Stair ascent and descent control for powered lower limb devices

• US 10,441,439 B2
• Filed: 03/15/2013
• Issued: 10/15/2019
• Est. Priority Date: 05/15/2012
• Status: Active Grant

• Current Assignee: Vanderbilt University
• Original Assignee: Vanderbilt University
• Inventors: Goldfarb, Michael, Lawson, Brian, Varol, Atakan Huseyin
• Primary Examiner(s): Sweet, Thomas
• Assistant Examiner(s): Bahena, Christie L
• Application Number:
  US13/832,431
• Publication Number:
  US 20130310949A1
• Time in Patent Office: 2,405 Days
• Field of Search: 623 24, 623 27, 623 39- 40, 623 47, 600595, 901 1, 602 23, 602 26, 602 27
• CPC Class Codes: A61F 2/60 Artificial legs or feet or parts thereof
• CPC Class Codes: A61F 2/64 Knee joints
• CPC Class Codes: A61F 2/6607 Ankle joints
• CPC Class Codes: A61F 2/70 electrical
• CPC Class Codes: A61F 2002/607 Lower legs
• CPC Class Codes: A61F 2002/701 operated by electrically controlled means, e.g. solenoids or torque motors
• CPC Class Codes: A61F 2002/704 computer-controlled, e.g. robotic control
• CPC Class Codes: A61F 2002/7625 for measuring angular position
• CPC Class Codes: A61F 2002/7635 for measuring force, pressure or mechanical tension

Abstract

Systems and methods of operating a lower limb device having at least a powered joint are provided. A method includes configuring the device to a first state in a finite state model for a current activity mode including a stair ascent mode or a stair descent mode. The method also includes, based on real-time sensor information, transitioning the device between different states in the finite state model when pre-defined criteria for transitioning among the different states are met. In the method, the finite state model for stair ascent includes lifting and swing phases, where the lifting phase includes a powered knee extension and a powered ankle push-off. The finite state model for stair descent includes yielding and swing states, where the swing states include providing a powered plantarflexion of the powered ankle joint and the yielding states include providing a resistive and passive plantarflexion of the powered ankle joint.

Claims

1. A lower limb device comprising:

a powered knee joint with a motor for providing an output torque at the powered knee joint;

a powered ankle joint with a motor for providing an output torque at the powered ankle joint;

a plurality of sensors for generating real-time sensor information for the lower limb device, wherein at least one of the plurality of sensors comprises an inertial measurement unit and a position sensor for angle measurements of the powered knee joint;

at least one processor communicatively coupled to the plurality of sensors, the powered knee joint and the powered ankle joint; and

a computer-readable medium comprising stored instructions for causing the at least one processor to perform the steps comprising;

selecting one of a plurality of phases of a finite state model defining a stair ascent controller to yield a current phase;

operating the motor of the powered knee joint and the motor of the powered ankle joint according to the current phase,

based on the real-time sensor information for the lower limb device and a transition trigger for the current phase, transitioning from a current phase to a subsequent phase in the finite state model,

computing an estimate of an orientation with respect to gravity of a thigh associated with the powered knee joint based on the real-time sensor information from the at least one inertial measurement unit, and

repeating the operating, the transitioning, and the computing until a criteria for exiting the stair ascent controller is met, wherein the criteria is met when the estimate indicates a presence of thigh flexion for a period of time.

2. The lower limb device of claim 1, the computer-readable medium further comprising stored instructions for causing the at least one processor to perform, prior to the selecting, the steps of:

computing an estimate of an orientation with respect to gravity of a thigh associated with the powered knee joint based on the real-time sensor information from the at least one inertial measurement unit and the position sensor; and

initiating the stair ascent mode when the estimate of the orientation with respect to gravity for the thigh associated with the powered knee joint indicates thigh extension for a period of time.

3. The lower limb device of claim 1, wherein the operating of the motor of the powered knee joint and the motor of the powered ankle joint according to the current phase comprises performing, for each one of the powered knee joint and the powered ankle joint:

identifying a stiffness coefficient, a damping coefficient, and an equilibrium angle pre-associated with the current phase; and

controlling the motor to cause the output torque according a pre-defined relationship between the output torque, an angular position, and an angular velocity that is based on the real-time sensor information, the stiffness coefficient, the damping coefficient, and the equilibrium angle.

