Motors: expansible chamber type – Plural relatively movable or rigidly interconnected working... – Single valve for relatively movable working members driving...
Reexamination Certificate
2002-07-10
2004-02-03
Look, Edward K. (Department: 3745)
Motors: expansible chamber type
Plural relatively movable or rigidly interconnected working...
Single valve for relatively movable working members driving...
C092S092000, C092S137000, C901S021000, C901S022000
Reexamination Certificate
active
06684754
ABSTRACT:
FEDERALLY SPONSORED RESEARCH
Not Applicable
SEQUENCE LISTING OR PROGRAM
Not Applicable
BACKGROUND
1. Field of Invention
The present invention relates to mechanical actuators. More specifically, the present invention relates to artificial muscle analogs which do not rely on electromagnetism for their motive force. Certain aspects of the present invention relate to robust, low-cost exoskeletal limbs powered by such muscle analogs, which have applications in many fields including robotics or prosthetics. A final aspect of the present invention relates to a particularly simple means of controlling such exoskeletal limbs, using a novel electro-pneumatic feedback loop.
2. Description of Prior Art
Historically, machines have been invented using an almost endless array of different mechanical methods to induce physical movement.
In recent years, however, most such machines have been based on some form of electromagnetic force, especially electrical motors. In some cases, this has been due more to the overwhelming prevalence of electricity and especially the electrical motor as a motive device, rather than the inherent superiority of this approach for the particular problem. Although an electrical motor is arguably the best solution for many problems, especially those requiring rotary motion, other situations exist where alternate approaches have inherent advantages. Yet relatively little research has been done into other motive methods, due in part to the tremendous popularity of the electrical motor.
One situation where an electromagnetic actuator is not particularly ideal is when a relatively-slow, controlled but powerful linear actuation is desired, especially when the desired machine must copy biological motion. For example, consider what is needed to build a mechanical limb that mimics a human arm. First limiting consideration only to the major biceps and triceps muscles used for flexion and extension of the arm, respectively, one requires a pair of “muscle analogs” or artificial muscles which can each contract with hundreds or thousands of kilograms of force, yet which weigh only a few kilograms. The muscle analogs should each be capable of pulling a “tendon analog” or cable through several centimeters of travel. Ideally, the muscle analogs should be adjustable in their “pull” throughout the range of motion. A feedback mechanism should be available so that the mechanical limb can be held at any position throughout the range of motion. And to best mimic biological muscles, the available pulling force on the cable should be greatest when the muscle analog is extended, falling as the muscle contracts; this tendency offsets the reduced mechanical advantage inherent in biological jointed limbs at extension.
The characteristics of a typical electrical motor are not well suited for such an application. For a variety of reasons, a power-conversion assembly is generally required to convert the output power of the motor to a usable form. First, the rotary motion is not desired, and must be converted to a linear motion through the addition of further hardware such as gears or threaded shafts. Also, a typical motor has optimal performance at a higher RPM than is desirable for such an application, so further speed-reducing gearing is also required. And since a powerful actuation is desired, the gearing used to reduce the speed and linearize the motion must efficiently take advantage of the mechanical advantage involved in such a speed reduction. Finally, an electrical motor does not exert force unless it is drawing power; so if it is desired to maintain an actuator in a particular position once it is moved there, some further hardware in the form of a braking means is also required. And the weight of this overcomplicated solution, including the electrical motor, the associated power-conversion assembly, and braking means, can be substantial. Clearly, there are inherent disadvantages in using electrical motors for such applications.
Specialized motors, such as stepper motors, are sometimes used for such applications, since they can generally run at slower speeds, closer to the desired speed of such actuators. But these are even more complicated and expensive to make than standard motors. And stepper motors are generally a less powerful form of electric motor; so if the stepper motor is run at a slow speed without speed-reduction/force-increasing gearing, a motor large enough to develop sufficient force will generally be even heavier than a more-typical motor with its associated gearing.
Solenoids would seem to be a better fit than motors for use as such a linear actuator, since they are inherently linear in function. A solenoid is an electromagnetic coil which pulls a ferrous core or piston into the coil's center when current is allowed to flow. However, solenoids are also relatively heavy and inefficient for a given strength, and again they require power to hold their position. And solenoids tend to be “digital” in nature, since the pull of the solenoid actually increases as the piston is pulled in. The maximum strength of the pull is at the wrong end of the motion, resulting in biologically-unnatural jerky motions which tend to accelerate toward an abrupt stop. And, using a typical solenoid, there is no good way to stop the motion partway, or lock the solenoid in a partially-closed position.
With these inherent problems using electromagnetic force, alternate approaches have been investigated and indeed are used today in certain instances of linear actuation.
For example, one novel approach uses “memory-wire” which changes length and/or shape when a current is passed through it. Such muscle analogs are extremely simple, light and small. However, this technology is still primarily a lab curiosity, since the available force is very small and the power efficiency is quite low.
Other non-electromagnetic approaches in the prior art have been more successful. For example, hydraulic cylinders enjoy a dominant role where the linear actuation desired must be particularly powerful. A hydraulic cylinder is a piston/cylinder arrangement wherein a piston is forced out when a essentially-incompressible pressurized fluid is allowed to enter a cylinder chamber when a control valve is opened. Double acting hydraulic cylinders are also available, wherein two opposing chambers are present so that the piston can be forced out or pulled in, in order to achieve the desired motion, a user opens the appropriate control valves to pressurize one chamber while simultaneously venting the opposing chamber to an unpressurized fluid reservoir. In a hydraulic cylinder, the available force is essentially constant throughout the range of motion, dependent only on the pressure of the fluid and the cross-sectional area of the piston. And since the fluid is incompressible, hydraulic cylinders lock the actuator solidly in position when all the appropriate valves are closed. However, hydraulic cylinders are generally quite slow and heavy, and are used only when the desired motion need not be particularly fast.
Analogously, pneumatic cylinders are sometimes used, with compressed air or other gases replacing the essentially incompressible but comparatively viscous hydraulic fluid in a hydraulic cylinder. Such pneumatic cylinders generally actuate much faster than comparable hydraulic cylinders. Double acting pneumatic cylinders are also available, able to exert force both during extension and retraction by pressurizing one chamber and venting the opposing chamber (probably to the atmosphere if compressed air is used). And, like hydraulic cylinders, the available force is essentially constant through the range of motion, dependent only on the gas pressure and the piston cross section. However, since the fluid now used is a compressed gas which is not incompressible in any sense, the actuator does not truly “lock” in place when the valves are closed. Instead, the pneumatic cylinder acts more like a spring, with an increasing force opposing deviation from the desired actuator position. This is actually more akin to biological mu
Leslie Michael
Look Edward K.
LandOfFree
Pneumatic muscle analogs for exoskeletal robotic limbs and... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Pneumatic muscle analogs for exoskeletal robotic limbs and..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Pneumatic muscle analogs for exoskeletal robotic limbs and... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3348676