Payton, D.W. 1990. Internalized plans: A representation for action resources. In Designing Autonomous Agents, Pattie Maes, (Ed.), MIT Press: Cambridge, MA, pp. 89--103.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....arises from purely reactive decision making (see Figure 2 15) This system is suited to situations where the environment is static and all the obstacles within it have been mapped. It is not ideal for dynamic situations or situations with incomplete information. 2.7. 3 Internalised Plans, Payton [[23]] When plans are constructed using state space search, the planning process generally finds an optimal path, returns this as the result, and if the assumptions under which the planning was done change, the process is repeated from the beginning. In particular, if the robot is forced to make even ....

Payton, D. W.: Internalized plans : a representation for action resources, Journal of robotics and autonomous systems(1&2), June 1990.

....advance, but are inflexible at run time due to their inability to store information dynamically. Hybrid or layered architectures attempt a compromise between purely reactive and planner based approaches, Hybrid systems span a large body of research (Firby 1987, Georgeff Lansky 1987, Arkin 1989, Payton 1990, Connell 1991) and tend to separate the control system into two or more communicating but otherwise independent parts; the low level reactive process takes care of the immediate safety of the robot, while the higher level uses the planner to select action sequences. While this approach works well ....

Payton, D. (1990), Internalized plans: a representation for action resources, in P. Maes, ed., `Designing Autonomous Agents: Theory and Practice from Biology to Engineering and Back', MIT Press, pp. 89--104.

....of planning and problem decomposition can be turned into a PM. Some of the ones we have been considering are abstraction hierarchies [70] 71] 72] case based planning [28] 29] 37] constraint based planners [80] 81] plan reuse [20] mixedinitiative planning [65] agent planning [2] 59] [67], and others [6] 23] 84] The one we describe here is a distributed planning PM called the Horde Planner (HP) The HP tries to collect together a complete set of resources that can solve a problem (or a set of problems) before any resources are applied. The HP uses the information in the WKB ....

David W. Payton, "Internalized Plans: A Representation for Action Resources", pp. 89-104 in [60]

....the discussion and conclusions. 1 2 Behaviour based control The behaviour based design paradigm is the result of work initiated and pursued at the MIT AI Lab by Brooks and his co workers [Bro86, Bro90b, Bro91, Con89, Mat90] and augmented by research conducted in other places [Ark90, Pay90, KR90] The main idea consists of building a controller for an autonomous system out of a set of primitive reactive behaviours that couple sensory inputs of the system to its actuating outputs. Ultimately a system built this way should reveal the ability to cope effectively with dynamic and ....

David W. Payton. Internalized plans: A representation for action resources. Robotics and Autonomous Systems, 6:89--103, 1990.

....where a problem solving ability seems to be necessary. This has lead some researchers to try to combine the reactive approach with planning in different ways. One of the greatest insights gained from this work is that plans should be considered more as resources than as programs to execute (Payton 1990). The immediate sensory readings from the environment should always have precedence over an internal plan. There have also been some attempts to combine an internal world model with reinforcement learning. The DYNA architecture proposed by Sutton (1992) is one noticeable example of this. Using an ....

Payton, D. W., (1990), "Internalized plans: a representation for action resources", in P. Maes (ed.) Designing autonomous agents, 89--103, Cambridge, MA: MIT Press.

.... et al. 1997) Examples of cooperative mechanisms are the potential field method (Khatib 1986) motor schemas used in (Arkin 1989) blending of fuzzy behaviors (Saffiotti et al. 1995, Yen Pfluger 1995) the dynamic systems approach (Large, Christensen Bajcsy 1997) and voting mechanisms (Payton et al. 1990, Rosenblatt 1995) A similar approach to voting is introduced in (Riekki Roning 1997) by combining action maps. 1.3 Related work overview 9 In the potential field approach (Khatib 1986) outputs generated by behaviors are interpreted as potential fields (attractive and repulsive forces) and ....

....following thesis can provide evidence in support of Thesis 1. Thesis 2 (P R produces satisficing Pareto optimal solutions) The command fusion mechanism proposed by Payton and Rosenblatt (P R) using voting schemes produces satisficing Pareto optimal solutions to the action selection problem. In (Payton et al. 1990) and later refined in (Payton et al. 1992, Rosenblatt 1997) David Payton and Kenneth Rosenblatt proposed an action selection mechanism based on voting schemes. Their approach is highly heuristic and governed by empiric and experimental justification of voting as a suitable mechanism for command ....

