This paper proposes a new logic programming language called GOLOG whose interpreter automatically maintains an explicit representation of the dynamic world being modeled, on the basis of user supplied axioms about the preconditions and effects of actions and the initial state of the world. This allows programs to reason about the state of the world and consider the effects of various possible courses of action before committing to a particular behavior. The net effect is that programs may be written at a much higher level of abstraction than is usually possible. The language appears well suited for applications in high level control of robots and industrial processes, intelligent software agents, discrete event simulation, etc. It is based on a formal theory of action specified in an extended version of the situation calculus. A prototype implementation in Prolog has been developed.

This paper describes a novel approach to high-level agent programming based on a highly developed logical theory of action. The user provides a specification of the agents' basic actions (preconditions and effects) as well as of relevant aspects of the environment, in an extended version of the situation calculus. He can then specify behaviors for the agents in terms of these actions in a programming language where one can refer to conditions in effect in the environment. When an implementation of the basic actions is provided, the programs can be executed in a real environment; otherwise, a simulated execution is still possible. The interpreter automatically maintains the world model required to execute programs based on the specification. The theoretical framework includes a solution to the frame problem, allows agents to have incomplete knowledge of their environment, and handles perceptual actions. The theory can also be used to prove programs correct. A simple meeting sc...

: Agents interacting with an incompletely known dynamic world need to be able to reason about the effects of their actions, and to gain further information about that world using sensors of some sort. Unfortunately, sensor information is inherently noisy, and in general serves only to increase the agent's degree of confidence in various propositions. Building on a general logical theory of action formalized in the situation calculus, developed by Reiter and others, we propose a simple axiomatization of the effect on an agent's state of belief of taking a reading from a noisy sensor. By exploiting Reiter's solution to the frame problem, we automatically obtain that these sensor actions leave the rest of the world unaffected, and further, that non-sensor actions change the state of belief of the agent in appropriate ways. Keywords: situation calculus, theories of action, knowledge, degree of belief. Declaration: This paper has not already been accepted by and is not currently under rev...

in this paper, we develop an account of the kind of deliberation that an agent that is doing planning or executing high-level programs under incomplete information must be able to perform.

The notion of computational agents has become very fashionable lately [24, 32]. Building such agents seems to be a good way of congenially providing services to users in networked computer systems. Typical applications are information retrieval over the internet, automation of common user activities, smart user interfaces,

In this paper, we propose a definition of goal achievability: given a basic action theory describing an initial state of the world and some primitive actions available to a robot, including some actions which return binary sensing information, what goals can be achieved by the robot? The main technical result of the paper is a proof that a simple robot programming language is universal, in that any effectively achievable goal can be achieved by getting the robot to execute one of the robot programs. The significance of this result is at least two fold. First, it is in many ways similar to the equivalence theorem between Turing machines and recursive functions, but applied to robots whose actions are specified by an action theory. Secondly, it provides formal justifications for using the simple robot programming language as a foundation for our work on robotics. 0 1

In this paper, we use an extended version of the situation calculus to formalize goals and rational action. We then use these notions and a definition of ability (Lesp'erance et al. 1995b) to show that an agent that is acting rationally will achieve its goals when it is able to do so. 1 Introduction This paper describes work on rational action that arose from our efforts to create an explicit representation of the goals of agents in the situation calculus. The utility of an explicit representation of the goals of agents is evident when we consider domains with multiple interacting agents. In domains where agents are communicating and cooperating to perform a task, the ability to specify the knowledge and goals of the agents becomes useful in order to determine that the agents can perform their parts of the task and have the required commitment to see their parts to completion. Designers of agents can use this information to help predict the behaviors of the agents they create. The ag...

. This paper addresses the problem of ensuring that agents' plans are epistemically feasible in multiagent systems specifications. We propose some solutions within the CASL formalism. We define a subjective execution construct Subj that causes the plan to be executed in terms of the agent's knowledge state rather than in therms of the world state. The definition assumes that the agent does not do planning or lookahead and chooses arbitrarily among the actions allowed by the plan. We also define another deliberative execution operator Delib for smarter agents that do planning. We show how these notions can be used to express whether a plan is epistemically feasible for an agent in several types of situations. 1

Agent theories and agent programs are two very different styles of specification of agent behavior. The former are declarative in nature, while the latter have an imperative flavor. In this paper, we combine ideas from both areas, yielding a powerful mode of agent specification that also gives the specifier a good deal of control over the complexity of the specified agent. In particular, we extend Shapiro et al.’s [16] agent theory to handle prioritized goals and then integrate it with the IndiGolog agent programming language. The result is a new IndiGolog construct that transforms a given nondeterministic, concurrent program δ into a new program δ ′ that can be described as a rational implementation of the original program, in the sense that δ ′ is an implementation of δ, and furthermore, δ ′ is the most rational of all implementations of δ relative to a given set of prioritized goals and the agent’s knowledge. With this construct, we can specify an agent that will attempt to achieve as many goals as possible in priority order even if the agent does not know of a plan that is guaranteed to achieve all the goals. In this case, the agent will select a plan that she thinks has the best chance of achieving the goals.

Abstract not found