Review of Autonomous Undersea Vehicle (AUV) Developments 
SENSORS 
The NSB panel that focused on sensor technologies recommended the following [2]: 
Based on the technology trends and historical growth patterns, the panel anticipates that future sensor 
technology will be characterized by the following: 
e fver-decreasing size and cost as microelectronics evolves into nanoelectronics within the limits and 
constraints implied by the physics of the interfaces. 
e Migration of the analog-to-digital conversion to the front end of the sensor, leaving only those analog 
elements absolutely necessary for interfacing with the physical phenomenon to be sensed--e.g., 
microwave LNA, filters and power amplifiers, fiber-optic transducers, MEMS transducers, and the like. 
e Ever-increasing application of computer processing as gigaflops grow to teraflops and then to 
petaflops. 
e Development of monolithic smart sensors, combining sensing transduction, ADC, digital signal 
processing, communication input and output, and perhaps power conditioning on a single chip. This 
offers interesting possibilities for very small, very smart weapons such as affordable smart bullets 
(figure 2). 
Note that not all sensors can be small, even though the electronics can be. Size depends very much 
on the physics of the physical interface constraints. For example, propagation-based sensors such 
as RF radar and sonars typically require many wavelengths across the T/R aperture for good spatial 
resolution. Optical and millimeter-wave sensors, however, with their small wavelengths, and all 
MEMS-mediated sensors can and will become small and integrated. 
e As increasingly capable sensors evolve, it will be natural to deploy collections of autonomous, mobile, 
communicating sensors that can cooperate to function as a single, higher-level metasensor (figure 2). 
In a sense, the Navy's CEC already functions as a metasensor but is not yet viewed as such. In CEC, the 
radars are thought of as individual, independent sensors that are cooperating. The meta interpretation 
views the cooperating radars as a single sensor, which happens to have distributed and mobile 
components. 
In the future, as individual sensors grow smaller and more capable, and perhaps become autonomously 
mobile, they will be deployed in environments where each can see only a small part of the scene and can 
communicate only in a limited sense with other close-by minisensors. Under these conditions it becomes 
natural to think of the individual sensors as members of a distributed ant-like society that, through only 
local communications and simple local protocols, manages to behave as a single purposeful entity—that 
is, ametasensor. Investigations of the dynamics and potentially chaotic behavior of such distributed 
systems (e.g., flocks of birds or schools of fish) have recently begun to appear in the physics literature. 
This direction of research should be carefully nurtured. 
USER USER 
£4 
se7=i> od x 
Hlement Element 
: Smart er 
eee Monokithic Distributed 
ensor Sande Meta-Sensor 
Figure 2. Sensor Technologies Development 
17 
