From: brutzman@taurus.cs.nps.navy.mil (Don Brutzman) Subject: PUBS: Articles available on AUV simulation/visualization/etc. Date: 11 Aug 92 22:12:05 GMT Organization: Naval Postgraduate School, Monterey CA Selected papers are now available via anonymous ftp regarding the Naval Postgraduate School Autonomous Underwater Vehicle. Additional references are available upon request. Currrent: aaai92ws.ps "From virtual world to reality: designing an autonomous underwater robot" by Don Brutzman AAAI Fall 92 Symposium: Applications of Artificial Intelligence for Autonomous Mobile Robots, 23-25 OCT 92 oceans92.ps "Autonomous Sonar Classification using Expert Systems" Don Brutzman, Mark Compton, Yutaka Kanayama Proceedings of IEEE Oceans Engineering Society Conference OCEANS 92, 26-29 OCT 92 auv92.ps "NPS AUV Integrated Simulator" Don Brutzman, Yutaka Kanayama and Michael Zyda Proceedings of IEEE Oceans Engineering Society Conference AUTONOMOUS UNDERWATER VEHICLES 92, 3-4 JUN 92 To obtain a copy of compressed postscript papers: ftp ftp.cs.nps.navy.mil login: anonymous password: youraccountname@site.domain ftp> cd pub ftp> cd auv ftp> ls You will now see what is available. To get a paper: ftp> get auv92.ps.Z ftp> quit Back on your unix system you must uncompress the postscript file prior to printing: unix> uncompress auv92.ps.Z unix> lpr auv92.ps comments or questions: e-mail is welcome. regards, Don brutzman@cs.nps.navy.mil Abstracts follow. ___________________________________________________________________ From virtual world to reality: designing an autonomous underwater robot Design of autonomous underwater robots is particularly difficult due to the physical and sensor challenges of the underwater environment. Inaccessibility during operation and low probability of failure recovery makes robot stability and reliability paramount. Building an accurate and complete virtual world simulation is proposed as a necessary prerequisite for design of an autonomous underwater robot. A virtual world can include actual robot components and models for all other aspects of the world. Robot design can be fully tested using a virtual world and then verified using the real world. Additional testing can be performed in the virtual world that is not feasible in the real world. Visualization of robot interactions within a virtual world permits sophisticated analysis of robot performance that is otherwise unavailable. All aspects of world modeling and robot design must be mastered and coordinated in order to build an authentic virtual world and capable autonomous robot. ------------------------------------------------------------ AUTONOMOUS SONAR CLASSIFICATION USING EXPERT SYSTEMS An expert system can process active sonar returns, perform geometric analysis and autonomously classify detected underwater objects. Autonomous classification of objects is an essential requirement for independent operation by autonomous underwater vehicles (AUVs). Most AUVs are only capable of rudimentary sensor analysis, since standard approaches to evaluation and classification of sonar data require excessive signal processing and computational power to be practical. This paper describes how to develop an autonomous sonar classification expert system for a working AUV. A fundamental approach is presented for applying geometric reasoning and expert system heuristics to sonar classification. Preliminary sonar processing is performed using parametric regression line fitting. A polyhedron-building algorithm correlates the parametric regression line segments into geometric objects. After quantifying geometric object attributes, objects are classified using rule-based evaluation of quantitative and qualitative attributes combined with sonar classification heuristics. A summary of expert systems describes their salient features pertinent to autonomous sonar classification systems. The expert system paradigm, knowledge representation, reasoning using facts and rules, rule sets, control of execution flow and expert system development are outlined. Expert system self-diagnosis and self-correction are also discussed. Implementation was performed using the "C" Language Integrated Production System (CLIPS) expert system shell. Real-time graphic simulation and scientific visualization are employed to evaluate results. Experimental sonar classification results are presented using actual mission data from the Naval Postgraduate School (NPS) AUV. Successful classifications of walls and a mine-like object are demonstrated. --------------------------------------------------------------- INTEGRATED SIMULATION FOR RAPID DEVELOPMENT OF AUTONOMOUS UNDERWATER VEHICLES The development and testing of Autonomous Underwater Vehicle (AUV) hardware and software is greatly complicated by vehicle inaccessibility during operation. Integrated simulation remotely links vehicle components and support equipment with graphics simulation workstations, allowing complete real-time, pre-mission, pseudo-mission and post-mission visualization and analysis in the lab environment. Integrated simulator testing of AUV software and hardware is a broad and versatile method that supports rapid diagnosis and robust correction of system faults. Pre-mission simulator AUV testing permits experimental evaluation of developmental software. Pseudo-mission simulator testing of AUV processes employs an identical laboratory microprocessor or remote communication with a testbench-mounted operating AUV, permitting end-to-end testing of all software and hardware. Post-mission simulator playback of recorded telemetry, sensor data and system state transitions supports in-depth reenactment, playback and analysis of in-water operational results. High-resolution three-dimensional graphics workstations can provide real-time representations of vehicle dynamics, control system behavior, mission execution, sonar processing and object classification. Use of well-defined, user-readable mission log files as the data transfer mechanism allows consistent and repeatable simulation of all AUV operations. Examples of integrated simulation are provided using the Naval Postgraduate School (NPS) AUV, an eight foot, 387-pound untethered robot submarine designed for research in adaptive control, mission planning, mission execution, and post-mission data analysis. The flexibility, connectivity and versatility provided by this approach enables sophisticated visualization and analysis of all aspects of AUV development. Integrated simulator networking is recommended as a fundamental requirement for comprehensive and rapid AUV research and development.