One of the central principles upon which intelligent devices seem to operate is that of maintaining internal models of their external environments. How difficult this is, depends upon both the complexity of the model and its method of representation. In particular, it is usually easy when the problem is posed in the classical heuristic search paradigm, and the data structures used to represent static configurations of the puzzle are relatively unproblematic (arrays, lists, and so on). The lack of side-effects reflects the simplicity of the physics which such models embody. This limitation to elementary forms of interaction is not, of course, intrinsic to the heuristic search method; but when more complex models are constructed it becomes less trivial to pursue the consequences of performing an action. This approach is more general than the heuristic search method (but the latter -- when it has sufficient expressive power -- wins at present by its computational advantage). Assertions mentioning several different situations can then be used to describe dynamical laws which move us from one situation to another. But in some ways the resulting sharp separations between states of affairs are an embarrassment. For if we distinguish two situations s1 and s2, then from the fact, if such it be, that a predicate p is true of Si, nothing whatever follows concerning s2. And this is true even when s2 is directly associated with sl. Say s2 results from s1 by the performance of some action: s2 do (a, si) then no matter how remote -- speaking intuitively -- the connection between the property p and the action a, it still does not follow that p is true of s2. If we want it to so follow we must state this explicitly. Now, unfortunately, there are innumerable facts which might remain unchanged when actions are performed. So instead of writing a law of motion' in the form A(s) B(do(a, s)) where A and B are fairly short expressions, we are apparently obliged to list systematically all conceivable facts which are not changed. So that the law looks more like (Ci(s)& Ci(do(a, s))& & C„(do(a, s))&B(do(a, s)) for some very large n. This works for small problems (such as the familiar hungry anthropoid), but these are usually better formalized in the heuristic search paradigm anyway.

Amsterdam: North-Holland.

New York: American Elsevier.

New York: McGraw-Hill.

A recognition procedure with a range finder has been developed for the eye of the ETL-ROBOT, an intelligent robot studied at the Electrotechnical Laboratory. The range finder employs a vertical slit projector which projects a light beam on the objects. While the beam is moved in a field of view, the picture at each instant is picked up by a TV camera. The distance to each point can be obtained by means of trigonometrical calculation. The information thus obtained is utilized for the recognition procedure, where (1) each point is classified into lines, (2) each line is classified into planes, (3) 3-dimensional position of each plane is calculated, and (4) the object is recognized by the relationship between planes.

A paradigm enabling heuristic problem solving programs to exploit an analogy between a current unsolved problem and a similar but previously solved problem to simplify it s search for a solu­tion is outlined. It is developed in detail for a first-order resolution logic theorem prover. Descriptions of the paradigm, implemented LISP programs, and preliminary experimental results are presented. This is believed to be the firs t system that develops analogical information and exploits it so that a problem-solving program can speed its search.IJCAI-71, British Computer Society, London, 1971. Revised version in Artificial intelligence 2(2):147- 178, fall, 1971.

This paper echoes, from a philosophical standpoint, the claim of McCarthy and Hayes that Philosophy and Artificial Intelligence have important relations. Philosophical problems about the use of "intuition" in reasoning are related, via a concept of analogical representation, to problems in the simulation of perception, problem-solving and the generation of useful sets of possibilities in considering how to act. The requirements for intelligent decision-making proposed by McCarthy and Hayes are criticised as too narrow, and more general requirements are suggested instead. Introduction The aim of this paper is to illustrate the way in which interaction between Philosophy and A.I. may be useful for both disciplines. It starts with a discussion of some philosophical issues which interested me long before I knew anything about A.I., and which I believe are considerably enriched and clarified by relating them to problems in A.I., which, they, in turn, help to clarify. This discussion is followed by some general speculations about the conceptual and perceptual equipment required by an animal or machine able to cope with our spatiotemporal environment. Finally, there are further vague, general and programmatic remarks about the relations between Philosophy and A.I. The paper was inspired mainly by discussions with Max Clowes, but also to some extent by the attempts made by McCarthy and Hayes (12), and Hayes (8) to relate philosophical issues to problems in the design of intelligent robots.

An initial version of the program has been implemented in LISP on a PDP-10 and is being used in conjunction with robot research at SRI. STRIPS is a member of the class of problem solvers that search a space of "world models" to ind one in w hich a given goal is achieved. For any world model, we assume that there exists a set of appllcable ope rators, each of w hi eh transforms the world model to some other world model. The task of the problem solver is to find some composl11on of ope rat ors that trans forms a given initial worId mode] into one t hat satisfies some stated goa1 condltion. This f rarnewo rk for probl em so 1 v i ng has l een cen t ra 1 to much of t he research I n artificial Intel licence (1). Ou r p nmary interest he re is in the class of p robJ ems faced by a robot in rea rranging ob]ec t s and in navigatlng, l.e.

Session No. 8 Robots and Integrated Systems 385 TRAJECTORY CONTROL OF A COMPUTER ARM* by Richard Paul Stanford Artificial Intelligence Project Stanford University Stanford, California USA This paper describes the programming of a computer controlled arm. The programming is divided logically into planning and execution Communication between planning and execution is by a data file which specifies the arm trajectory with reapect to time, and actions that the arm should perform. The servo program which moves the arm along the trajectory is based on Legrangian mechanics and takes into account coupling between links, and the variation of inertial loading with change of arm configuration. Key words: INTRODUCTION arm, trajectory, servo We are Interested in driving a computer controlled arm such as the one shown in Figure 1. This arm [1] has six degrees of freedom with a vise grip hand and a useiuL working area about equivalent to that of a human arm. The arm is powered by printed circuit eietric motors with harmonic drive gear reductions.

MENTAL can be used as a guestion-answering system with formatted input /output, as a vehicle for experimenting with various theories of semantic structures or as the memory management portion of a natural language question-answering system. 1. Introduction In order to develop machines capable of "understanding" natural language, it is extremely valuable, if not necessary, to design a method of organizing a corpus of data to facilitate the storage and retrieval of information on many subjects, some in depth, some in breadth; to facilitate the storage, retrieval and use of the many complex relationships among real-world concepts; to facilitate the storage, retrieval and use of information which tells how other information in the corpus may be used to further explicate implied relationships among concepts; and to facilitate the identification from the vast corpus of data of those pieces of information most directly relevant to any given topic. This paper describes a data structure (MENS) and procedures for manipulating it The research reported herein was partially supported by a grant from the National Science Foundation (GJ-583) and partially by USAF Proj.