The Artificial Intelligence and Information Sciences Office of the Research and Engineering Directorate at the Johnson Space Center (JSC) has as its basic responsibility the function of "consulting through research," that is, matching technology in universities, industry, and other NASA centers to space station applications. This requires staying abreast of the state of the art by conducting technology development and applications research in knowledge-based systems, machine vision, and robotics. A significant contribution of the AI office is the support of space station system engineering and integration (SE&I) activities. The AI office was formed in October 1984 and currently has a staff of twelve civil service personnel. Facilities include a Symbolics 3600 and a Symbolics 3670; a VAX 11/780 operating under ULTRIX (a Unix lookalike); and a robotics laboratory with a multiple-arm mobile robot, vision sensors, and mock-ups for simulating satellite servicing.

Mission Planning and Analysis Division Technology Development and Applications Branch The Technology Development and Applications Branch of the Space Operations Directorate has the following functional responsibilities: (1) the evaluation, assessment, and implementation of new technology from NASA research centers, industries and universities relative to JSC space operation applications; (2) development of in-house capabilities for assessment and implementation of new technologies; and (3) prototype development and demonstrations of new technologies. Specific applications of artificial intelligence include robotic simulations, interactive graphics, information communication, man-machine interface, expert systems, automated software development, intelligent computer-aided instruction, intelligent computeraided engineering, expert system builders, parallel processing, and software verification and validation. Robotic Systems Development This project will develop a robotic software test bed that integrates artificial intelligence and graphics system technologies for dynamic system simulations. Specific objectives are (1) to develop a clearer understanding of the computing, communications, and control capabilities needed for coordinated robotic manipulation of objects in a spacebased environment; (2) to develop an understanding of tradeoffs and the complementary relationships between graphic simulation modules and empirical test-bed modules with respect to robotic subsystem design, development, and proof-of-concept demonstrations; (3) to identify requirements for integrating coordinated intelligent robotic units with the evolutionary space station network of computing systems, communications and control systems, and common service and hardware-software elements; and (4) to identify short-and long-term technology needs. In the process, technical issues such as perception, automated aids for problem solving, cognition and learn-Kathleen Healey is Special Assistant at the Simulation and Avionics Integration Division, Research and Engineering Directorate, NASA Johnson Space Center, Houston, Texas 77058.

For more detailed information about any of MITRE's projects please contact Joseph Katz (KATZQMITRE-Bedford) or Richard Brown CLINUS!BROWN@MITRE-Bedford) at the Bedford center or Peter 6onasso (BONASSOQMITRE) at the Washington center Subsequently, Rome Air Development Center took over support of the project and continues to fund part of our AI research effort. MITRE's current research is summarized below. The Bedford center is supported by 15 Symbolics Lisp machines netted to two Vax-780 file servers, while the Washington center is supported by both a classified and an unclassified facility, with 2 Lambdas and 2 Symbolics Lisp machines respectively netted to Vax-780 file servers. Both centers support creative groups of people who generate exciting new ideas. Planning MITRE Bedford AI Programs and Reasoning Research into planning and reasoning started with the development of the KNOBS system.

AI is currently applied in seven AI&DS research projects in five overlapping areas: software aids, image understanding, hypothesis formation, distributed AI, and AI for decision making. In addition, AI&DS has an interest in the areas of knowledge-based system development for various applications, advanced user interfaces, robotics, integration of information from multiple sources [Drazovich & Wishner-8 I], and fuzzy aids for control [Tong-801 and decision making [Tong & Bonissone-801 Current AI Research Projects This section briefly describes each research project at AI&DS that has some AI component. After the title of each ' research project, the principal investigator is listed. This research is to evaluate the need for and to design prototype tools that apply advanced software technologies to the problem of maintaining large software systems written in ADA. The technologies being considered include artificial intelligence, automatic programming, program verification, very high-level languages, program transformations, program optimization, and interactive programming aids.

