Wheel-less walking support and rehabilitation device

• US 6,805,677 B2
• Filed: 09/20/2001
• Issued: 10/19/2004
• Est. Priority Date: 09/20/2000
• Status: Active Grant

• Current Assignee: John Castle Simmons
• Original Assignee: John Castle Simmons
• Inventors: Simmons, John Castle
• Primary Examiner(s): Winakur, Eric F.
• Assistant Examiner(s): McCrosky, David J.
• Application Number:
  US09/960,293
• Publication Number:
  US 20030191507A1
• Time in Patent Office: 1,125 Days
• Field of Search: 600/595, 700/245-264, 340/825.19
• US Class Current: 600/595
• CPC Class Codes: A61F 4/00 Methods or devices enabling patients or disabled persons to operate an apparatus or a device not forming part of the body operating or control means for prostheses A61F2/48, A61F2/68
• CPC Class Codes: B25J 9/0006 Exoskeletons, i.e. resembling a human figure
• CPC Class Codes: B25J 9/1689 Teleoperation
• CPC Class Codes: G05B 2219/40136 Stereo audio and vision
• CPC Class Codes: G05B 2219/40144 Force sensation feedback from slave
• CPC Class Codes: G06F 3/011 Arrangements for interaction with the human body, e.g. for user immersion in virtual reality blind teaching G09B21/00

Abstract

An apparatus and process for empowering those in wheelchairs and others with loss of limb control to walk, climb stairs, sit in ordinary chairs, use normal bathroom facilities and stand at normal height with their peers. It also provides a means to speed rehabilitation of the injured while providing superior physical therapy without draining professional resources. The apparatus is worn under normal clothing allowing them to also look normal.

Claims

1. A limb control or assistance device comprising:

a first reporting means for reporting joint dynamics of a first body; and

a first computing means associated with the first body, operatively connected to first reporting means and equipped with a first memory means for storing the joint dynamics over time for one or more body parts as a body executes an action or series of actions, thus capturing the joint dynamics of one or many thus-learned actions or types of learned actions;

a joint control means for controlling or modifying some or all of the joint dynamics of a second body;

a second computing means associated with the second body, operatively connected to joint control means, and equipped with a second memory means for storing and recalling learned actions, to direct the joint control means to replicate or approximate these stored joint dynamics over a time period; and

a learned action selection means, associated with the second computing means, for selecting a previously stored learned action to be replicated or approximated;

wherein

learned action selection means, whether in the form of a user directing the computing means to pick the next of any number of available learned actions, or in the form of a software-driven computer selection of the most currently appropriate of the learned actions, or any combination thereof, can direct the body through a series of potentially complex learned actions by selecting, as needed, learned actions which are then transitioned-to and replicated or approximated.

2. The device of claim 1 wherein joint dynamics comprise joint or body part position, angle, direction, force, contact pressure, speed, attitude, or acceleration, or any combination thereof.

3. The device of claim 1 wherein the first and second computing means, the first and second body, the first and second memory means, or any other elements capable of useful application to both the capture of a learned action or its replication or approximation, or any combination thereof, may be singular in that an element used at capture can also be used to perform similar or related tasks during replication or approximation;

whereby

instead of having redundant elements performing similar functions for the first and second body or even necessarily involving 2 different bodies, any portion of the same equipment elements, or body, or both may be used for both the modeling of a learned action and for its replication or approximation.

4. The device of claim 1 wherein the learned action selection means selects any number of learned actions to be used one after the other.

5. The device of claim 1 further comprising:

a second reporting means for reporting the joint dynamics of the second body to enable the second computing means to automatically select, as necessary, the next learned action, wherein the second computing means, responsive to current data from the second reporting means, selects a learned action to replicate or approximate based on at least a portion of the preferred learned action'"'"'s stored joint dynamics being a closer match to the congruent joint dynamics being currently reported by the second reporting means, or based on a recognized condition or set of conditions suggesting the choice of a specific or specific kind of learned action, or any combination thereof.

6. The device of claim 1 further comprising:

a user interface operatively connected to the second computing means, wherein a user guides the selection of learned actions of the learned action selection means in order to meet currently desired objectives or obstacles through the user interface.