[Article contains additional citation context not shown here]

Payton, D. W. (1990), Internalized plans: A representation for action resources, in P. Maes, ed., `Designing Autonomous Agents', MIT/Elsevier, The MIT Press, Cambridge, Massachusetts, London, England, pp. 89--104.

....Despite their di#erences, most techniques can be classified within two major streams, i.e, reactive and deliberative navigation. In reactive methods, there is a stimulus response relationship between sensors and actuators, with very limited or no world modeling at all; e.g. see [4] 5] 6] [7], 8] 9] In deliberative techniques, a world model is used to formulate plans to which the robot is more or less committed; one example of this kind is given in [10] While reactive navigation proves to be flexible by virtue of its modular design approach, it may fail when confronted with ....

D. W. Payton, "Internalized plans: A representation for action resources," Robotics and Autonomous Systems, vol. 6, pp. 89-- 103, 1990.

....(e.g. collision avoidance) as well as goal oriented attraction based behaviors. The high level introduces intelligence into the system allowing for action selection and increased control over the low level. Hybrid architectures have recently increased in popularity (e.g. Arkin [10] Payton [125], and Connell [42] They have been found to successfully overcome the limitations exhibited by the individual strategies, discussed in previous sections. 3.4 The Move Toward Layered Architectures Two common complaints with pure behavioral control are that users cannot control behaviors at a ....

David W. Payton. Internalized plans: A representation for action resources. In Pattie Maes, editor, Designing Autonomous Agents, pages 89--103. MIT Press, 1990.

....but if the knowledge is incorrect or if the robot s actual position becomes su#ciently di#erent from its perceived position, much of that computation is wasted, and in the worst case, out of date information can cause the robot to collide with an obstacle. One example of a deliberative system is (Payton 1990) where a priori knowledge is used to create a gradient field which is used in conjunction with a simple reactive obstacle avoidance technique to choose how to move the robot. Neither of these approaches are adequate for what we want to do with our robot: we want our robot to be able to build and ....

Payton, D. W. 1990. Internalized plans: A representation for action resources. In Maes, P., ed., Designing Autonomous Agents. MIT Press.

....Ch. 5] and sophisticated control architectures [9] Several approaches have been investigated, ranging from reactive to deliberative behaviors. In the former, there is a stimulus response relationship between sensors and actuators, with as little as possible world modeling [10] for example, see [9, 11, 12, 13]. In deliberative approaches, a world model is used to plan actions, to which the robot is then more or less committed; examples of this kind are given in [14, 15] The fuzzy map concept is well suited for use in deliberative or mixed approaches, which are still preferable for finding good ....

D. W. Payton, "Internalized plans: A representation for action resources," Robotics and Autonomous Systems , 6, pp. 89--103, 1990.

....Figure 4.9: Hybrid decomposition of a robot system. generating activities that have to be executed. It can even stop execution if required. However, actual execution of the activities is done independently of the reasoning process. 4.4. 1 Example: Payton s hybrid architecture In [4] 5] and [6] a robot system based upon multiple behaviors is described. It was used for reflexive control of an autonomous land vehicle (ALV) The authors divided the control system into separate perception and planning units. Payton and his colleagues also defined four layers, see figure 4.10. The higher ....

D.W. Payton, Internalized Plans: a representation for action resources, Robotics and Autonomous Systems 6, 1990.

....to move straight forward if the obstacle is at a safe distance. Using standard fuzzy inferencing, e.g. max prod in this example, the rules are combined into a multivalued output that encodes the (grade of) desirability of each action from the behavior s point of view. Payton and Rosenblatt [17, 25] demonstrated that multivalued behaviors define a more suitable communication interface because they reduce the loss of information (compared to single valued behaviors) and hence define a suitable basis for finding consensus and compromize solutions. Furthermore, Payton and Rosenblatt [18] ....

David W. Payton. Internalized plans: A representation for action resources. In Pattie Maes, editor, Designing Autonomous Agents, pages 89--104. MIT/Elsevier, The MIT Press, Cambridge, Massachusetts, London, England, 1990.

