N.J. Nilsson, "Shakey the robot," Technical Report 223, SRI International, 1984.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....stable. 2 Related Work This section is a brief review of mobile manipulation and its relation to the present work. A more thorough survey can be found in[1] Several preceding systems have explored the connection between manipulation and locomotion. One of the earliest influential robots, Shakey[2], pushed boxes and other objects. More recently, the task domain of Sojourner in Mars included elements of manipulation. A more direct approach is to attach a manipulator to a mobile platform. The JPL Cart[3] provides an early example, while Romeo and Juliet[4] provide a current example. These ....

N.J. Nilsson, "Shakey the robot," Technical Report 223, SRI International, 1984.

....stable. 2 Related Work This section is a brief review of mobile manipulation and its relation to the present work. A more thorough survey can be found in[1] Several preceding systems have explored the connection between manipulation and locomotion. One of the earliest influential robots, Shakey[2], pushed boxes and other objects. More recently, the task domain of Sojourner in Mars included elements of manipulation. A more direct approach is to attach a manipulator to a mobile platform. The JPL Cart[3] provides an early example, while Romeo and Juliet[4] provide a current example. These ....

N.J. Nilsson, "Shakey the robot," Technical Report 223, SRI International, 1984.

....the two systems, the challenges for ROGUE include developing a communication mechanism for control and feed back, as well as extending the planner to handle the dynamics of a real world task. 1.2. Planning and Execution in ROGUE There are a few approaches to creating plans for execution. Shakey [Nilsson, 1984] was the first system to use a planning system on a robot. This project was based on a classical planner that ig nored real world uncertainty [Fikes, Hart, Nilsson, 1972] and followed a deterministic model to generate a single executable plan. When execution failures occurred, replanning was ....

Nilsson, N.J. (1984). Shakey the robot. Technical Report 323, AI Center, SRI International, Menlo Park, CA.

.... developing a clear semantics for robots and their environments, within the principled framework of logical description, will lead to further advances in our overall understanding of intelligence in robots and artificial agents [8] Despite a promising early start with projects such as SRI s SHAKEY [13, 20], the rigours of applying formal logic to real robots remain problematic. In part, these problems arise from the semantic knife edge [14] the observation that logic representations are apparently inevitably balanced between a lack of expressivity, which renders them useless for the tasks they ....

Nilsson, N. J. (1984) "Shakey the Robot", SRI Technical Note no. 323, SRI, Menlo Park, CA.

....or overhead view imagery. Our path planner can also be run independently of WITS, using a separate GUI. 2. PREVIOUS WORK IN ROVER PATH PLANNING Autonomous mobile robot research has been underway for around 30 years. One of the first such robots, Shakey, was developed between 1966 and 1972 [2]. Already then was it recognized that visual perception was one of the most important issues. Other well known robotic ground vehicles capable of autonomous navigation include the HERMIES series of robots developed at the Oak Ridge National Laboratory [3] and the NAVLAB series of Autonomous Land ....

N. J. Nilsson, "Shakey the Robot," SRI AI Center, Technical Report 323, April, 1984.

....to an agent s procedure for deciding what to do, i.e. choosing actions. Over time, many action selection mechanisms have been designed, some of which have lead to a rethinking of what it means for an agent to decide what to do . In the 1970s, action selection was done via deliberative thinking [58], i.e. planning. Planning involves using a symbolic model of the agent s goals, capabilities and environment to generate a series of actions that meet the agent s goals. This plan is then carried out by an executive that simply performs the actions and (therefore) achieves the goals. While it is ....

....This was fortunate because inferencing over a collection of facts (the world model) proved to be computationally complex [20] and slow in practice. In the mid 1980 s, Brooks [17] challenged both the world model assumptions of planners and the sense plan act model of robot architectures (e.g. [58]) with his subsumption architecture. In this architecture, there is no internal model of the world and thus there can be no time consuming deliberation process. This changes the notion of action selection from deciding upon a series of actions, to making a momentary decision on what to do now. ....

Nilsson, N. 1984. Shakey the Robot. Technical Note 323. SRI International. Menlo Park, California.

