|
|
|
|
| Research | Component Technologies |
| |
| Prime Mover (Switched Reluctance Motor) |
Over the past decade, RRG has come up with many innovative switched reluctance motor designs for the
varying types of actuator requirements. The below are six different concepts that were generated over
the years. With our experience in SRM design the focus of our research is now on Design Synthesis which
involves obtaining a complete parametric description of the motor (which includes amongst others the
torque model, the thermal model and the acoustic model) and then using rules of thumb and algebraic
elimination techniques to quickly arrive at a design solution.
 |
 |
 |
 |
 |
 |
|
Figure 1: Conceptual Design of Switched Reluctance Motors
|
|
|
| Gear Transmission (Hypocyclic Gear Train) |
The demanding requirements for electromechanical actuators include lower volume/weight, higher torque
capacity (or higher torque density), higher stiffness (for higher precision and shock resistance), lower
input inertia (for quicker speed changes and stops/starts), higher efficiency (minimum power loss, less
heat generation), minimum backlash, minimum transmission error, and lower noise/vibration. Not
surprisingly, the most crucial component to satisfy these requirements is the gear transmission.
RRG has developed the actuator gear transmission technology for 15 years. The development includes a
complete parametrical description of 3K Paradox epicyclic gear train. The parametric model consisted
of 226 parameters and 139 constraints initially, and 87 parameters were successfully eliminated from
the mathematical model. This model reduction and solution techniques were effectively applied to design
a series of epicyclic gear trains providing the range of reduction ratio from 10:1 to 600:1.
The latest gear train under investigation is Hypocyclic Gear Train (HGT). This unique device has been
proposed in the EMAA development as a principal component. It is believed to “be more compact,
exhibit lower inertia, be simpler and more easily assembled, be of lower cost, and provide higher
stiffness than anything developed before.” EM actuators with the HGT should see use in a numerous
fields of applications. They include the all-electric ship for the Navy, more-electric aircraft for
Air Force, future combat systems for Army, intelligent machines, space robotics, and nuclear systems.
|
|
| Release Mechanism |
Actuator release mechanism plays a vital role in the successful operation of any torque/force summing
fault tolerant actuator system wherein it is utilized to quickly decouple a failed actuator unit from
the system. The mechanism is mounted at the output end of the actuator and typically acts like a clutch,
the fundamental difference being its duty cycle. While a clutch is designed to disengage and engage the
power train numerous times during its operational life, the release mechanism is meant solely for a single
disengagement in the event of failure. Conventional clutches, on account of their high duty cycle, have a
complex arrangement, are bulky, heavy, slow and have high inertia. As a result they are not ideally
suited to be used as release mechanisms.
The RRG has been actively involved in the development of the architecture for release mechanisms. Three
different architectures have been developed and these include actuation using Shape Memory Alloy (SMA),
Energetic Materials and Pyrotechnics. Among these different actuation methods, pyrotechnics offers the
most advantageous set of features to arrive at a mechanism that is light weight, simple construction,
compact, fast acting, stiff and has a high torque density. Pyrotechnic actuated release mechanisms are
the current focus of research.
|
|
Figure 2: SMA Actuated Release Mechanism
|
|
|
| Sensors |
Typical robot actuators utilize a very small number of sensors, sometimes only a single position sensor,
and assume a conservative linear model for servo-control. Actuators, however, are highly complex
electromechanical systems.The actuator model is extremely nonlinear and parametrically coupled.
One of the key features of the Intelligent Actuator is its expanded sensor suite. The Intelligent
Actuator will utilize an array of ten or more advanced sensors to provide the actuator controller with
information about the state of the actuator.This additional information will provide the opportunity for
superior actuator operation and will enable the implementation of additional features such as
multi-criteria decision making, operational fault tolerance, condition-based maintenance, actuator
resource management, and metrology.The 10 parameters that are proposed for inclusion into the
Intelligent Actuator architecture are position, velocity, acceleration, torque, current, voltage,
magnetic field, temperature, vibration, and noise.
Justification of the Ten Sensor Architecture:
-
Multi-criteria decision making: Multiple sensors will provide additional parameters to create meaningful criteria to allow for decision making based on user set priorities.
- Operational fault tolerance: The additional sensor information will be used to detect, identify, determine the severity, and recover from a wide range of failure modes.
- Condition-based maintenance: The additional sensor information will also be processed to detect conditions and trends that represent a pending failure before it becomes critical to system operation.
- Actuator resource management: The information from multiple sensors will specify the operating state of the actuator. This will be compared to the charted performance envelope to allow higher performance.
- Metrology: Measurement of critical actuator and system parameters is required to achieve the high end effecter position accuracy required by highly demanding tasks.
|
| Interface |
We are currently working to design modular robot interfaces with a focus on high connection accuracy.
The concepts of reconfigurability and modularity call for self-contained individual modules that are
interchangeable. Also, changing out a module or reconfiguring a system must be an efficient process
for reconfigurability to be a worthwhile expense. The interface is of critical importance, because the
connection must be simple, yet highly accurate. If sufficient connection accuracy can be guaranteed,
it becomes possible to predict the system geometry without the need of further calibration. Our goal
is to design an interface where the accuracy can be determined analytically based on geometry and
manufacturing tolerances.
The most important part of the analysis of interface design is the relationship between the magnitudes
of the deformations in the interface and that of the manufacturing tolerances. As we understand the
problem now, tolerances are not critical when deformations are at least an order of magnitude larger
than tolerances. However, when the two are roughly equal, tolerances can become extremely critical to
the connection accuracy. The question we hope to answer is, how close can tolerances come to the size
of the deformations before tolerances become important? Secondly, when does it make sense to pay for
improved tolerances, and when will it be a waste of money because it no longer improves accuracy?
The RRG interface, which employs 16 tooth pairs in contact, has high, controlled deformations relative
to the manufacturing tolerances, which allows deformations to dominate and tolerance to be a non-critical
factor. An analytical simulation has been written to determine the expected connection accuracy based
on geometry, material properties, and tolerances
|
|
| Publications |
|
|
|
|
