| P. Maes, `Situated agents can have goals', Journal for Robotics and Autonomous Systems, 6(1), 49--70, (1990). |
....fields. B. Scale to more sophisticated motion tasks Extensive simplification of the action selection mechanism can unnecessarily deflate the reactive capabilities to achieving only simple tasks; information useful to selecting actions has not been fully exploited. Arbitration strategies [8] [19] are severely victimized by their policy of one winning behavior among a group of competing ones to assume full control of the robot until the next selection cycle. This precludes the execution of several, possibly conflicting motion tasks (e.g. target reaching and obstacle avoidance) in ....
P. Maes. Situated agents can have goals. In P. Maes, editor, Designing Autonomous Agents: Theory and Practice From Biology to Engineering and Back, pages 49--70. MIT Press, Cambridge, MA, 1990. Also appeared in Special Issue of Robotics and Autonomous Systems, 6(1-2).
....military units such as division, battalion, company, platoon, squad, and vehicle. Agent architectures such as are popular in the current literature are typically organized from the perspective of agents. Examples include personnel with rank such as major, captain, lieutenant, sergeant, private [13, 14, 15]. The ISAM architecture can be viewed either as a hierarchy of agents or a hierarchy of organizational units. In both cases, it is important to maintain a clear distinction 28 between organizational units and the agents that belong to them. If this distinction is not carefully observed, the system ....
Maes, P. (1990) "Situated Agents Can Have Goals," Robotics and Autonomous Systems, 6, pp.49-70.
....plan and thus on the internal model. To avoid the risk of a premature commitment to a potentially incorrect or incomplete system model, various behavior based control architectures have been designed in which behavior is constructed on line from combinations of primitive, reactive behaviors [19, 93, 31, 6, 72]. These approaches avoid the complexity of monolithic controllers, and since models can be expensive to build and misleading as predictive devices, traditional models are avoided. In the subsumption architecture [19] a number of parallel sensor action mappings are are arranged in layers and used ....
Maes, Pattie. Situated agents can have goals. Robotics and Autonomous Systems 6, 1 (1990), 49-70.
....order to achieve higher and higher levels of competence. Both the layers of behaviors and the integration of the behaviors are handcoded by the developer. During the following few years after Brooks invention of the behavior based approach, a number of re searchers like L. Steels [10] P. Maes [9], R. Arkin [1] etc. developed different architectures for the behaviorbased approach to robotics. The architectures use different representations and different behavioral coordination methods. In general, simple behaviors are handcoded, and the behaviors are coordinated through competitive ....
P. Maes. Situated agents can have goals. In P. Maes, editor, Designing Autonomous Agents, Cambridge, MA, 1990. MIT Press.
....with the ordinary plan: controlling a manipulator and the artificial ant problems. The results show that evolving reactive plans requires much less computational effort than ordinary plans. Key Words: Reactive planning, robot planning, evolutionary computation. 1. Introduction Reactive planning [1,2,3,4] is a form of robot planning where the plan focuses on response to the environment. The main aim of this type of planning is to create systems that sense the environment and act in real time. To achieve this goal reactive systems have strong coupling between sensing and action. The cycle ....
Maes, P. 1990. Situated agents can have goals. Robotics and Autonomous Systems 6:49-70.
....its own set of behaviours, and all the layers run asynchronously. This was designed to simplify the management of behaviour interaction, as each layer contains only a small number of active behaviours at any given point. Behaviour selection is performed on the basis of level of activation (c.f. [6]) a main factor of which is perceptual input. world models (abstraction) L:ayer 0 Layer 1 . selection action behaviour instantiation info extraction . generic behaviours Fig. 1. The layered design of the behaviour based system. As shown in the figure, there are a ....
P. Maes. Situated agents can have goals. In P. Maes, editor, Designing Autonomous Agents. MIT Press, Cambridge, MA, 1990.
....sensors on one hand and manipulators and speech synthesizers on the other have already reached high standards. Rather, it is the interpretation of the sensory data that poses the major obstacle for a robot to become truly situated [30] or embedded [29] with all the theoretical implications [22]. We claim that, currently, situatedhess does not make much of a difference for a robot because the resulting information is simply not available to its control system. Even representations close to the raw sensory data, which have been proposed as a way out of this problem, need a level of ....
Pattie Maes. Situated agents can have goals. In Pattie Maes, editor, Designing Autonomous Agents, pages 49-70. MIT press, third edition, 1994. Reprinted from Robotics and Autonomous Systems, volume 6, num- bers i & 2 (June 1990).
No context found.
P. Maes, `Situated agents can have goals', Journal for Robotics and Autonomous Systems, 6(1), 49--70, (1990).
No context found.
P. Maes. Situated agents can have goals. In Patti Maes, editor, Designing Autonomous Agents, pages 49--70. MIT Press, 1990.
No context found.
Maes, P.: Situated Agents can have Goals. Ed. P. Maes, Designing Autonomous Agents: 49-70, MITT Press, Cambridge, MA, 1990.
No context found.
P. Maes. Situated agents can have goals. In P. Maes, editor, Designing Autonomous Agents: Theory and Practice from Biology to Engineering and Back, pages 49--70. MIT Press, Cambridge, MA, 1990.
No context found.
Patti Maes. Situated agents can have goals. In Patti Maes, editor, Designing Autonomous Agents, pages 49--70. MIT Press, 1990.
No context found.
P. Maes. Situated agents can have goals. Robotics and Autonomous Systems, 6:49--70, 1990.
No context found.
P. Maes. Situated agents can have goals. Robotics and Autonomous Systems, 6:49--70, 1990.
No context found.
Pattie Maes, Situated agents can have goals, Robotics and autonomous systems 6, 1990, 49-70. 49
No context found.
Patti Maes. Situated agents can have goals. In Patti Maes, editor, Designing Autonomous Agents, pages 49--70. MIT Press, 1990.
No context found.
Patti Maes. Situated agents can have goals. In Patti Maes, editor, Designing Autonomous Agents, pages 49--70. MIT Press, 1990.
No context found.
Maes, P., "Situated Agents Can Have Goals," Robotics and Autonomous Systems, 6, North-Holland, 1990.
No context found.
P. Maes. Situated agents can have goals. Journal for Robotics and Autonomous Systems, 6(3):49--70, June 1990.
No context found.
P. Maes, Situated agents can have goals, Robotics and autonomous systems 6, 1990, 49-70.
No context found.
Maes, P.: Situated Agents can have Goals. Ed. P. Maes, Designing Autonomous Agents: 49-70, MITT Press, Cambridge, MA, 1990.
No context found.
P. Maes. Situated agents can have goals. Robotics and Autonomous Systems, 6:49--70, 1990.
No context found.
Maes, P. "Situated Agents can have Goals", Robotics and Autonomous Systems, Vol 6.
No context found.
Pattie Maes. Situated agents can have goals. In Pattie Maes, editor, Designing Autonomous Agents, pages 49--70. MIT Press, 1990. 159
No context found.
Maes, P. 1990: Situated Agents can have Goals. In: Designing Autonomous Agents, Ed. P. Maes, pp. 49--70, MITT Press, Cambridge, MA.
First 50 documents Next 50
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC