheses are limited in their function.

SUMMARY

The object of the invention is to provide an improved prosthesis to replace a missing extremity of a living being, particularly a human being, which is economical to produce and has expanded functionality.

The invention results from the features of the main claims. Advantageous further embodiments and designs are the subject matter of the dependent claims. Further features, application options, and advantages of the invention result from the following description, and explanation, of example embodiments of the invention, which are represented in the figures.

The invention is achieved with a prosthesis for replacing a missing extremity of a living being, particularly a human being. The proposed prosthesis includes one or more prosthetic links driven by actuators, wherein the proximal prosthetic link has a mechanical interface for fixing the proximal prosthetic link to the living being.

The prosthesis further includes first sensors, which sense a current state ZUS(t) of the prosthesis, particularly a state of contact between the prosthesis and the environment. Advantageously, the first sensors include one or more of the following sensors: Joint sensor(s) for sensing a motor position and/or a motor speed and/or a motor acceleration, output sensor(s) for sensing an output position and/or an output speed and/or an output acceleration, acceleration sensor(s), for example for sensing an acceleration in the joints of the prosthesis or an acceleration in the prosthetic links, the acceleration due to gravity, an acceleration and motion of the mechanical interface for fixing the proximal prosthetic link, a relative acceleration and motion of the prosthesis on the residual limb, force sensor(s) for determining the forces transferred to an environment due to the individual prosthetic links and for determining forces transferred from an environment onto the individual prosthetic links, as well as for determining interaction forces between the prosthesis and the prosthesis wearer, torque sensor(s) for determining the torques transferred to an environment by the individual prosthetic links and for determining torques transferred from an environment to the individual prosthetic links, tactile sensor(s), particularly an artificial skin, for local sensing of acting forces and torques, temperature sensor(s) for sensing temperatures acting upon the prosthesis, humidity sensor(s) for sensing humidity acting upon the prosthesis. All sensors (including individual ones) may also advantageously be used to determine mechanical relations (forces, accelerations, etc.).

The prosthesis further includes an interface to second sensors, which sense biosignals SIG_BIO(t) in the living being for controlling the missing extremity. The second sensors advantageously include one or more electromyography sensors and/or one or more electroencephalography sensors. The individual second sensors are advantageously either applied to the living being or implanted in the living being.

The prosthesis further include third sensors for sensing data D_UMG(t), which describe a current environment of the prosthesis, particularly objects and/or other living beings located in the environment. The third sensors may include one or more of the following sensors: optical sensors, for example camera or video sensors, ultrasound sensors, laser sensors, etc.

The prosthesis further include a prediction unit, which determines a model M_A(t) of an action A to be performed by the prosthesis on the basis of the biosignals SIG_BIO(t) and the state ZUS(t) of the prosthesis and the data D_UMG(t) and predicts motions B_eweg(M_A(t)), dependent on the model M_A(t), of the prosthetic links for a time period [t, t+Δt]. In the simplest case, the prediction takes place for a next period of time. This can be the control cycle or control cycles of the prosthesis. The prediction time period Δt is advantageously selected in a range of from 0.3 to 30 seconds. The prediction time period Δt may advantageously vary depending on the processor load in the prediction unit. Advantageously, the prediction unit is designed and set up such that the sensing of model M_A(t) and/or M_A′(t) is implemented as a learning-capable process, which autonomously learns motions B_eweg(M_A(t))/B_eweg(M_A′(t)) of the prosthetic links based on historical data, for executing an action A/A′ for this. They may also consider predicted/estimated contact and/or other multimodal information.

The prosthesis further includes an evaluating unit, which can determine the discrete decision E to replace the action A with another action A′(E) on the basis of an evaluation of the biosignals SIG_BIO(t), of the state ZUS(t), of the data D_UMG(t). and of the predicted motions B_eweg(M_A(t)) in accordance with a specified evaluation scheme, wherein the action A′(E) can define a reflexive motion and/or protective motion of the prosthesis that is autonomously controlled in an open-loop/closed-loop manner, and wherein, if the action A′(E) does not define such a reflexive motion and/or protective motion of the prosthesis that is autonomously controlled in an open-loop/closed-loop manner, the prediction unit determines a model M_A′(t) of the action A′ to be performed by the prosthesis and predicts motions B_eweg(M_A′(t)), dependent on the model M_A′(t), of the prosthetic links for a time period [t, t+Δt].

The evaluating unit is advantageously designed and set up such that a decision E for an action A′(E), which defines a reflexive motion and/or protective motion of the prosthesis that is autonomously controlled in an open-loop/closed-loop manner independently of the biosignals SIG_BIO(t), is determined on the basis of the current state ZUS(t) and/or of the data D_UMG(t).

