Richard Uhrich & James Walton
Naval Command, Control and Ocean Surveillance Center
RDT&E Division, Code 561
53560 Hull Street
San Diego, CA 92152-5000
Presented at the Ninth International Symposium on Unmanned Untethered
Submersible Technology, Durham NH, September 25, 1995.
At what point does communication make an AUV an ROV? Obviously the practice of classifying unmanned underwater vehicles as either ROVs or AUVs is unsatisfactory for free-swimming vehicles with communications, especially those capable of accepting real- time commands. We will resist the temptation to introduce yet another acronym (eg., Semi-autonomous Unmanned Untethered Undersea Vehicles), and will refer to them as semi-autonomous or supervised vehicles.
Like a purely autonomous vehicle, a supervisory controlled one is not encumbered by a cable, the proverbial tail which wags the dog of deep towed or tethered systems. Thus it can move at relatively high speeds and perform sharp, precise maneuvers, or it can hover stably without expending power fighting cable pull. Furthermore, the operators are relieved of tedious piloting and navigation functions, because all critical loops are closed on the vehicle. In effect, they have a remote autopilot: they can even direct the vehicle to go to a location miles away, then go to lunch trusting it to be waiting there when they return.
On the other hand, with supervisory control, operational decisions need not be preprogrammed or based on limited machine intelligence. As with a tethered vehicle, the operator can make operational decisions based on detailed, up-to-date information, and can immediately see the results of new commands. Decision thresholds and control loop parameters themselves can be reprogrammed on the fly. Objects or geographic areas of major interest can be both identified and fully explored on the same dive rather than on subsequent ones, thereby saving considerable time and resources. Even the response to critical or emergency situations can be flexible. If new information warrants, the operators can make decisions which instantly and totally redirect the mission.
On the other hand, long range is obtained at the cost of resolution. Unless the object of the search is a nearly intact ship or submarine, sonar is unlikely to unambiguously distinguish it from worthless objects, termed false targets. A search system must do more than identify the location of all likely or possible targets; it must confirm whether each is the object sought.
Generally, a search is not considered finished until a human sees optical images of the real target or of all potential targets.
Once the object sought is recognized, a detailed optical inspection can be conducted immediately. Many options are available during the inspection phase. Previously transmitted images can be retransmitted at higher resolution. New optical images can be requested from different altitudes and positions. A documentary film camera can be turned on or off. AUSS can back away and perform more sonar scans at higher resolution or longer range. If the object of interest is very large or found to be highly fragmented, the vehicle can perform a small photomosaic search pattern, taking overlapping pictures guaranteeing total optical coverage of a defined area.
An AUV only slightly smarter than the AUSS vehicle might recognize strong or unusual sonar returns, and close and photograph each one it finds in the defined search area, all in one deployment. A really smart AUV might even be capable of optically recognizing certain objects, and be authorized to terminate the search when it thinks the job is done. We would want such an AUV to be conservative, erring to search too much area rather than failing to find the object or failing to prove it not to be in the search area. It would therefore investigate many false sonar targets, often "seeing" the sought object but continuing the search because it wasn't "sure" of what it saw.
In either case, simple or smart, only after the AUV returned and humans reviewed the optical images could the search director be told whether the object was found or the area was thoroughly searched. Detailed inspection, if required, would await an ROV or a manned submersible.
Even if autonomous search were to be proven reliable, it would almost certainly be slower and more expensive than supervised search. An AUV, deprived of human expertise, insight and curiosity, could not perform inspection and exploration as AUSS does.
There are some missions which require acoustic silence, and there are ones wherein noise or geometry make acoustic communications physically impossible. There are cases where the mission is so simple or the cost of failure so low that communication is of little importance. However, if communication can be used, why not let humans monitor the remote system, handle unexpected situations, recognize difficult objects and make important decisions in real time?
A vehicle system designed from the start with supervisory control in mind will not likely turn out to be just an AUV with an acoustic link added on. It would probably employ completely different methods than previously considered, which might result in broader capabilities with a system which is simpler overall than a purely autonomous one.
Mackelburg, G. R., S. J. Watson, and W. D. Bryan. 1992. "Advanced Unmanned Search System (AUSS) Acoustic Communication Link Development." NRaD TR 1531 (Nov). Naval Command, Control and Ocean Surveillance Center, RDT&E Division, San Diego, CA.
Uhrich, R. W., and S. J. Watson. 1992. "Deep-Ocean Search and Inspection: Advanced Unmanned Search System (AUSS) Concept of Operation." NRaD TR 1530 (Nov). Naval Command, Control and Ocean Surveillance Center, RDT&E Division, San Diego, CA.
Walton, J. 1992. "Evolution of a Search System: Lessons Learned with the Advanced Unmanned Search System." NRaD TR 1529 (Nov). Naval Command, Control and Ocean Surveillance Center, RDT&E Division, San Diego, CA.
Address all questions/comments to:
Last update: 2 December 1998