In this article, I discuss the emerging field of artificial intelligence and legal reasoning and review the new book by Anne v.d.L. Gardner, An Artificial Intelligence Approach to Legal Reasoning, published by Bradford/MIT Press (1987, 225 pp., $22.50) as the first book in its new series on the subject. Dworkin 1977, 1985) offer insights valuable to AI, their jurisprudential analyses often raise more questions than they answer and their insights, couched in philosophical discourse, are difficult to harness computationally. Much of the knowledge used in legal reasoning is published, codified, and highly indexed. The legal system maintains extremely detailed records of its cases and commentary on them and except for the lowest-level courts, all cases are published and indexed commercially. For example, Shepard's Citations records and updates all forward and backward pointers for cases; that is, for a given case, all the succeeding cases citing it as well as all the cases it cited.

Bell & Howell Mail-Processing Systems (Skokie, Ill.), a manufacturer of mail sorting, folding, and inserting equipment, has implemented an expert system solution at the core of its integrated service information technologies ( The goal is to facilitate the design of exhaust mufflers of inlet manifolds in hours instead of days. Aerospace manufacturer Lockheed Martin (Bethesda, Md.) and the U.S. Air Force Manufacturing Technology Directorate (MANTECH) (Wright-Patterson Air Force Base, Ohio) are working together on a virtual manufacturing project to apply modeling and simulation technology during the design process to prove out and select optimal new concepts. David Blanchard is the editor of Intelligent Systems Report, Intelligent Manufacturing, and the new Electronic Commerce Update.

John Deere Dubuque Works (Dubuque, Iowa), a manufacturer of agriculture and industrial equipment, has implemented a virtual-reality system to use in its construction division. The system enables John Deere to use virtual-product prototypes to assess key design factors in construction equipment, such as visibility and the ability to reach controls. Nabisco Biscuit (East Hanover, N.J.), a manufacturer of cookies and crackers, has installed an intelligent process-operating guidelines (POG) system. This POG system uses expert system technology to provide realtime process control information to the bakeries. Automotive manufacturer Ford Powertrain Operations (Dearborn, Mich.) has developed a flexible manufacturing system that is being controlled by an intelligent cell controller.

The system will help eliminate avoidable pollution and save pharmaceutical and chemical manufacturers millions of dollars. The Batch Design Kit will be commercially available in the summer of 1995. Photosound (Saffron-Walden, England) is using virtual reality to simulate and visualize the effects of different pharmaceutical drugs on the human body. The results are being used as the focal point of exhibition stands designed by Photosound for such pharmaceutical firms as Smith-Kline Beecham. Westinghouse Electric (Pittsburgh, Penn.) and Carnegie Group (Pittsburgh, Penn.) are working with the Pittsburgh Cancer Institute on a knowledge-based intelligent system for the management of national clinical trials.

Inference Corp. (El Segundo, Cal.) has entered into a joint agreement with software giant Microsoft Corp. (Redmond, Wash.) that will enable Microsoft to leverage Inference's casebased reasoning technology and engineering services in future versions of the Microsoft Windows operating system. Inference's client/server, objectoriented, 32-bit technology will be used for intelligent user assistance as a component of Windows. Bank Bandeirantes (Sao Paulo, Brazil) has teamed up with another Brazilian company, Software Marketing International, to develop SISCRED, a credit analysis expert system that operates in real-time. The intelligent system has helped the bank experience dramatic gains in development productivity. The bank anticipates deploying the system in all 100 Brazil offices.

Prairie Virtual Systems Corp. (Chicago, Ill.) has developed a barrier-free design virtual reality system that checks building access. The system, which helps designers meet the requirements of the American with Disabilities Act, assists in the design of interiors that are accessible and safe. It allows users to travel through a virtual world that simulates a proposed interior environment. United Technologies Corp. (Hartford, Conn.) has developed intelligent maintenance aiding technology to reduce operation and support costs and increase the availability of complex equipment, systems and processes. The company is applying AI and expert systems technology in the development, manufacture and support of products using a patented AI technique called qualitative reasoning, which compares faulty operations to knowledge about how a system normally operates.