Balloons/Aerobots for Planetary Exploration
Task Objective
|
Zero Pressure |
|
Super-pressure -- Mars (artist's rendition) |
|
Zero-pressure -- Titan (artist's rendition) |
|
Super-pressure (pumpkin) (artist's rendition) |
|
IR Mongolfier -- Jupiter (artist's rendition) |
|
Zero-pressure -- Venus (artist's rendition) |
The exploration of the solar system has advanced from fly-bys and
static landers to sophisticated orbiters, landers, rovers, and
other advanced exploratory vehicles. The possibility of
using robotically-controlled balloons (also called "Aerobots") in
missions for science exploration has tremendous promise for two
reasons: (1) The nearness to the surface allows the use of relatively
inexpensive imaging equipment, and (2) The mobility of the balloon
allows us to explore a large area.
The goal of this research is to show how moderate vertical mobility
can be used with improved wind models to achieve the horizontal
mobility necessary to perform target-focused science missions with
formations of balloons.
The near-term benefits of this research are quite simple: No one knows
how to fly a single aerobot or a formation of aerobots to move from one
place to another---much less how to maintain a formation during flight.
This research will allow consideration of formations of aerobots for
focused planetary science missions. Formations of balloons will
increase usable science
data returned as well as allow for some redundancy of science
exploration vehicles.
The long-term goal of this line of research is to convincingly
demonstrate that aerobots or formations of aerobots can be highly
useful for the scientific exploration of our solar system, including
the earth.
Task Description
We have designed and implemented software for simulating multiple
aerobots/balloons incorporating complete balloon thermodynamic models
for high simulation fidelity. This is the first flexible simulation
software to allow simulation of multiple balloons of various types. The
object-oriented software in C++ simplifies implementation of new types
of balloons, ballast control systems, motion controllers, and
environments. The software incorporates new concepts for balloon flight
train mechanics based on multi-body dynamics inward-outward
recursion. The software also allows visualization of planetary
wind fields as shown in the following figures:
|
|
Earth wind fields |
Mars wind fields |
This wind field visualization software was used to visualize the
trajectories that balloons might follow in a planets winds with a
specified vertical trajectory. See the following figure for
an example trajectory at Mars:
|
Mars balloon trajectory |
The key questions that the research has
attempted to
address is whether a balloon can move to desired locations with only
limited
vertical control by using the planet's wind structure. These questions can be
divided into two separate issues:Reachability
and Path Planning. Reachability
is answering the question of whether it is even possible for an aerobot
to move
from on location to another desired location. Assuming that a path can be achieved
(the target is reachable), then the
second issue is how do we plan and execute a path that reaches the
target. These two issues can
also be further
divided by regional extent involved.First, the winds in a relatively small region can be treated as
being
constant or varying linearly across the region.Obviously, as the region gets larger and
reaches any significant portion of the globe, such simplifications
about wind
are unlikely to be valid and different techniques and approaches will
be necessary.
These software tools were also used to visualize cones of reachability
from a starting point with various vertical
trajectories. The following figure shows the the
reachable areas for a variety of vertical control decisions in a
specific non-constant wind field.
|
Wind field in rectangular volume with reachable volume illustrated. |
Dr. Ranjan Mukherjee of Michigan State University implemented C-based
simulation of single hot-air balloon for path planning research. They
demonstrated bang-bang altitude control algorithms. This enables
testing and demonstration of path planning concepts in wind fields of
limted extent. Optimal control algorithms were developed and
demonstrated optimal control of balloon motions in linear velocity
fields. This research proved that optimal paths involve minimal number
of vertical motion reversals. These results greatly enhance our
ability to plan and execute balloon paths in localized regions.
An eary phase of this research address the issue of pointing
instruments from balloon platforms to enable high-resolution surface
imagery/sensing from single and multiple balloon gondolas for planetary
exploration by developing improved pointing and stabilization
algorithms and technology.
|