....for Mobile Navigation. Adapted from [53] The advantages of preferential behaviors are demonstrated in [44] where it is argued that preferential outputs decrease the loss of information and serve a suitable interface between behaviors regardless of their level of planning. Further in [43] the authors demonstrate that this level of abstraction is well suited for integration of high level deliberative planners with low level reactive behaviors. DAMN has been successfully applied to in door as well as out door real world applications such as the out door CMU Navlab vehicles, and ....

....among the behaviors. The outputs generated by the fuzzy behaviors of the fuzzy approaches can also be considered multivalued, since their outputs are fuzzy membership functions. It seems that multivalued outputs define an appropriate interface and means of communication between the modules. In [43] Payton explains why multivalued outputs portray an appropriate interface between system components. One of his points is that the amount of loss of information is reduced thus it defines a better basis for communication. One problem, however, encountered by Rosenblatt and Payton is that the ....

David W. Payton. Internalized plans: A representation for action resources. In Pattie Maes, editor, Designing Autonomous Agents, pages 89--104. MIT/Elsevier, The MIT Press, Cambridge, Massachusetts, London, England, 1990.

....design of robot control systems. This view of behavior organization in insects and in vertebrates has led to the development of architectures such as behavior based or reactive control, situated agents , or subsumption , which have been used successfully by Brooks [9, 10] Maes [11] Payton[12, 13], Arkin , 14] and others in the design and fabrication of a number of autonomous robots. It should be noted that behavior based control does not necessarily preclude all forms of planning. For example, Maes [11] points out that action selection can be based on some form of planning , such as ....

....points out that action selection can be based on some form of planning , such as looking ahead or evaluating alternative goals, while preserving the task orientation of behavior based systems and avoiding the need for complete remapping of the world and replanning at every decision point. Payton [12] indicates that plans can be internalized and used to enhance performance, without giving up the ability to react to sensory stimuli. Similarly, Agre and Chapman [15] have suggested that plans can be used as resources in deciding on actions, rather than as rigid formulas for execution. Arkin [14] ....

Payton, D.W., Internalized plans: a representation for action resources. Robotics and Autonomous Systems, 1990. 6: p. 89103.

....On the other hand, even if the designer can pre code all reaction rules, it is often time consuming and too complex. The architecture to be presented in this paper will be between the above two extremes. This kind of compromise architecture, sometimes called hybrid or mixed architecture [3, 15], has a set of pre compiled reaction rules and a deliberative planner to deal with unpredicted situations in the reaction rules. In contradistinction to existing hybrid architectures, our situated agent will always act according to its situated knowledge and intentions. At any time point, only ....

Payton, D., Internalized plans: A representation for action resources, in: P. Maes (ed.), Designing Autonomous Agents, MIT Press, 1991, 89--103

.... Reactive Action Packages (RAPs) higher level primitives for planning which hide and take care of the details of execution (Firby 1987) and PRS (Procedural Reasoning System) an architecture for flexible control rule invocation (Georgeff Lansky 1987) Schemas (Arkin 1989) and several others (Payton 1990, Connell 1991) These systems tend to separate the control system into two or more communicating but otherwise independent parts. In most cases, the low level reactive process takes care of the immediate safety of the robot, while the higher level uses the planner to select action sequences. ....

Payton, D. (1990), Internalized Plans: a representation for action resources, in P. Maes, ed., `Designing Autonomous Agents: Theory and Practice from Biology to Engineering and Back', The MIT Press.

....importance is available. I. Threaded Reactive Robot Navigation The specific task examined in our research is robot navigation to a known goal position across an unmapped world potentially cluttered with obstacles. Many robot control systems have been proposed as solutions to this problem (e.g. [1, 6, 4, 7]) The system implemented in this research is based on the Autonomous Robot Architecture (AuRA) 2] AuRA consists of both reactive and deliberative components. The deliberative component sets high level goals and selects appropriate behaviors to achieve them. The reactive component of AuRA ....

D. Payton. Internalized plans: A representation for action resources. In P. Maes, editor, Designing Autonomous Agents. MIT Press, 1991.

....on explicit world models and representational knowledge is avoided during execution. ffl They are particularly well suited for dynamic and unstructured domains. Brooks subsumption architecture is a wellknown example of this control paradigm [15] Other representative examples include [3] 30] [46], 39] 49] These various strategies differ in several significant ways including the organization and decomposition of behaviors and whether arbitration, action selection, or concurrent processing is used. The interested reader is referred to [8] for a more complete review. Schema based ....

