Brooks, R. A. (1989). A robot that walks: Emergent behaviors from a carefully evolved network. Technical Report AI MEMO 1091, MIT.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....needed. In order to generate personal semantic evaluations we have to extend our very simple embodied agent from above (Fig. 2) This agent is reflex based (reactive) and its structure is that of a purely syntactic system. Now we design a so called subsumption architecture of loops upon loops [5] (see Fig. 3) This way we 14 will argue meaning (semantics) can be objectively (i.e. in a measurable way) constructed by the intrinsic observation process performed by the outer loop(s) on the inner loop(s) When the agent has come to life there is only the innermost loop working. This ....

Rodney A Brooks. A robot that walks; emergent behaviors from a carefully evolved network. Technical Report 1091, MIT AI Lab, 1989.

....planning [Brooks, 1986; Agre and Chapman, 1987] is the general approach of building systems that make the right decision by design. Agents based on these ideas are normally built from combinatorial circuits plus a small timing circuitry. The combinatorial circuitry may be adaptive, as in [Brooks, 1989] , with the capability to converge on the desired behavior after a certain training period of interaction with the environment. Universal planning [Schoppers, 1987] is a similar approach which is, in a sense, more restrictive than reactive planning since it does not allow for emergence of the ....

R. A. Brooks. A robot that walks: Emergent behavior from a carefully evolved network. Neural Computation,1(2):253--262, 1989.

....soldiers. Tu and Terzopoulos [7] created autonomous fishes living in a physically modeled virtual marine world. Hodgins et al. 8] described dynamic athletic behaviors. Unuma et al. 9] modeled human figure locomotion with emotions. Other papers presented behavioral architectures. Brooks [10] [11] developed a subsumption architecture built in layers of behaviors. Maes [12] presented an action selection model where action selection is an emergent property of activation inhibition dynamics among actions. Beer et al. 13] developed neural networks for insects. Ahmad et al. 14] described ....

Brooks, R.A.. "A Robot That Walks; Emergent Behaviors from a Carefully Evolved Network. MIT AI Lab Memo 1091, February 1989.

....terminate at the source of another wire (an output site ) Any message flowing in the extra wire will not only inhibit the other wire, but will be inserted on to it, suppressing and replacing the normal dataflow. A variant of this type of connection appears in later versions of Subsumption [[9]] called defaulting, where priority is given to the original wire, with dataflow from the new wire only being accepted when there is no data on the original wire for a given time. Augmented Finite State Machine s d i Suppressing Defaulting Inhibiting Inputs Outputs Figure 2 9: The use of ....

Brooks R.A.: A Robot That Walks: Emergent Behaviors from a Carefully Evolved Network, AI-Laboratory, Massachusetts Institute of Technology, Cambridge, MA, AIMemo 1091, 1989.

....each one separately from others. The left and right avoid obstacles behavior modules are based on the following pairs of fuzzy variables: sensor[O] sensor[1] and (sensor[4] sensor[5] respec tively. The avoid front obstacles behavior module receives as inputs the pair (max(sensor[2] sensor[3]) sensor[2] sensor[3] the first being a measure of the distance to front obstacles and the second a measure of the orientation of the robot in relation to the obstacle. Every fuzzy variable representing distance is defined by three terms (mentioned earlier) while the orientation fuzzy ....

....others. The left and right avoid obstacles behavior modules are based on the following pairs of fuzzy variables: sensor[O] sensor[1] and (sensor[4] sensor[5] respec tively. The avoid front obstacles behavior module receives as inputs the pair (max(sensor[2] sensor[3] sensor[2] sensor[3]) the first being a measure of the distance to front obstacles and the second a measure of the orientation of the robot in relation to the obstacle. Every fuzzy variable representing distance is defined by three terms (mentioned earlier) while the orientation fuzzy variable is defined by five ....

Rodney Brooks. A Robot That Walks: Emergent Behaviors From a Carefully Evolved Network. In Neural Computation, 1:2, pages 253 262, Summer, 1989.

....under the assumption of fully modeled, deterministic robot and robot environments. This changed radically in the mid1980s, when the paradigm shifted towards reactive techniques. Approaches such as Brooks s behavior based architecture generated control directly in response to sensor measurements [4]. Rejections of models quickly became typical for this approach. Reactive techniques were arguable as limited as model based ones, in that they replaced the unrealistic assumption of perfect models by an equally unrealistic one of perfect perception. Since the mid 1990s, robotics has shifted its ....