7. The device of claim 1 wherein certain learned actions modeled and stored in the second computing means can be treated as cyclical and thus be repeated any number of times, wherein a potentially complex and repetitive activity can be executed cyclically for as long as desired.

8. The device of claim 1 wherein a learned action can be transitioned to from multiple points in a current learned action to a similar, or congruent or predetermined, or related, or any combination thereof, point in another learned action, wherein the learned action selection means can easily switch between learned actions as needed and can do so from potentially any point in the execution of a current learned action.

9. The device of claim 1 wherein a learned action can be transitioned to from multiple points in a current learned action to a point in another learned action that is chosen because its joint dynamics data, or other data stored with the learned action, or any combination thereof, at tat point in the learned action, is closest to or most favorable to the current actual joint dynamics or other sensor values.

10. The device of claim 1 wherein a learned action can be transitioned to from the current learned action to a point in another learned action using intermediate adjustments between the learned actions.

11. The device of claim 1 further comprising:

a user interface means associated with the second computing means for sending or adjusting instructions to the joint control means, wherein a user equipped with a pointing or other form of user interface device can direct the second computing means to send instructions to the joint control means to direct an action or, as the replication or approximation of a learned action is in progress, adjust the joint control instructions of the currently executing learned action, or any combination thereof.

12. The device of claim 1 wherein the second computing means, during the replication or approximation of a learned action, compares joint dynamics reported by a second reporting means to comparable stored joint dynamics for the current point in the current learned action and, when finding a difference that is beyond an established or calculated norm, or when the joint dynamics currently being reported from the second reporting means fit a recognizable pattern, or any combination thereof, responds by selecting and executing another learned action whose joint dynamics at some point are closer to or are a better match for the joint dynamics currently being reported, or by selecting another learned action known to be a solution for said recognizable pattern or norm, wherein second computing means may sense an undesirable, dangerous, destabilizing, or anticipated condition or any combination thereof and respond to it automatically.

13. The device of claim 1 wherein the second computing means, during the replication or approximation of a learned action, compares joint dynamics reported by a second reporting means to comparable stored joint dynamics for the current point in the current learned action and, when finding a difference beyond an established or calculated norm, or when the joint dynamics currently being reported from the second reporting means fit a recognizable pattern or characteristic, or any combination thereof, adjusts through commands to joint control means to correct.

14. The device of claim 1 wherein the second computing means, during the replication or approximation of a learned action, compares joint dynamics reported by a second reporting means to comparable stored joint dynamics for the current point in the current learned action and, when finding a difference, beyond an established or calculated norm, or when the joint dynamics currently being reported from the second reporting means fits a recognizable pattern or characteristic, or any combination thereof, responds by sending adjusted values to joint control means to offset or minimize differences, or by selecting another learned action, or any combination thereof, wherein second computing means can potentially recognize and respond appropriately to a condition much earlier by comparing these potentially large numbers of often interrelated factors and, potentially, vast amounts of elegantly synchronized multi-joint data allowing the recognition of an out of balance or other condition requiring adjustment before problematic momentums develop.

15. The device of claim 1 further comprising:

a first proximity sensing means operatively connected to the first computing means for measuring over time the proximity to an environmental presence, a distance in a direction to an object or potential impediment, or any combination thereof, from some point or points on or proximal to the body or body-worn equipment, during the capture of a learned action with the resulting data stored as part of or relatable to the learned action; and

a second proximity sensing means operatively connected to the second computing means for measuring proximity to any environmental presence, distance in a direction to an object or potential impediment, or any combination thereof from some point or points on or proximal to the body or body worn equipment, as a learned action is replicated or approximated for the purpose of comparison with the congruent or relatable proximity data stored with the learned action for recognition of differences indicating needed responses, wherein the second computing means may rapidly recognize and respond to an unanticipated placement of objects, terrain, or other environmental presences, or any combination thereof.