Payton, D.W., 1991. Internalized Plans: A Representation for Action Resources, in Designing Autonomous Agents, Maes, P. (ed), MIT Press, pp. 89103.

....systems is the abstraction of data information between different modules. Fully reactive systems (Brooks 1986, Agre Chapman 1987, Rosenschein Kaelbling 1986) behavior based systems (Matari c 1991, Matari c 1992c, Maes 1989) and hybrid systems (Firby 1987, Arkin 1989, Georgeff Lansky 1987, Payton 1990, Connell 1991) have all been proposed as alternatives to solving the data abstraction problem in control. However, not much work has yet been done in scaling up such systems to problems that require complex enough control and challenging enough planning abilities to demand multiple interacting ....

Payton, D. (1990), Internalized Plans: a representation for action resources, in P. Maes, ed., `Designing Autonomous Agents: Theory and Practice from Biology to Engineering and Back', MIT Press.

....Basic reactions to stimuli are combined to generate a resultant behavior. Since these reactions are usually independent, they may be produced by concurrently executable threads of computation. 2. 2 AuRA The Autonomous Robot Architecture Many reactive robotic systems have been developed (e.g. [5, 7, 9]) The system implemented in this research is based on the Autonomous Robot Architecture (AuRA) 3] AuRA is a hybrid architecture integrating both deliberative and reactive control. Among other things, the deliberative component selects waypoints, or intermediate goals the robot must approach. ....

D. Payton. Internalized plans: A representation for action resources. In P. Maes, editor, Designing Autonomous Agents. MIT Press, 1991.

....the box canyon and other navigational problems. This hybrid approach constitutes a step towards symbolic systems, while maintaining the speed and simplicity of reactive systems. II. Related Work A. Reactive Control for Robot Navigation There are many instances of reactive robotic systems (e.g. [6, 11, 13]) Our system, the Autonomous Robot Architecture (AuRA) 3] is a hybrid architecture incorporating aspects of both deliberative and reactive control. The scope of this paper is restricted to the reactive execution component. Motor schemas are the basic unit of behavioral control in AuRA and are ....

....being supported; this paradigm is referred to as action oriented perception [3] B. Other Approaches Several existing strategies are related to our method in their representation of geographic knowledge or use of memory: A navigational strategy based on gradient fields was developed by Payton [13]. Gradient fields are constructed to describe mission plans and then movement decisions are based on them. The fields may be derived from world map data or other mission constraints. Our method operates at a lower reactive level and differs in that no a priori geographic knowledge is assumed. As ....

Payton, D., "Internalized Plans: A Representation for Action Resources", Designing Autonomous Agents, Maes, P. Ed., MIT Press (1991).

....A great deal of research on autonomous systems design is done within the behaviour based paradigm. This paradigm is the result of work initiated and pursued at the MIT AI Lab by Brooks and co workers [Bro86, Bro90b, Bro91, Con89, Mat90] and augmented by research conducted in other places [Ark90, Pay90, KR90]. The main idea consists of building a controller for an autonomous (possibly intelligent) system out of a set of primitive reactive behaviours that couple sensory inputs of the system to its actuating outputs. Ultimately a system built this way should reveal the ability to cope effectively with ....

David W. Payton. Internalized plans: A representation for action resources. Robotics and Autonomous Systems, 6:89--103, 1990.

....to develop mobile robots capable of navigating autonomously in unstructured and or unexplored environments. The practicality of such a robot is a function of its ability to utilize a map to locate itself and its goals as it navigates throughout. In some situations a map may be available a priori [Miller89, Arkin90, Gat90, Payton90], in which case goal locations on the map are specified and path planning algorithms are employed to direct the robot s motion while avoiding known (mapped) obstacles. A more desirable, and more challenging situation is for the robot to be able to generate and maintain its map autonomously. That ....

Payton, D.W., Internalized Plans: A Representation for Action Resources, in Robotics and Autonomous Systems 6, pp. 89-103, 1990.

....by the system itself. A lot of research on this topic is currently being done within the behaviororiented paradigm. This paradigm is the result of work initiated and pursued at the MIT AI Lab by Brooks and his co workers [1, 3, 5, 6, 8, 15] and augmented by research conducted in other places [2, 16, 13]. The main idea consists of building a controller for an autonomous (possibly intelligent) system out of a set of primitive reactive behaviors that couple sensory inputs of the system to its actuating outputs. Ultimately a system built this way should reveal the ability to cope effectively with ....