The evaluating unit is advantageously designed and set up such that a decision E for an action A′(E), which defines a reflexive motion and/or protective motion of the prosthesis that is autonomously controlled, is then triggered when:

• • a current motion of the prosthetic links deviates from the predicted motion B_eweg(M(t)) or B_eweg(M_A′(t)) of the prosthetic links by more than a specified limit (this limit may be time-variant and/or depend on the state ZUS(t)); and/or
  • a current motion of the prosthetic links deviates from a learned model, which observes the nominal states of the prosthesis and corresponding critical deviations from the nominal states; and/or
  • it is discernible based on the current state ZUS(t) and the data D_UMG(t) that the action A or A′(E) to be executed with the prosthesis was not or will not be executed or was or will be executed in a flawed manner (thereby, for example, a re-grasping can occur when there is a faulty grasping of an object by the prosthesis); and/or
  • an unintentional collision of one or more prosthetic links with an object in the environment has occurred or will occur; and/or
  • a temperature recorded by a temperature sensor of the prosthesis reaches or exceeds a limit; and/or
  • a distal end of the prosthesis has approached an object up to a specified distance or environment envelope (this is used particularly for executing so-called standard skills such as, for example, the (partially) autonomous grasping of an object, the moving of a key on a keyboard, opening of a door, the grasping and bringing to the mouth of a bottle of beer, etc.); and/or
  • the current state ZUS(t) of the prosthesis corresponds to a state to which the following applies: ZUS(t)∉Z_ZUS,erlaubt, wherein Z_ZUS,erlaubt indicates the quantity of all allowed states ZUS(t)-(ZUS(t) may also be derived variables. Thus, any varieties, observing elements, models can be learned with methods of autonomous learning.

The prosthesis finally includes a control unit, which derives control signals Sig(t) for controlling the actuators and controls/regulates the actuators based on the control signals Sig(t), based on the currently valid, predicted motions B_eweg(M_A(t)) or B_eweg(M_A′(t)) or based on the reflexive and/or protective motion autonomously controlled in an open-loop/closed-loop manner.

Through evaluation of the SIG_BIO(t), through use of the listed sensors to sense the current state ZUS(t) of the prosthesis and a state of an environment of the prosthesis, and through model formation and prediction, the proposed prosthesis enables an improved and more natural control of the prosthesis, wherein the prosthesis is capable of performing autonomous motions and actions, particularly reflexive and/or protective autonomous motion.

An advantageous refinement of the prosthesis is characterized in that the control unit has an input interface, via which a user of the prosthesis prompts, by an input, the control unit to stop any sort of motion of the prosthesis. This enables the wearer of the prosthesis to stop an activity of the prosthesis at any time.

An advantageous refinement of the prosthesis is characterized in that the control unit is designed and set up such that a further motion of the prosthesis takes place based on currently sensed biosignals SIG_BIO(t) and/or after an input of a user of the prosthesis by an interface of the control unit after an execution of a reflexive and/or protective autonomous motion, wherein the interface is designed and set up for manual optical and/or acoustic and/or tactile input. Advantageously, the control unit is formed from multiple multimodal control units. It is further advantageous when the control unit is designed and set up such that the control system of the actuators has torque control with friction compensation and/or an impedance regulation and/or a force regulation.

An advantageous refinement of the prosthesis is characterized in that the control unit provides multiple apps (control programs and control parameter sets) for controlling the prosthesis, wherein each app defines an operating mode of the prosthesis, and wherein the control unit has an interface, particularly a wireless interface, to a mobile unit, via which a respective app can be selected and configured for controlling the prosthesis. The mobile unit is advantageously a notebook or a smart phone or a mobile tablet computer.

An advantageous refinement of the prosthesis is characterized in that the control unit is designed and set up such that an automatic adapting of the mechanical active and/or passive impedance, particularly the stiffness of the prosthesis, takes place through a corresponding change in the control signals Sig(t), based on the current state ZUS(t) and the data D_UMG(t).

The proposed prosthesis is advantageously designed as a hand prosthesis, an underarm prosthesis with and without elbows, a complete arm prosthesis, a leg prosthesis, or a prosthesis with an exoskeleton. One, more, or all of the prosthetic links with the proposed prosthesis can be driven by actuators.

An advantageous refinement of the prosthesis is characterized in that a response interface is available, by which feedback on the current state ZUS(t) of the prosthesis and its interaction with the environment is transmitted to the prosthesis wearer. This feedback occurs preferably haptically or through electrical stimulation.

Other advantages, features, and details result from the following description, in which at least one exemplary embodiment is described in detail—optionally with reference to the drawing.

Equivalent, similar, and/or functionally equivalent parts have been given the same reference numbers.