16. The device of claim 15 wherein as a learned action is being or is about to be replicated or approximated, computing means, responsive to a potential obstacle, which is sensed as a currently sensed distance response that is not found in the current learned action or is placed differently, either

adjusts the length of the stride of the current learned walking action or selects a learned walking action with a desirable stride length or other applicable stride factors so that the foot doesn'"'"'t encounter the distance-sensed obstacle at an unfavorable point;

or

goes around the obstacle by adjusting the current learned action'"'"'s joint dynamics before sending them to the joint control means to modify the lateral direction or by accomplishing the same direction change by selecting a learned action that turns the body;

or

adjusts the height or placement of the foot or other body parts at points in the current learned walking action, or selects a learned walking action with a desirable clearance or other applicable factors, or any combination thereof.

17. A device according to claim 1 wherein a learned action or any series of learned actions are replicated or approximated to safely and precisely guide a living body though rehabilitation or other exercises.

18. A limb control or assistance device comprising:

a joint control means for the control of joint dynamics of a body wherein joint dynamics comprise joint or body part position, angle, direction, force, contact, pressure, speed, attitude, or acceleration, or any combination thereof;

a reporting means for reporting the joint dynamics of the parts of the body;

a computing means, operatively connected to said joint control means and said reporting means, equipped with memory means for storing the joint dynamics over time for one or multiple body parts simultaneously as a body models a learned action by performing an action or a series of actions, and for recalling those stored joint dynamics and directing the same or a similar joint control means on the same or another joint-control-means- equipped body to replicate those thus-learned actions over some time period; and

a learned action adjustment means for adjusting joint dynamics through commands to joint control means, wherein

learned action adjustment means, responsive to current conditions or objectives, whether the learned action adjustment means is computer software, or a user with a user interface, or any combination thereof, makes it possible to adjust the replication or approximation of a potentially complex learned action with potentially simple adjustments.

19. The device of claim 18 wherein:

for some or all of the equipment that is used both in learning a new learned action and for replicating or approximating it, one piece of equipment may be used for learning a learned action and another piece of equipment can serve the same or similar purpose as it is replicated or approximated.

20. The device of claim 18 wherein the computing means, or a user directing computing means through a user interface, or any combination thereof, may, in response to conditions or current objectives, select another learned action whereupon computing means transitions to and begins executing the newly selected learned action.

21. A method for assisting or directing in part or in full the mobilization of a body comprising, in any order, the steps of:

capturing joint dynamics data, where joint dynamics are defined as joint or body part position, angle, direction, force, contact, pressure, speed, attitude, or acceleration, or any combination thereof, while a learned action or multiple learned actions to be learned are modeled by a body being monitored for some or all of these values by sensors,

storing data for each such learned action in retrievable form,

placing a body in a joint-control means capable of controlling body part positions,

selecting, by decision of computing means or by user selection through a user interface, or any combination thereof, the desired learned actions to recapitulate or approximate,

using the data for each thus-selected learned action to direct the joint-control means to transition, when necessary, from current positions or any learned action already in progress, and replicate or approximate over time those stored values on affected body areas.

22. A method according to claim 21, further comprising:

directing changes to one or more instructions to the joint control means associated with a learned action responsive to a user interface, whereby the user can modify the body actions.

23. A method according to claim 21, further comprising the step of:

making offsetting, correcting, or other calculated adjustments through instructions to the joint control means responsive to current sensor readings when sensor values returned are different from comparable stored values or when the difference exceeds a tolerance.

24. A method according to claim 21, further comprising the steps of:

recognizing a needed adjustment to joint or body dynamics by comparing current data from one or more attitude sensors against congruent or comparable data that was captured and stored when the current learned action was modeled, to locate differences that are beyond a calculated or otherwise established value or that fit a pattern; and

directing any needed error offsets or otherwise-calculated adjustments through commands to joint control means, or by transitioning to another learned behavior whose stored values are closer to or more ideal for the current sensor values, whereby

a body is automatically balanced even when replicating or approximating a potentially complex learned action.

25. A method according to claim 21, further comprising the step of:

repeating a learned action in a cyclical manner.

26. A method according to claim 21, further comprising:

managing transitions from one learned action to another by selecting the point at which to leave the current learned action and the point in the next learned action at which to begin based on the two points being congruent or comparable in their respective sequences, or either of both of the points being predetermined as good transition points, or the points having similar or comparable joint dynamics, or any combination thereof.