....of control theory, pattern recognition, and artificial intelligence, with AI being the centerpiece of the system. Unfortunately, a mobile robot is more than just the sum of its parts. The earliest application of this straightforward method such as Shakey the robot and its immediate descendants[Nilsson, 1984] all have a fatal flaw: They all assume that the world is easy to model. Shakey s planning system, STRIPS, is a straightforward implementation of means ends analysis[Newell and Simon, 1963] This seminal planning method uses a recursive goal decomposition method to break the mission goals down ....

....on external sources for its plans, whether that be a human hand coding a mission plan for a specific demo or a sophisticated AI planner generating plans as it responds to changes in environment and goals. One of the historical models of planning involves a separation of planning and execution [Nilsson, 1984]. A planner comes up with a complete plan, sends it to an executor which does its best to perform the plan, and reports success or failure to the planner. In this paradigm there is a conceptual wall between planning and execution, and many researchers have shown the limitations of this ....

Nilsson, N. (1984). Shakey the robot. Technical Report 323, SRI, Menlo Park, CA.

....ran into great di#culty in coping with dynamic environments in real time. Action selection was based on constructive planning, which was in turn based on heuristic search (Newell and Simon, 1981) This system has been both demonstrated and theoretically proven intractable without major alteration (Nilsson, 1984; Chapman, 1987) These results led researchers both inside and outside of artificial intelligence towards a new paradigm of reactive intelligence (George# and Lansky, 1987; Agre and Chapman, 1990; Maes, 1991; Rosenschein and Kaelbling, 1995; Hendriks Jansen, 1996; van Gelder, 1998; Bryson, in ....

Nilsson, N. (1984). Shakey the robot. Technical note 323, SRI International, Menlo Park, California.

....ran into great diculty in coping with dynamic environments in real time. Action selection was based on constructive planning, which was in turn based on heuristic search (Newell and Simon, 1981) This system has been both demonstrated and theoretically proven intractable without major alteration (Nilsson, 1984; Chapman, 1987) These results led researchers both inside and outside of arti cial intelligence towards a new paradigm of reactive intelligence (George and Lansky, 1987; Agre and Chapman, 1990; Maes, 1991; Rosenschein and Kaelbling, 1995; Hendriks Jansen, 1996; van Gelder, 1998; Bryson, in ....

Nilsson, N. (1984). Shakey the robot. Technical note 323, SRI International, Menlo Park, California.

.... as a robot architecture, but has proven so reliable it has also been used in manned space flight and air traffic control (Georgeff Lansky 1987) It was developed at roughly the same time as subsumption architecture, as a follow up program to the longest running robot experiment ever, Shakey (Nilsson 1984). PRS aims to fix problems with the traditional planning architectures exposed by the Shakey project. Such problems include: Forming a complete plan before beginning action. This is a necessary part of the search process underlying planning a planner cannot determine whether a plan is ....

....are not conflicting. But for examples see Bates et al. 1992) Grand et al. 1997) and Reilly (1996) Bates et al. 1992) in particular shares the concept of plan libraries with PRS. Another interesting related set of research is the evolution of the Shakey project prior to the development of PRS (Nilsson 1984). Although Shakey had a traditional planner (called STRIPS) over the term of the project the concept of triangle tables was developed. A triangle table is a version of a plan which decomposes it into its steps and assumptions, and allows the plan to be restarted from any point when perception ....

Nilsson, N. (1984), Shakey the robot, Technical note 323, SRI International, Menlo Park, California.

....but has proven sufficiently reliable to be used extensively for tasks such as aircraft maintenance and defense simulations. It was originally developed at roughly the same time as subsumption architecture, as a part of a followup program to the longest running robot experiment ever, Shakey (Nilsson, 1984). PRS is designed to fix problems with traditional planning architectures exposed by the Shakey project. Such problems include: Forming a complete plan before beginning action. This is a necessary part of the search process underlying planning a planner cannot determine whether a plan is ....