BRIEF DESCRIPTION OF THE DRAWINGS

In the drawings:

FIG. 1 shows a schematic diagram of a proposed prosthesis.

DETAILED DESCRIPTION

FIG. 1 shows a schematic diagram of a proposed prosthesis to replace a missing extremity of a living being, particularly of a human being, which has: one or more prosthetic links driven by actuators 101, wherein the proximal prosthetic link has a mechanical interface for fixing the proximal prosthetic link to the living being; first sensors 102, which sense a current state ZUS(t) of the prosthesis, particularly a state of contact between the prosthesis and an environment; an interface to second sensors 103, which sense biosignals SIG_BIO(t) of the living being to control the missing extremity; third sensors 104 for sensing data D_UMG(t), which describe a current environment of the prosthesis, particularly objects and/or other living beings in the environment; a prediction unit 105, which determines, based on the biosignals SIG_BIO(t) and on the state ZUS(t) of the prosthesis and on the data D_UMG(t), a model M_A(t) of an action A to be executed with the prosthesis and predicts motions B_eweg(M_A(t)), dependent on the model M_A(t), of the prosthetic links for a period of time [t, t+Δt]; an evaluating unit 106 by which the discrete decision E to replace the action A with another action A′(E) can be determined on the basis of an evaluation of the biosignals SIG_BIO(t), of the state ZUS(t), of the data D_UMG(t), and of the predicted motions B_eweg(M_A(t)) in accordance with a specified evaluation scheme, wherein the action A′(E) can define a reflexive motion and/or protective motion of the prosthesis that is autonomously controlled in an open-loop/closed-loop manner, and wherein, if the action A′(E) does not define such a reflexive motion and/or protective motion of the prosthesis that is autonomously controlled in an open-loop/closed-loop manner, the prediction unit 105 determines a model M_A′(t) of the action A′ to be performed by the prosthesis and predicts motions B_eweg(M_A′(t)), dependent on the model M_A′(t), of the prosthetic links for a time period [t, t+Δt]; and a control unit 107, which derives control signals Sig(t) for controlling the actuators and controls/regulates the actuators based on the control signals Sig(t), based on the currently valid, predicted motions B_eweg(M_A(t)) or B_eweg(M_A′(t)) or based on the reflexive and/or protective motion autonomously controlled in an open-loop/closed-loop manner.

Although the invention has been illustrated and explained in more detail by preferred example embodiments, the invention is not limited by the disclosed examples and other variations may be derived by one of ordinary skill in the art without extending beyond the protective scope of the invention. It is thus clear that a plurality of variation options exist. It is likewise clear that example embodiments actually only represent examples, which are not to be interpreted in any manner as a limitation, for example, of the protective scope, the use options, or the configuration of the invention. Rather, the previous description and the description of figures should make one of ordinary skill in the art capable of specifically implementing the example embodiments, wherein one of ordinary skill in the art with knowledge of the disclosed concept of the invention can undertake various changes, for example with respect to the function or the arrangement of individual elements listed in an example embodiment, without going beyond the scope of protection, which is defined by the claims and the legal equivalents thereof such as, for example, more extensive explanations in the description.

LIST OF REFERENCE NUMBERS

• 101 Actuators
• 102 First sensors
• 103 Second sensors
• 104 Third sensors
• 105 Prediction unit
• 106 Evaluating unit
• 107 Control unit

Claims

1. A prosthesis to replace a missing extremity of a living being the prosthesis comprising: one or more prosthetic links driven by actuators, wherein a proximal prosthetic link has a mechanical interface to fix the proximal prosthetic link to the living being;first sensors to sense a current state ZUS(t) of the prosthesis, particularly a state of contact between the prosthesis and an environment;an interface to second sensors, the second sensors to sense biosignals SIGBIO(t) in the living being for controlling the missing extremity;third sensors to sense data DUMG(t) that describe a current environment of the prosthesis, particularly objects and/or other living beings located in the environment;a prediction unit to determine a model MA(t) of an action A to be performed by the prosthesis on the basis of the biosignals SIGBIO(t) and the state ZUS(t) of the prosthesis and the data DUMG(t), the prediction unit further to predict motions Beweg(MA(t)), dependent on the model MA(t), of the prosthetic links for a time period [t, t+Δt];an evaluating unit to determine a discrete decision E to replace the action A with another action A′(E) on the basis of an evaluation of the biosignals SIGBIO(t), of the state ZUS(t), of the data DUMG(t), and of the predicted motions Beweg(MA(t)) in accordance with a specified evaluation scheme, wherein the action A′(E) is enabled to define a reflexive motion and/or protective motion of the prosthesis that is autonomously controlled in an open-loop/closed-loop manner, and wherein, if the action A′(E) does not define such a reflexive motion and/or protective motion of the prosthesis that is autonomously controlled in an open-loop/closed-loop manner, the prediction unit is to determine a model MA′(t) of the action A′ to be performed by the prosthesis and to predict motions Beweg(MA′(t)), dependent on the model MA′(t), of the prosthetic links for a time period [t, t+Δt]; anda control unit to derive control signals Sig(t) that the actuators and control/regulate the actuators based on the control signals Sig(t), based on currently valid, predicted motions Beweg(MA(t)) or Beweg(MA′(t)), or based on the reflexive motion and/or protective motion that is autonomously controlled in an open-loop/closed-loop manner.