4. The lower limb device of claim 1,

wherein the finite state model comprises a stance phase, a push off phase, a swing extension phase, a swing flexion phase and a pre-landing phase,

wherein the transition trigger from the stance phase to the push off phase is detecting that respective angles of the powered knee joint and the powered ankle joint meet a straightening criteria,

wherein the transition trigger from the swing flexion phase to the swing extension phase is detecting that an angular velocity for the powered knee joint inflects, and

wherein the transition trigger from the swing extension phase to the stance phase is detecting that the angle of the powered knee joint during the knee flexion meets a criteria for stance.

5. The lower limb device of claim 1, wherein the estimate is further based on at least one of:

an estimation of a location of a segment of the lower limb device or a user through the at least one inertial measurement unit;

a joint angle or angular velocity measurements from at least one of the plurality of sensors;

a measurement of a degree of load or moment bearing in the lower limb device from a load sensor in the plurality of sensors;

or

input received from a sensor.

6. A lower limb device comprising:

a powered knee joint with a motor for providing an output torque at the powered knee joint;

a powered ankle joint with a motor for providing an output torque at the powered ankle joint;

a plurality of sensors for generating real-time sensor information for the lower limb device, wherein at least one of the plurality of sensors comprises an inertial measurement unit and a position sensor for angle measurements of the powered knee joint;

at least one processor communicatively coupled to the plurality of sensors, the powered knee joint and the powered ankle joint; and

a computer-readable medium comprising stored instructions for causing the at least one processor to perform the steps comprising;

selecting one of a plurality of states of a finite state model defining a stair descent controller to yield a current state;

operating the motor of the powered knee joint and the motor of the powered ankle joint according to the current state,

based on the real-time sensor information for the lower limb device and a transition trigger for the current state, transitioning from a current state to a subsequent state in the finite state model,

computing an estimate of an orientation with respect to gravity of a thigh associated with the powered knee joint based on the real-time sensor information from the at least one inertial measurement unit, and

repeating the operating, the transitioning, and the computing until a criteria for exiting the stair descent controller is met, wherein the criteria is met when the estimate indicates an absence of thigh flexion for a period of time.

7. The lower limb device of claim 6, the computer-readable medium further comprising stored instructions for causing the at least one processor to perform, prior to the selecting, the steps of:

computing an estimate of an orientation with respect to gravity of a thigh associated with the powered knee joint based on the real-time sensor information from the at least one inertial measurement unit and the position sensor; and

initiating the stair descent controller when the estimate of the orientation with respect to gravity for the thigh associated with the powered knee joint indicates thigh flexion for a period of time.

8. The lower limb device of claim 6, wherein the operating of the motor of the powered knee joint and the motor of the powered ankle joint according to the current state comprises performing, for each of the powered knee joint and the powered ankle joint:

identifying a stiffness coefficient, a damping coefficient, and an equilibrium angle pre-associated with the current phase; and

controlling the motor to cause the output torque to a pre-defined relationship between the output torque, an angular position, and an angular velocity that is based on the real-time sensor information, the stiffness coefficient, the damping coefficient, and the equilibrium angle.

9. The lower limb device of claim 6, wherein the finite state model comprises a swing extension phase, a forefoot strike phase, a stance knee flexion phase, and a swing flexion phase and wherein the plurality of sensors further comprises a load sensor,

wherein the transition trigger from the swing extension phase to the forefoot strike phase is detecting, by the load sensor, that a load on the lower limb device exceeds a threshold,

wherein the transition trigger from the forefoot strike phase to the stance knee flexion phase is detecting that an angle of the powered ankle joint meets a threshold phase,

wherein the transition trigger from the stance knee flexion phase to the swing flexion phase is detecting, by the load sensor, that the load on the lower limb device falls below the threshold, and

wherein the transition trigger from the swing flexion phase to the swing extension phase is detecting that an angular velocity for the powered knee joint inflects.

10. The lower limb device of claim 6, wherein the estimate is further based on at least one of:

an estimation of a location of a segment of the lower limb device or a user through the at least one inertial measurement unit;

a joint angle or angular velocity measurements from at least one of the plurality of sensors;

a measurement of a degree of load or moment bearing in the lower limb device from a load sensor in the plurality of sensors, wherein the measurement of a degree of load is provided by a load sensor;

or

input received from a sensor.