 |
Alan Jennings
PhD Candidate at University of Dayton, Graduating 2012
Research Assistant at the Air Force Institute of Technology |
|
|
Alan Jennings
PhD Candidate at University of Dayton, Graduating 2012
Research Assistant at the Air Force Institute of Technology |
University of Dayton: IEEE Student Branch
Abstract
LabVIEW is a graphical programming language. Rather than executing
code line by line, functions are wired together and whatever function
has all its inputs can evaluate. LabVIEW has traditionally had major market share for test and
measurement control, but has continued to make gains into machine
vision, analysis and simulation. It is used by major companies such as
GE, Procter & Gamble, US military, Google, Motoman, Boeing, Microsoft,
Panasonic, and more. Monster list nearly 300 jobs with LabVIEW
programming specifically as a desired skill. Frequently, jobs in the
Dayton area will also pop up.
LabVIEW a much different programming style that can result in a
steep learning curve. It will be compared and contrasted with
traditional and other graphical programming, so participants can
build on existing skills. Basic examples will be built and
demonstrated to show flow of execution, advantages of the graphical
layout and specific challenges to programming.
(27th of March 2012)
2011 Great Midwestern Space Grant Regional Meeting
Abstract
People are not initially competent, but can become experts in almost any task.
However, many optimizations and artificial intelligence techniques use a finite resolution that prevents “life-long” learning.
This work presents a general purpose waveform shaping algorithm that continues to increase resolution as experience increases.
The waveforms can represent motions, such as jumping where the height of the jump is controlled.
A learned response for any output value comes by interpolating optimized waveforms.
Performance is measured by 1) output accuracy 2) cost compared to its lower limit and 3) number of trials required.
Comparison is made to bootstrapping a direct optimization on a mathematical system.
Results on a physical system are also presented.
(4th of October 2011)
Abstract
Finding optimal inputs for a multiple input, single output
system is taxing for an system operator. This work presents a
population-based optimization to create sets of functions to
approximate a locally optimal input as an operator selects an
output. Output and cost functions are modeled by neural
networks. Neural network gradients are used to optimize a
population of agents by minimizing the cost for the agent's
current output. When an agent reaches an optimal input for its
current output, additional agents are generated to step in the
output gradient directions. The agent then settles to the local
optimum for the new output value. The set of associated optimal
points forms a inverse function, via spline interpolation, from
a desired output to an optimal input. In this manner, a locally
optimal function is created for each settled agent. These
functions are naturally clustered in input and output spaces
allowing for a continuous optimal function. The best cluster
over the anticipated range of desired outputs can be chosen and
the process optimized on-the-fly to respond to different set
points. Results are shown for a diverse set of functions.
(6th of June 2011)
Abstract
Optimal control problems are challenging to solve even on
simple systems. Few problems afford analytical solutions
because of the boundary valued differential equations. Kinetic
systems, such as robotic linkages, can be complicated
to solve because of nonlinearities and degrees-of-freedom.
A numeric control optimization software, DIDO, is coupled
to a numeric kinetic solver, SimMechanics, within
MATLAB. The kinetic model is created directly from a
solid model assembly eliminating human errors. A pendulum
with control saturation is tested to validate satisfaction
of theoretical conditions (< 10% optimality residuals,
typically < 5%). The numeric method is contrasted to
a linear-quadratic-regulator (LQR) and the optimal linear
state transfer. A four degree-of-freedom, arm robot pick-and-
place command is also optimized and realizes a 50%
decrease in energy used over a ramp to constant velocity
maneuver. This coupling obtains near optimal solutions
without intense, model specific analysis.
(2nd of Nov 2010)