....(e. g Huber, 1999; d Inverno et al. 1997) The original development lab for PRS, SRI, is now focusing effort on a much more modularized AI architecture, built under a multi agent paradigm (Wilkins and Myers, 1998) The pre history of PRS, the Shakey project, also has relevant evolutionary trends (Nilsson, 1984). Although Shakey had a traditional planner (called STRIPS) over the term of the project the concept of triangle tables was developed. A triangle table decomposes a plan into its steps and assumptions, then creates a contingency table allowing the plan to be restarted from any point. Perception ....

[Article contains additional citation context not shown here]

Nilsson, N. (1984). Shakey the robot. Technical note 323, SRI International, Menlo Park, California.

....processing and so forth) Early AI driven robots took advantage of having large amounts of knowledge programmed into them in by designers who knew the domain. These designers programmed both the proximal and distal behaviour of their robots. For example the SRI International robot, SHAKEY (see (Nilsson, 1984)) moved about in a carefully engineered world. Analysed images from its onboard camera were used by a first order predicate calculus model of the world before the STRIPS planning system generated sequences of action. However, such robots seemed to have reached a ceiling in terms of what they could ....

Nilsson, N. (1984). Shakey the robot. Technical Note 323, SRI International, Menlo Park, California.

....white and color cameras. 1.2. Related Work Historically, Artificial Intelligence researchers have divided their investigation into separate studies of reason, perception and action. This phenomenon is very clear in the early robot programs, beginning with the SRI Shakey robot project (Nilsson [30]) and the Stanford Cart (Moravec [29] but it is still evident in more recent systems such as MAUV (Herman and Albus [20] Carnegie Mellon s NAVLAB (Thorpe, Hervert, Kanade and Shafer [35] the University of Maryland (a) b) c) Figure 3 Experiments have been performed using three different ....

Nilsson, Nils J. "Shakey the Robot". SRI Technical Note 323, 1984.

....problems of identifying, formalising, and representing knowledge [38] The emphasis on knowledge leads almost automatically to a focus on disembodied intelligence. Classical AI systems therefore do not include a physical body nor sensing or acting. If intelligent robots have been considered (as in [90], sensing and action has been delegated to subsystems which are assumed to deliver symbolic descriptions to the central planning and decision making modules. Moreover knowledge oriented theories do not include environmental pressures on the self preservation of the agent, and the role of ....

Nilsson, N. (ed.) (1984) Shakey the Robot. SRI AI center. Technical Note 323.

....ground. 2. Related Work While an extensive body of work exists for range based obstacle detection, little work has been done in appearancebased obstacle detection (Everett 1995) Interestingly, Shakey, the first autonomous mobile robot, used a simple form of appearance based obstacle detection (Nilsson 1984). Because Shakey operated on textureless floor tiles, obstacles were easily detected by applying an edge detector to the monochrome input image. However, Shakey s environment was artificial. Obstacles had non specular surfaces and were uniformly coated with carefully selected colors. In addition, ....

Nilsson, N.J. 1984. Shakey the Robot. Technical Note 323, SRI International.

No context found.

N.J. Nilsson, "Shakey the robot," Technical Report 223, SRI International, 1984.

No context found.

Nilsson, N. (1984). Shakey the robot. Technical Report 323, SRI International.

No context found.

N. J. Nilsson, "Shakey the robot," Tech. Rep. 323, Artificial Intelligence Center. SRI International, Menlo Park, USA, 1984.

No context found.

N.J. Nilsson, "Shakey the robot," Technical Report 223, SRI International, 1984.

No context found.

Nilsson N.J., (1984) "Shakey the robot", SRI A.I. Center Technical Note 323, April

No context found.

Nilsson, N.J., (1984). "Shakey the robot", SRI A.I. Center Technical Note 323, April

No context found.

Nilsson, N.J., "Shakey the robot", SRI A.I. Center Technical Note 323, April 1984

No context found.

Nilsson, N.J., (1984). "Shakey the robot", SRI A.I. Center Technical Note 323, April

No context found.

Nilsson, N. (1984). Shakey the robot. Technical report, A.I. Center Technical Note 323, SRI International.

No context found.

Nilsson, N. J. (1984). Shakey the Robot. Technical Report 323.

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