David W. Payton. Internalized plans: A representation for action resources. Robotics and Autonomous Systems, 6:89--103, 1990.

....enough for it to make sense to plan far ahead. In other words, this corresponds to making a 1 year weather forecast. Another reason for not planning far ahead is that it is computationally intractable and that it has shown to lead to behaviours that do not respond very well to dynamic events [1] [11]. In a certain respect the idea of planning to the end of a problem is self contradictory as stated by Pearl [12] In fact, once we have the facility to run a complete analysis of the problem, the need to evaluate and rank information sources vanishes, since the optimal strategies found already ....

David W. Payton. Internalized plans: A representation for action resources. In Pattie Maes, editor, Designing Autonomous Agents: Theory and Practice from Biology to Engineering and Back, pages 89--103. MIT/Elsevier, 1990.

....this topic is currently being done within the behavior based design paradigm. This paradigm is the result of work initiated and pursued at the MIT AI Lab by Brooks and his co workers [Ang89, Bro86, Bro90b, Bro91, Con89, FBIB89, Mat90] and augmented by research conducted in other places [Ark90, Pay90, KR90] The main idea consists of building a controller for an autonomous (possibly intelligent) system out of a set of primitive reactive behaviors that couple sensory inputs of the system to its actuating outputs. Ultimately a system built this way should reveal the ability to cope effectively ....

David W. Payton. Internalized plans: A representation for action resources. Robotics and Autonomous Systems, 6:89--103, 1990.

....4: The Distributed Architecture for Mobile Navigation. The advantages of preferential behaviors are demonstrated in [10] where it is argued that preferential outputs decrease the loss of information and serve a suitable interface between behaviors regardless of their level of planning. Further in [33] the authors demonstrate that this level of abstraction is well suited for integration of high level deliberative planners with low level reactive behaviors. DAMN has been successfully applied to in door as well as out door real world applications such as the CMU Navlab project and the No hands ....

....the DAMN approach corresponds to the action that fulfills the combined preferences of the behaviors, whereas in the fuzzy approach defuzzification is used. It seems that multivalued outputs, fuzzy or preferential, define an appropriate interface and means of communication between the modules. In [33] Payton explains why multivalued outputs portray an appropriate interface between system components. One of his points is that the amount of loss of information is reduced thus it defines a better basis for communication. 3.6 Formalism One of the main challenges of mobile robotics is to devise ....

Payton, D. W. Internalized plans: A representation for action resources. In Designing Autonomous Agents, P. Maes, Ed. MIT/Elsevier, The MIT Press, Cambridge, Massachusetts, London, England, 1990, pp. 89--104.

....for the layered architecture. The last section contains the conclusions. 2 Behaviour based control The behaviour based design paradigm is the result of work initiated and pursued at the MIT AI Lab by Brooks and his co workers [5, 7, 8, 9, 24] and augmented by research conducted in other places [3, 15, 28]. The main idea consists of building a controller for an autonomous system out of a set of primitive reactive behaviours that couple sensory inputs of the system to its actuating outputs. Ultimately a system built this way should reveal the ability to cope effectively with dynamic and ....

David W. Payton. Internalized plans: A representation for action resources. Robotics and Autonomous Systems, 6:89--103, 1990.

.... planner to find a path in the model, then send each of the steps of the plan to the robot s wheels ) 37, 66, 45] Hybrid systems attempt a compromise between the thinking and acting extremes by employing a reactive system for low level control and a planner for higher level decision making [30, 36, 8, 73, 51, 22]. Hybrid systems tend to separate the control system into two or more communicating but largely independent parts. Behavior based systems are an alternative to hybrid systems; they enable fast real time responses through simple reactive behaviors that directly link sensors and effectors, but also ....

David Payton. Internalized plans: a representation for action resources. In Pattie Maes, editor, Designing Autonomous Agents: Theory and Practice from Biology to Engineering and Back, pages 89--104. MIT Press, 1990.