R. A. Brooks. A robot that walks; emergent behaviors from a carefully evolved network. Neural Computation, 1(2):253, 1989.

....the field of robotics, researchers have proposed alternative languages and methodologies for programming robots. None of these approaches integrates learning at the architectural level, and none supports computation with uncertain information. For example, Brooks s popular subsumption architecture [9, 10] 28 provides a modular way for programming robots, by coupling together finite state machines that map sensor readings more or less directly into motor commands. Unfortunately, this approach does not address the uncertainty typically arising in robotic domains, and as a consequence it fails to ....

R. A. Brooks. A robot that walks; emergent behaviors from a carefully evolved network. Neural Computation, 1(2):253, 1989.

....accurate and complete. Recognizing this limitation, some researchers have advocated model free reactive approaches. Instead of relying on planning, these approaches require programmers to program controllers directly. A popular example of this approach is the subsumption architecture [12], where controllers are composed of small finite state automata that map sensor readings into control while retaining a minimum of internal state. Some advocates of this approach went as far as refusing the need for internal models and internal state altogether [12, 20] Observing that the world ....

....the subsumption architecture [12] where controllers are composed of small finite state automata that map sensor readings into control while retaining a minimum of internal state. Some advocates of this approach went as far as refusing the need for internal models and internal state altogether [12, 20]. Observing that the world is its own best model [13] behavior based approaches usually rely on immediate sensor feedback for determining a robot s action. Obvious limits in perception (e.g. robots cannot see through walls) pose clear boundaries on the type of tasks that can be tackled with ....

R. A. Brooks. A robot that walks; emergent behaviors from a carefully evolved network. Neural Computation, 1(2):253, 1989.

....The robot device must be precise enough to accurately execute plans that were generated using the internal models of the world. Recent research on autonomous robots has changed the design of autonomous agents and pointed out a promising direction for future research in robotics (see for exam ple [Brooks, 1989] and several papers in [Maes, 1991] Reactivity and real time operation have received considerably more attention than, for example, optimality and completeness. Many approaches have dropped the assumption that perfect world knowledge is available some systems even operate in the extreme where no ....

Rodney A. Brooks. A robot that walks; emergent behaviors from a carefully evolved network. Neural Computation, 1(2):253, 1989.

....[18] develop a gait controller and low level motor programs to generate legged motion. Zeltzer [26] analyzes various approaches to the control of complex animated objects and considers their integration. Raibert and Hodgins [22] describe control systems for several legged creatures. Brooks [4] produces complex walking behavior in a physical, insect like robot from a distributed network of low level finite state machines. Other Work in Robotics Some further insights on control can be gained from examining the literature in the field of robotics. While this field deals with controlling ....

R. A. Brooks. A robot that walks: emergent behaviors from a carefully evolved network. In Proceedings of the

....properties, such as safety and liveness. This approach allows one to specify formally, and verify, models of embedded control systems. Our goal is to develop it as a practical tool for building real, complex, sensor based robots. It can be seen as a development of Brooks Subsumption Architecture [Bro88] that enhances its modular advantages while avoiding the limitations of the augmented finite state machine approach. 9 The names are inspired by Monty Python s soccer playing philosophers sketch. Zeno and Heraclitus were particularly concerned with dynamic worlds. 10 In fact, the Clod Buster ....

....behaviour based approach is common to much of the work on the situated agents. However, we feel that the current mechanisms for arbitrating among behaviours are inadequate. Our method for dynamic action selection is based on inter behaviour bidding, an approach more general than other approaches [Bro88; KR90] 5 The main idea behind this theory of action selection is that each behaviour is best able to identify how applicable it is in a given situation. Each behaviour independently evaluates the world and reports a utility estimate or bid to the other behaviours. The behaviour with the ....

R. A. Brooks. A robot that walks: emergent behaviors from a carefully evolved network. Massachusetts Institute of Technology, Cambridge, MA, 1988.

....and action modules of GOFAIR not only can t be built but also do not correspond to natural scientific domains with clean interfaces and limited interaction amongst them. Perception, reasoning and action correspond only to labels that we use to caricature aspects of the agent s behavior. Brooks [8] has correctly pointed out that the traditional divide andconquer AI approach to robotics, by slicing intelligence into perception, reasoning and action, has pursued a strategy that does not scale up. This, incidentally, implies that any research program based on that division will be sterile. ....