2. The prosthesis according to claim 1, wherein the control unit has an input interface via which a user of the prosthesis prompts, by an input, the control unit to stop motion of the prosthesis.

3. The prosthesis according to claim 1, wherein the evaluating unit is further to determine a decision E for an action A′(E), which defines a reflexive motion and/or protective motion of the prosthesis that is autonomously controlled in an open-loop/closed-loop manner independently of the biosignals SIGBIO(t), on the basis of the current state ZUS(t) and/or of the data DUMG(t).

4. The prosthesis according to claim 1, wherein the evaluating unit is further to trigger a decision E for an action A′(E), which defines a reflexive motion and/or protective motion of the prosthesis that is autonomously controlled, when: a current motion of the prosthetic links deviates from the predicted motion Beweg(M(t)) or Beweg(MA′(t)) of the prosthetic links by more than a specified limit; and/ora current motion of the prosthetic links deviates from a learned model, which observes nominal states of the prosthesis and corresponding critical deviations from the nominal states; and/orit is discernible based on the current state ZUS(t) and the data DUMG(t) that the action A or A′(E) to be executed with the prosthesis was not or will not be executed or was or will be executed in a flawed manner; and/oran unintentional collision of one or more prosthetic links with an object in the environment has occurred or will occur; and/ora temperature recorded by a temperature sensor of the prosthesis reaches or exceeds a limit G2; and/ora distal end of the prosthesis has approached an object up to a specified distance or environment envelope; and/orthe current state ZUS(t) of the prosthesis corresponds to a state to which the following applies: ZUS(t)∉ZZUS,erlaubt, wherein ZZUS,erlaubt indicates the quantity of all allowed states ZUS(t).

5. The prosthesis according to claim 1, wherein the control unit is designed and set up such that a further motion of the prosthesis takes place based on currently sensed biosignals SIGBIO(t) and/or after an input of a user of the prosthesis via an interface of the control unit after an execution of a reflexive and/or protective autonomous motion, wherein the interface is designed and set up for manual optical and/or acoustic and/or tactile input.

6. The prosthesis according to claim 1, wherein the first sensors comprise one or more of the following sensors: a joint sensor or joint sensors to sense a motor position, motor speed, and/or a motor acceleration;output sensors;acceleration sensors;force sensors to determine forces transferred to an environment due to individual prosthetic links and for determining forces transferred from the environment onto the individual prosthetic links as well as to determine interaction forces between the prosthesis and the living being wearing the prosthesis;torque sensors to determine torques transferred to the environment due to the individual prosthetic links and to determine torques transferred from the environment onto the individual prosthetic links;tactile sensors to locally sense effective forces and torques;temperature sensors to sense temperatures acting upon the prosthesis; andhumidity sensors to sense humidity acting upon the prosthesis.

7. The prosthesis according to claim 1, wherein the control unit is to provide multiple apps that control the prosthesis, wherein each app defines an operating mode of the prosthesis, and wherein the control unit has an interface to a mobile unit to select a respective app and configure the selected app to control the prosthesis.

8. The prosthesis according to claim 1, wherein the second sensors comprise one or more electromyography sensors and/or one or more electroencephalography sensors, wherein individual sensors are applied to the living being or implanted in the living being.

9. The prosthesis according to claim 1, wherein the control unit is designed and set up such that an automatic adapting of the mechanical active and/or passive impedance takes place through a corresponding change in the control signals Sig(t), based on the current state ZUS(t) and the data DUMG(t).

10. The prosthesis according to claim 1, wherein the prediction unit is designed and set up such that the sensing of model MA(t)/MAt) is implemented as a learning-capable process, which autonomously learns motions Beweg(MA(t))/Beweg(MA′(t)) of the prosthetic links based on historical data, for executing an action A/A′ for this.

11. The prosthesis according to claim 1, wherein the living being is a human being.

12. The prosthesis according to claim 6, wherein tactile sensors are comprised in an artificial skin to locally sense effective forces and torques.

13. The prosthesis according to claim 7, wherein an app comprises a control program and a control parameter set.

14. The prosthesis according to claim 9, wherein the mechanical active and/or passive impedance is a stiffness of the prosthesis.