....systems is the abstraction of data information between different modules. Fully reactive systems (Brooks 1986, Agre Chapman 1987, Rosenschein Kaelbling 1986) behavior based systems (Matari c 1991, Matari c 1992c, Maes 1989) and hybrid systems (Firby 1987, Arkin 1989, Georgeff Lansky 1987, Payton 1990, Connell 1991) have all been proposed as alternatives to solving the data abstraction problem in control. However, not much work has yet been done in scaling up such systems to problems that require complex enough control and challenging enough planning abilities to demand multiple interacting ....

Payton, D. (1990), Internalized Plans: a representation for action resources, in P. Maes, ed., `Designing Autonomous Agents: Theory and Practice from Biology to Engineering and Back', MIT Press, pp. 89--104.

....and at the same time the robot can reach these places using navigation strategies based on world features rather than on physical positions. In this way the routes can be interpreted as suggestions like in the work of Agre and Chapman [1] and like in the potential fields approach of Payton [12], so that the route towards a goal place can be established through navigation strategies activated in each reference place. An example of route planning should be the following: go ahead towards the wall, then turn to the right and take the first way on the left . A Petri net can be defined as ....

Payton, D.W.: Internalized Plans: A representation for action resources. In: P. Maes (ed.): Designing Autonomous Robots. MIT Press, Cambridge, Massachusetts (1991) 89-103

.... used in Reactive Action Packages (RAPs) higher level primitives for planning that hide and take care of the details of execution (Firby 1987) PRS (Procedural Reasoning System) an architecture for flexible control rule invocation (Georgeff Lansky 1987) Schemas (Arkin 1989) Internalized Plans (Payton 1990), Contingency Plans (Connell 1991) and others. Hybrid solutions tend to separate the control system into two or more communicating but otherwise independent parts. In most cases, the low level reactive process takes care of the immediate safety of the agent, while the higher level uses the ....

Payton, D. (1990), Internalized Plans: a representation for action resources, in P. Maes, ed., `Designing Autonomous Agents: Theory and Practice from Biology to Engineering and Back', MIT Press.

No context found.

Payton, D.W. 1990. Internalized plans: A representation for action resources. In Designing Autonomous Agents, Pattie Maes, (Ed.), MIT Press: Cambridge, MA, pp. 89--103.

No context found.

D.W. Payton, Internalized Plans: A Representation for Action Resources, in Designing Autonomous Agents, ed. Pattie Maes, MIT Press, Cambridge, Mass, 1990, pp. 89-103.

No context found.

Payton, D.W. "Internalized Plans: A Representation for Action Resources, in Designing Autonomous Agents," ed. Pattie Maes, MIT Press, Cambridge, Mass, pp. 89-103, 1990.

....the current state of the system can be identified within that state space. We refer to plans formulated and used in this manner as internalized plans, since they embody the complete search and lookahead performed in planning, without providing an abstracted account of an explicit course of action [19]. 1) Avoiding Unnecessary Abstraction: The difference between the use of internalized plans and conventional plans is best illustrated in the context of the earlier example of a deficient route plan. In contrast to the abstract route plan, consider a gradient description of a plan to achieve the ....

D. W. Payton, "Internalized Plans: A Representation for Action Resources," Proceedings of the Workshop on Representation and Learning in an Autonomous Agent, Lagos, Portugal, November 16-18, 1988.

....the network. Second Level of Competence GRADIENT FIELD The above behaviors constitute a basic level of competence which endows the robot with the ability to avoid obstacles indicated in VMT input. Another source of information that is used in deciding upon a vehicle heading is the gradient field [9], which indicates the globally optimal route to a goal from any location. A representative gradient field is shown in Figure 14. The optimal path from any given point to the goal can be traced out by following the arrows. During the execution of a planned route, the gradient field is indexed by ....

D. W. Payton, "Internalized Plans: A Representation for Action Resources", presented at the Workshop on Representation and Learning in an Autonomous Agent, Faro, Portugal, November 16-18, 1988.

No context found.

David W. Payton. Internalized plans: A representation for action resources. In Pattie Maes, editor, Designing Autonomous Agents, pages 89--103. MIT Press, 1990.

No context found.

Payton, D.W. (1990). Internalized Plans: A Representation for Action Resources. Robotics and Autonomous Systems, 6, 89--103.

No context found.

D.W. Payton. Internalized Plans: A Representation for Action Resources. Robotics and Autonomous Systems, 6(12) :89--103, June 1990.

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