The control and integration of distributed, multi-sensor perceptual systems is a complex and challenging problem. The observations or opinions of different sensors are often disparate incomparable and are usually only partial views. Sensor information is inherently uncertain and in addition the individual sensors may themselves be in error with respect to the system as a whole. The successful operation of a multi-sensor system must account for this uncertainty and provide for the aggregation of disparate information in an intelligent and robust manner. We consider the sensors of a multi-sensor system to be members or agents of a team, able to offer opinions and bargain in group decisions. We will analyze the coordination and control of this structure using a theory of team decision-making. We present some new analytic results on multi-sensor aggregation and detail a simulation which we use to investigate our ideas. This simulation provides a basis for the analysis of complex agent structures cooperating in the presence of uncertainty. The results of this study are discussed with reference to multi-sensor robot systems, distributed Al and decision making under uncertainty.

Scheduling in the factory setting is compounded by computational complexity and temporal uncertainty. Together, these two factors guarantee that the process of constructing an optimal schedule will be costly and the chances of executing that schedule will be slight. Temporal uncertainty in the task execution time can be offset by several methods: eliminate uncertainty by careful engineering, restore certainty whenever it is lost, reduce the uncertainty by using more accurate sensors, and quantify and circumscribe the remaining uncertainty. Unfortunately, these methods focus exclusively on the sources of uncertainty and fail to apply knowledge of the tasks which are to be scheduled. A complete solution must adapt the schedule of activities to be performed according to the evolving state of the production world. The example of vision-directed assembly is presented to illustrate that the principle of least commitment, in the creation of a plan, in the representation of a schedule, and in the execution of a schedule, enables a robot to operate intelligently and efficiently, even in the presence of considerable uncertainty in the sequence of future events.

The discovery that the minimax decision rule performs poorly in some games has sparked interest in possible alternatives to minimax. Until recently, the only games in which minimax was known to perform poorly were games which were mainly of theoretical interest. However, this paper reports results showing poor performance of minimax in a more common game called kalah. For the kalah games tested, a non-minimax decision rule called the product rule performs significantly better than minimax. This paper also discusses a possible way to predict whether or not minimax will perform well in a game when compared to product. A parameter called the rate of heuristic flaw (rhf) has been found to correlate positively with the. performance of product against minimax. Both analytical and experimental results are given that appear to support the predictive power of rhf.

One of the most important aspects of current expert systems technology is the ability to make causal inferences about the impact of new evidence. When the domain knowledge and problem knowledge are uncertain and incomplete Bayesian reasoning has proven to be an effective way of forming such inferences [3,4,8]. While several reasoning schemes have been developed based on Bayes Rule, there has been very little work examining the comparative effectiveness of these schemes in a real application. This paper describes a knowledge based system for ship classification [1], originally developed using the PROSPECTOR updating method [2], that has been reimplemented to use the inference procedure developed by Pearl and Kim [4,5]. We discuss our reasons for making this change, the implementation of the new inference engine, and the comparative performance of the two versions of the system.

The fundamental elements of evidential reasoning problems are described, followed by a discussion of the structure of various types of problems. Bayesian inference networks and state space formalism are used as the tool for problem representation. A human-oriented decision making cycle for solving evidential reasoning problems is described and illustrated for a military situation assessment problem. The implementation of this cycle may serve as the basis for an expert system shell for evidential reasoning; i.e. a situation assessment processor.

This paper presents a methodology for research and development of the inferencing and knowledge representation aspects of an Expert System approach for performing reasoning under uncertainty in support of a real time vehicle guidance and navigation system. Such a system could be of major benefit for non-terrain following low altitude flight systems operating in foreign hostile environments such as might be experienced by NOE helicopter or similar mission craft. An innovative extension of the evidential reasoning methodology, termed the Sum-and-Lattice-Points Method, has been developed. The research and development effort presented in this paper consists of a formal mathematical development of the Sum-and-Lattice-Points Method, its formulation and representation in a parallel environment, prototype software development of the method within an expert system, and initial testing of the system within the confines of the vehicle guidance system.

Although many investigators affirm a desire to build reasoning systems that behave consistently with the axiomatic basis defined by probability theory and utility theory, limited resources for engineering and computation can make a complete normative analysis impossible. We attempt to move discussion beyond the debate over the scope of problems that can be handled effectively to cases where it is clear that there are insufficient computational resources to perform an analysis deemed as complete. Under these conditions, we stress the importance of considering the expected costs and benefits of applying alternative approximation procedures and heuristics for computation and knowledge acquisition. We discuss how knowledge about the structure of user utility can be used to control value tradeoffs for tailoring inference to alternative contexts. We address the notion of real-time rationality, focusing on the application of knowledge about the expected timewise-refinement abilities of reasoning strategies to balance the benefits of additional computation with the costs of acting with a partial result. We discuss the benefits of applying decision theory to control the solution of difficult problems given limitations and uncertainty in reasoning resources.

Ross D. Shachter and Leonard J. Bertrand Department of Engineering-Economic Systems, Stanford University (visiting the Center for Health Policy Research and Education, Duke University, PO Box GM, Durham, NC 27706) and Strategic Decisions Group, Menlo Park, CA for the Third Workshop on Uncertainty in Artificial Intelligence Seattle, Washington, July 10-12, 1987 The generalized fault diagram, a data structure for failure analysis based on the influence diagram, is defined. Unlike the fault tree, this structure allows for dependence among the basic events and replicated logical elements. A heuristic procedure is developed for efficient processing of these structures. Deterministic logic and conditional probabilities are both appealing frameworks in which to build a knowledge base. Each has a natural graphical representation, semantic network for logic and influence diagrams (Howard and Matheson, 1981) or bayes networks (Pearl, 1986) for probabilities.

This paper reports work in progress on an explanation facility for Bayesian conditioning aimed at improving user acceptance of probability-based decision support systems. Design of the facility, which appears to be reasonably domain-independent, is based on an information processing model that accounts both for biased and normative behavior in reasoning about conditional evidence. Preliminary results indicate that the facility is both acceptable to naive users and effective in improving understanding of Bayesian conditioning.

A method for computing probabilistic propositions is presented. It assumes the availability of a single external routine for computing the probability of one instantiated variable, given a conjunction of other instantiated variables. In particular, the method allows belief network algorithms to calculate general probabilistic propositions over nodes in the network. Although in the worst case the time complexity of the method is exponential in the size of a query, it is polynomial in the size of a number of common types of queries.