which tend to make the concept of the supervisory-controlled vehicle both feasible and excit- 

 ing. These are 



1 . The advent of strong, low-loss glass fibers, which allow many megahertz of signal 

 bandwidth over many kilometers of nonmetallic transmission line Uterally as thin 

 as a human hair. 



2. The development of high-energy Uthium batteries capable of providing the power 

 to propel torpedo-sized vehicles for many kilometers at respectable speeds. 



3. The appearance of the integrated circuit microprocessor, a remotely program- 

 mable computer-on-a-chip capable of making relatively complex control and inter- 

 facing decisions in situ and of guiding the vehicle safely home in the event of 

 tether breakage. 



Of course, if wide-bandwidth data transfer is not required for a particular mission, the 

 tether cable can be dispensed with entirely and an acoustic communication link can be sub- 

 stituted. Such a system is still supervisory-controlled because of its man-in-the-loop capabil- 

 ity. It appears possible to provide acoustic command control, and in some situations it may 

 be feasible to transmit some uplink data such as slow-scan television. While such communica- 

 tion is neither high-resolution nor real-time, some situations may benefit from such a cable- 

 less approach so that the degradation incurred is entirely acceptable (ref 2). However, even 

 in these cases the deployable optical fiber link may be superior as the primary communica- 

 tions channel because of its great noise immunity and lack of fade, with the acoustic capabil- 

 ity serving as a backup in the event of a tether breakage. 



In the most extreme case, the communications link could be eliminated entirely. A 

 vehicle with artificial intelligence and huge memory capacity, the totally autonomous robotic 

 vehicle (fig ID), can be postulated with no man in the loop (and, moreover, no need for a 

 support platform on station!). Such a robotic vehicle could be dropped (by aircraft or ship) 

 into an area to perform an inspection mission, and, after performing its task, be recovered. 

 The totally autonomous vehicle is not just a "blue-sky" concept; rather, it is a logical exten- 

 sion of the supervisory-controlled vehicle, and clearly will be feasible once already foresee- 

 able advances in the technology permit its realization. In practice, it is expected that the 

 distinction between the two vehicle types will be less than well defined as developments 

 continue. Both approaches can be lumped together under the generic category of "free 

 swimmers" or "free-swimming vehicles." 



This report concentrates on the feasibility of a supervisory-controlled free swim- 

 mer; specifically, a system that deploys an optical fiber tether for communications purposes, 

 with a supervisor that resides on a surface support platform. 



1.3 FREE-SWIMMING VEHICLE TESTBEDS 



There are at least two possible testbeds currently suitable for use as supervisory- 

 controlled inspection vehicles upon which we can demonstrate the use of fiber-optic tethers. 

 These are the vehicles which constitute the platform technology portion of the total U.S. 

 Geological Survey Free-Swimming Submersible Technology Development Program HAVE 

 (Experimental Autonomous VEhicle) (ref 3). One of these has been designed and fabricated 

 at NOSC, and the other at the University of New Hampshire as a product of the Sea Grant 

 Program. Both vehicles have "open-frame" construction, which is ideahy suited to the pipe- 

 line and structures inspection requirements of hovering and maneuvering at zero and low-to- 

 medium speeds. This type of construction also permits control and inspection sensors to be 



10 