27. A method according to claim 21, further comprising the steps of:

recording distances, or vectors, or both from body parts to objects or other environmental presences using sensors on or adequately proximal to the body during the capture of a learned action in a form relatable to points in the progress of the learned action;

sensing, as the learned action is replicated or approximated, using the same or other comparable sensors, the current distances or vectors or both between the body parts and objects or other environmental presences;

comparing congruent or comparable data elements to recognize unplanned presences, absences, or changes in location, or any combination thereon; and

responding, when a difference or change exceeds a tolerance, by selecting and transitioning to a learned action whose stored values are closer to or more ideal for the current sensor values, or by adjustment commands to joint control means, or any combination thereof.

28. A method according to claim 21, further comprising the steps of:

identifying, by a computer program, or a user interface means, or by any combination thereof, a currently desired direction of travel as a learned action is being or is about to be replicated or approximated,

comparing current sensor values for distance, or direction, or any combination thereof, from a plurality of body parts to objects or other environmental presences, against congruent or comparable sensor values that were captured as the behavior was being modeled, and

responding by transitioning as necessary to a learned action, or by adjusting the joint dynamics through commands to the joint control means, or by any combination thereof,

to bring the current joint dynamics closer to or more ideal for some or all of the learned behavior data and the desired direction whereby;

unanticipated objects or terrain may be automatically accommodated.

29. A method according to claim 21 further comprising the steps of:

recording distances, or vectors, or both from body parts to objects or other environmental presences using sensors on or adequately proximal to the body during the capture of a learned action,

sensing, as the learned action is replicated, using the same or other comparable sensors, the current distances or vectors or both between the body parts and objects or other environmental presences,

comparing congruent or comparable data elements in the recorded and currently-sensed data to recognize unplanned presences, absences, or changes in location, or any combination thereof, and

responding by transitioning as necessary to a learned action, or by adjusting the joint dynamics through commands to the joint control means, or by any combination thereof, to modify the length of a walking stride, or the height or placement of the foot or other body parts, or the direction of travel, or other body part direction or placement, or any combination thereof, wherein the body favorably addresses or avoids altogether an unanticipated presence.

30. A method according to claim 21 further comprising the steps of:

gathering, during the execution of a learned action, a plurality of distance-to-the-nearest-object data from a plurality of distance sensors which effectively sweep or scan an area,

locating, in the matrix defined by the data, lines or patterns associated with the predictable and sharply retreating or squared-off distance pattern of a line associated with the top or edge of a step, curb or similar obstacle, or another recognizable scanned distance pattern associated with an object or obstacle, or any combination thereof, to identify the location and placement of an object or obstacle; and

modifying the length of the stride, the body'"'"'s direction with respect to the upcoming presence, foot elevation, other body part position characteristics, or any combination thereof, by adjusting some of the instructions to the joint-control apparatus, or by transitioning to another learned action, or any combination thereof, whereby

the body is caused to address the step, curb, hole, or other potential obstacles in a position already conducive to a safe negotiation.

31. A method according to claim 21 further comprising the steps of:

capturing, while a learned behavior is being modeled, information from sensors relating to the distance, or direction, or both, from locations on the body to objects or other environmental presences;

storing this sensor data in a form relatable to points in the progress of the learned behavior;

comparing, as a learned behavior is being replicated or approximated, current sensor data with congruent or comparable stored sensor data; and

modifying, when a difference exists or exceeds a tolerance, the position of the affected body part or parts so that the data from current distance sensors, direction sensors, or any combination thereof more closely match or are more ideal for the comparable stored sensor data, wherein

the body can be automatically directed to interact more precisely with objects.

32. A method according to claim 21 further comprising the steps of:

modifying, during the replication or approximation of a learned action and assisted by a pointing device or other user interface, the movement or positioning of a body part wherein a user can guide a body part to a more precise or desirable location, or have it arrive at a position at a more ideal time, or any combination thereof.

33. A method according to claim 21 further comprising the steps of:

recognizing a potentially unstable body position by the user looking for potential problems and alerting computing means though a user interface, or

by computing means comparing current sensor values with known or calculated norms for stability, or by comparing current sensor values with comparable or congruent sensor values captured when the learned behavior was modeled to recognize a significant difference,

transitioning to a stabilizing position or learned action whereby a fall or other danger may be averted.