....on provable properties of its subsystems. This approach allows one to specify formally, and verify, models of embedded control systems. Our goal is to develop it as a practical tool for building real, complex, sensor based robots. It can be seen as a development of Brooks subsumption architecture [8] that enhances its modular advantages while avoiding the limitations of the augmented finite state machine approach. A robot situated in an environment is modeled as three machines: the robot plant, the robot control and the environment. Each is modeled separately as a dynamical system by ....

R. A. Brooks, A robot that walks: emergent behaviors from a carefully evolved network. Cambridge, MA: Massachusetts Institute of Technology, 1988.

....of their behavior is internally generated, and how much emerges from the interaction of simple response patterns within a complex task environment. It seems doubtful that true progress in understanding the properties of intelligent behavior can be made by studying disembodied, syntactic systems [16, 17, 1, 8, 18]. Intelligent behavior, at least in biological organisms, seems built upon a foundation of fast and robust pattern recognition and completion, both of static and temporally extended, often vague and noisy patterns. This observation is suggestive of several features that may be necessary in the ....

Rodney A. Brooks. A robot that walks: Emergent behaviors from a carefully evolved network. In Randall Beer, R. Ritzmann, and T. McKenna, editors, Biological Neural Networks in Invertebrate Neuroethology and Robotics. Academic Press, 1993.

....Therefore, they must be embodied and situated in the real world. These beliefs have driven the design and implementation of our robots. It may be noted that embodying robots in the early prototyping stages is a method that has gained considerable support in recent years. Prominent examples include [3, 5, 4] and the various Navlab vehicles [10] used for off road as well as highway tests. Other examples of successful integration and cross platform validation using hardware platforms include the Dante robot built at CMU for exploring volcanic craters, which went through two iterations and the Rocky ....

R. A. Brooks. A robot that walks: Emergent behaviors from a carefully evolved network. Neural Computation, 1(2):365--382, 1989.

....can be easily modified. For example, pairs of legs and actuators can be replicated to produce a mechanical centipede with flexible joints between leg segments. The robot is intended for use as a research and educational platform to study computational sensors [3, 4] subsumption architectures [5], neural gait control [6] behavior of social insects [7] and machine vision [8] The robot may be powered and controlled through a tether, or autonomously with on board power supply and electronics [9] It is capable of carrying up to 50 grams while walking at a speed of 3 to 10 centimeters per ....

Brooks, R. 1990. A robot that walks: Emergent behaviors from a carefully evolved network. Neural Computation 1 (2): pp. 253-262.

No context found.

Brooks, R. A. (1989). A robot that walks: Emergent behaviors from a carefully evolved network. Technical Report AI MEMO 1091, MIT.

No context found.

Brooks, R. A. (1989). "A Robot That Walks: Emergent Behaviors from a Carefully Evolved Network." Neural Computation (pp. 254-262).

No context found.

Brooks, R., A Robot that Walks; Emergent Behaviors from a Carefully Evolved Network, Neural Computation, Vol. 1, No. 2, 1989, pp. 253--262.

No context found.

Brooks R. A.: A Robots that Walks: EmergentBehaviors from a Carefully Evolved Network. Neural Computation 1:2, Summer, (1989)

No context found.

Brooks, R. A.: A Robots that Walks: Emergent Behaviors from a Carefully Evolved Network. Neural Computation 1:2, Summer, 1989.

No context found.

Brooks, R. A.: A Robots that Walks: Emergent Behaviors from a Carefully Evolved Network. Neural Computation 1:2, Summer, 1989.

No context found.

R. A. Brooks. A robot that walks; emergent behaviors from a carefully evolved network. Neural Computation, 1(2):253-262, 1989.

No context found.

Brooks R. A.: A Robots that Walks: EmergentBehaviors from a Carefully Evolved Network. Neural Computation 1:2, Summer, (1989)

No context found.

R.A. Brooks, 'A Robot that Walks; Emergent Behaviors from a Carefully Evolved Network', Neural Computation, Vol 1, pp 253-262, 1989.

No context found.

R. A. Brooks. A robot that walks; emergent behaviors from a carefully evolved network. In Neural Computation, volume 1, pages 253--262. 1989.

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