This paper describes an implementation of the 3T robot architecture which has been under development for the last eightyears. The architecture uses three levels of abstraction and description languages whichare compatible between levels. The makeup of the architecture helps to coordinate planful activities with real-time behaviors for dealing with dynamic environments. In recent years, other architectures have been created with similar attributes but two features distinguish the 3T architecture: 1) a variety of useful software tools have been created to help implement this architecture on multiple real robots;, and 2) this architecture, or parts of it, have been implemented on a varietyofvery different robot systems using different processors, operating systems, effectors and sensor suites.

The Spatial Semantic Hierarchy is a model of knowledge of large-scale space consisting of multiple interacting representations, both qualitative and quantitative. The SSH is inspired by the properties of the human cognitive map, and is intended to serve both as a model of the human cognitive map and as a method for robot exploration and map-building. The multiple levels of the SSH express states of partial knowledge, and thus enable the human or robotic agent to deal robustly with uncertainty during both learning and problem-solving. The control level represents useful patterns of sensorimotor interaction with the world in the form of trajectory-following and hill-climbing control laws leading to locally distinctive states. Local geometric maps in local frames of reference can be constructed at the control level to serve as observers for control laws in particular neighborhoods. The causal level abstracts continuous behavior among distinctive states into a discrete model ...

This paper describes Rogue, anintegrated planning and executing robotic agent. Rogue is designed to be aroving o ce gopher unit, doing tasks such as picking up & delivering mail and returning & picking up library books, in a setup where users can post tasks for the robot to do. We have been working towards the goal of building a completely autonomous agent which can learn from its experiences and improve uponits own behaviour with time. This paper describes what we have achieved to-date: (1) a system that can generate and execute plans for multiple interacting goals which arrive asynchronously and whose task structure is not known a priori, interrupting and suspending tasks when necessary, and (2) a system which can compensate for minor problems in its domain knowledge, monitoring execution to determine when actions did not achieve expected results, and replanning to correct failures. 1.

Vision systems that have successfully supported nontrivial tasks have invariably taken advantage of constraints derived from the task and environment to increase reliability and lower the complexity of perception. We propose that it is possible to build a general purpose vision system, that is, one that can support a wide variety of tasks, and take advantage of such constraints. The central idea within our proposed architecture is the reactive skill. Skills are concurrent control routines assembled at run time using instructions from a symbolic execution system. Visual modules are used as resources in the construction of these skills. Skills control the agent as continuous feedback loops but are constructed using discrete, symbolic instructions. The key to general-purpose vision is the ability to parametrize the primitive elements of the vision system and to compose visual and control routines in a variety of ways. We demonstrate the architecture in the context of an implemented examp...

khaigh�cs.cmu.edu

1 Introduction Modularity Issues in Reactive Planning The RAP reactive plan execution system is specifically designed to accept plans and goals at a high level of abstraction and expand them into detailed actions at run time. One of the key features of the RAP system is the use of hierarchical expansion methods that allow different paths of execution for the same goals in different situations. Another central reason for the hierarchy is to create modular expansion methods that can be used in the execution of many different tasks. However, experience using the RAP system to control the University of Chicago robot Chip in the 1995 IJCAI robot competition has shown that there are difficult trade-offs between modularity and correctness in a predefined plan hierarchy. This paper describes the RAP hierarchies used to control the robot while cleaning up a small office space and discusses some of the issues raised in writing these RAPs to be useful for other tasks as well...

The Spatial Semantic Hierarchy (SSH) is a set of distinct representations for large scale space, each with its own ontology and each abstracted from the levels below it. At the control level, the agent and its environment are modeled as continuous dynamical systems whose equilibrium points are abstracted to a discrete set of distinctive states. The control laws whose execution defines trajectories linking these states are abstracted to actions, giving a discrete causal graph representation for the state space. The causal graph of states and actions is in turn abstracted to a topological network of places and paths (i.e. the topological map). Local metrical models of places and paths can be built within the framework of the control, causal and topological levels while avoiding problems of global consistency. ...

. Hybrid agent architectures are particularly successful in implementing a broad spectrum of (sub-)cognitive abilities, such as reactive feedback, deliberative problem solving, and social coordination. They can be seen as special instances of boundedly rational systems, i.e., systems that trade off the quality of a decision vs the invested consumption of resources. For sophisticated domains, such as the virtual soccer environment of the RoboCup simulation, we argue that a generalised framework that combines a hybrid design with explicit resourceadapting mechanisms is reasonable. Based on the layered agent model InteRRaP, we describe an experimental setting that is to guide and to evaluate the development of reasoning about abstract resources. These are representations of general interdependencies between computational processes. 1 Introduction In the nineties, the complementary AI paradigms of deliberative, perfect rationality and of reactive, myopic emergence have found th...

One of the difficult aspects of building an autonomous system, such as a service robot, is representing the plans needed to carry out tasks in unstructured environments. In this paper we outline the plans used by the University of Chicago entry in the 1994 AAAI robot competition. The robot was controlled using the Animate Agent Architecture, and the skills and RAPs presented constitute its task knowledge.

The purposive approach to computer vision has been much discussed in recent years, but there have been few tests of the viability of the paradigm in a working vision system. We have tested the approach in the context of vision for a mobile robot performing tasks in a typical, unaltered oOEce environment. In particular, we examine the approach in the context of a trash collection task; the task that was the subject of the 1994 AAAI Robot Competition. Our experience demonstrates that the approach is viable for solving such a task, and we show how our design for the visual system takes advantage of tight integration between the perception and action systems. In particular, we demonstrate that because of the constraints inherent in the task, it is possible to eOEciently perform a search over the scene and over viewpoint for the object of interest. We address particular concerns with the purposive approach, such as how purposive systems can be adapted to new situations. 1 Introduction Ov...