Customer Reviews

Intelligent Robotic Systems for Space Exploration (The Springer International Series in Engineering and Computer Science)

5 star:		(0)
4 star:		(1)
3 star:		(0)
2 star:		(0)
1 star:		(0)

 

 

The title of this book is misleading. One would think that this would be a collection of essays on the latest generation of remote scientific space probes currently being developed. However, the bulk of the book is dedicated to the description of current research in the design of robots to assemble prefabricated objects in space. The goal of this research is, of course, the proposed space station Freedom. While that work is interesting in its own right, the scope of the book should have been expanded.
The problems of space assembly are of course formidable. Lacking the anchoring effect of gravity, the simple movement of tightening a bolt imparts a torque to the robot that must be compensated for. Without the scattering effect of an atmosphere, there is essentially no middle ground between glare and shadow. Developing a vision system that will differentiate between the proper and improper alignment of the parts in this environment is very difficult. This book does a good, but not complete job in presenting the problems. That is not meant to be a criticism per se, as that is to be expected, given the nature of the articles. The papers are by people reporting on their research, and are understandably somewhat biased on the positive side. However, as has been clear throughout the space programs of both the U. S. and the former Soviet Union, one must experience the problems created by free-fall before you can truly appreciate the complexities.
What is missing from the book is any sense of the intense and sometimes bitter debate being carried out within the space exploration community. The essence of that debate can be distilled down to two focal points that, in the current climate of budget restraints, are rapidly losing what common ground there was:

1) The overwhelming majority of space science, in space, should be done by machine intelligence.
2) Human intelligence in space should be a permanent condition.

Proponents of the first point consider the building of space station Freedom to be an exercise in political and scientific foolishness, with this research being only peripherally useful. This group is focused on the development of intelligent machines that can move around and sample their environment with little or no human input. The track record of these people is impressive, although some failures have occurred. However, even some of these have been compensated for by the brilliance of earthbound personnel. A case in point is the failure of the high-gain antenna on the Galileo craft to unfurl. Even though this prevents the high-speed transmission of data, changes in the collection and coding of the data should lead to the recovery of nearly all of the information.
Those who favor the second point consider a permanent human presence in space essential. Three primary arguments are put forward in support of this position:

(a) Only human space flight can capture the public imagination and guarantee the level of support necessary for adequate funding of the space program.
(b) No one can anticipate all possible problems and code for them, so only human intelligence is flexible enough to respond. This is generally an argument that questions the robustness of machines intelligence, and indirectly that of the programmers.
(c) All other space explorations, including the establishment of colonies, require a space station as a base.

This would have been a much better book if some of the points of this debate had been included.
Furthermore, remote scientific probes will require a higher level of intelligence than robots designed to construct objects in Earth orbit. The travel time in the cycle

machine encounters problems and requests instructions - >
request received and processed by humans, response transmitted - > instructions received by machine

is on the order of hours, while that in Earth orbit is nearly instantaneous. While harsh, the environment in Earth orbit is extremely consistent. A surface rover on a planet will have to dodge boulders, differentiate the interesting from the routine, carry out experiments and perhaps hibernate through a sandstorm. The level of intelligence required to do this dwarfs that of construction robots in space, and that should appear in this book. Also, the automated construction of space stations should not be considered space exploration.
In conclusion, if your primary interest is in how robots will construct objects in space, this book is for you. Otherwise, you need to do a little more exploring of the universe before you find what you want.

Published in "Journal of Automated Reasoning" reprinted with permission from Springer