|
1. |
EDITORIAL: Biorobotics |
|
Connection Science,
Volume 10,
Issue 3-4,
1998,
Page 163-166
Preview
|
PDF (77KB)
|
|
ISSN:0954-0091
DOI:10.1080/095400998116387
出版商:Taylor & Francis Group
年代:1998
数据来源: Taylor
|
2. |
Evolutionary Robotics: Exploiting the Full Power of Self-organization |
|
Connection Science,
Volume 10,
Issue 3-4,
1998,
Page 167-184
Stefano Nolfi,
Preview
|
PDF (236KB)
|
|
摘要:
In this paper I claim that one of the main characteristics that makes the evolutionary robotics approach suitable for the study of adaptive behavior in natural and artificial agents is the possibility of relying largely on a self-organization process. Indeed, by using artificial evolution the role of the designer may be limited to the specification of a fitness function which measures the ability of a given robot to perform a desired task. From an engineering point of view, the main advantage of relying on self-organization is the fact that the designer does not need to divide the desired behavior into simple basic behaviors to be implemented in separate layers (or modules) of the robot control system. By selecting individuals for their ability to perform the desired behavior as a whole, simple basic behaviors can emerge from the interaction between several processes in the control system and from the interaction between the robot and the environment. From the point of view of the study of natural systems, the possibility of evolving robots that are free to select their way to solve a task by interacting with their environment may help us to understand how natural organisms produce adaptive behavior. Finally, the attempt to scale up to more complex tasks may help us to identify what the critical features of natural evolution are which allowed the emergence of the extraordinary variety of highly adapted life forms present on the planet.
ISSN:0954-0091
DOI:10.1080/095400998116396
出版商:Taylor & Francis Group
年代:1998
数据来源: Taylor
|
3. |
Better Living Through Chemistry: Evolving GasNets for Robot Control |
|
Connection Science,
Volume 10,
Issue 3-4,
1998,
Page 185-210
Phil Husbands,
Tom Smith,
Nick Jakobi,
Michael O'Shea,
Preview
|
PDF (553KB)
|
|
摘要:
This paper introduces a new type of artificial neural network (GasNets) and shows that it is possible to use evolutionary computing techniques to find robot controllers based on them. The controllers are built from networks inspired by the modulatory eff ects of freely diff using gases, especially nitric oxide, in real neuronal networks. Evolutionary robotics techniques were used to develop control networks and visual morphologies to enable a robot to achieve a target discrimination task under very noisy lighting conditions. A series of evolutionary runs with and without the gas modulation active demonstrated that networks incorporating modulation by diff using gases evolved to produce successful controllers considerably faster than networks without this mechanism. GasNets also consistently achieved evolutionary success in far fewer evaluations than were needed when using more conventional connectionist style networks.
ISSN:0954-0091
DOI:10.1080/095400998116404
出版商:Taylor & Francis Group
年代:1998
数据来源: Taylor
|
4. |
Evolution and Development of Modular Control Architectures for 1D Locomotion in Six-legged Animats |
|
Connection Science,
Volume 10,
Issue 3-4,
1998,
Page 211-237
Jerome Kodjabachian,
Jean-Arcady Meyer,
Preview
|
PDF (451KB)
|
|
摘要:
An evolutionary approach is used to design neural control architectures for virtual sixlegged animats. Using a geometry-oriented variation of the cellular encoding scheme and syntactic constraints that reduce the size of the genetic search space, the developmental programs of straight locomotion controllers are first evolved. One such controller is then included as the first module in a larger architecture, in which a second neural module is evolved and develops connections to the first one, so as to set locomotion on or offaccording to sustained or instantaneous external control signals. Such an incremental approach should prove useful to the automatic design of relatively complex control architectures that might, in particular, implement some cognitive abilities over and above mere stimulus-response mechanisms.
ISSN:0954-0091
DOI:10.1080/095400998116413
出版商:Taylor & Francis Group
年代:1998
数据来源: Taylor
|
5. |
Construction of a Hexapod Robot with Cockroach Kinematics Benefits both Robotics and Biology |
|
Connection Science,
Volume 10,
Issue 3-4,
1998,
Page 239-254
Roger D Quinn,
Roy E Ritzmann,
Preview
|
PDF (515KB)
|
|
摘要:
Any device that moves by actuating several multi-segmented legs must solve fundamental problems in mechanics and control regardless of whether that device is made out of living tissue or metal. With this in mind, we believe that advances can be made both in the design of legged robots and in understanding how legged animals locomote by working on these issues in tandem. This basic philosophy has led us to build a hexapod robot with kinematics that are remarkably similar to those of the death head cockroach Blaberus discoidalis . This eff ort has required us to make detailed neurobiological and kinematic observations of cockroaches walking on a treadmill and climbing over barriers. The data acquired in this way were then input to our dynamic simulation tool to determine mechanically accurate parameters for the design of the robot. The resulting vehicle captures the cockroach kinematics so well that issues that are being addressed in controlling it are providing important notions regarding the neural control of locomotion in the animal. Thus, the project has led to benefits in both directions. Biological data inspired the initial design of the robot, while engineering eff orts in control are inspiring further neurobiological experiments in animal locomotion.
ISSN:0954-0091
DOI:10.1080/095400998116422
出版商:Taylor & Francis Group
年代:1998
数据来源: Taylor
|
6. |
Rapid Navigational Learning in Insects with a Short Lifespan |
|
Connection Science,
Volume 10,
Issue 3-4,
1998,
Page 255-270
Thomas S Collett,
Preview
|
PDF (289KB)
|
|
摘要:
The remarkable navigational skills of ants, bees and wasps are partly due to their ability to learn the appearance of features of their surroundings. This paper discusses strategies employed by these insects to ensure rapid and efficient learning.
ISSN:0954-0091
DOI:10.1080/095400998116431
出版商:Taylor & Francis Group
年代:1998
数据来源: Taylor
|
7. |
Mammalian Navigation, Neural Models and Biorobotics |
|
Connection Science,
Volume 10,
Issue 3-4,
1998,
Page 271-289
A. S Etienne,
Preview
|
PDF (263KB)
|
|
摘要:
Within their home range, rodents and other mammals navigate by updating their position through (internal) signals derived from locomotion and by using landmark-place associations. Single unit recordings in freely moving rats show that motion cues and external, mainly visual information act on place and head direction cells in the hippocampus and functionally related brain areas. Currently, a number of partially converging and diverging neural network models aim at representing the neural system that controls rodent navigation. It is argued that the attempt to build self-steering robots on the basis of behavioural and neurophysiological results from rodents and of computer simulations of these results is premature, given the lack of consensus in the literature on mammalian navigation.
ISSN:0954-0091
DOI:10.1080/095400998116440
出版商:Taylor & Francis Group
年代:1998
数据来源: Taylor
|
8. |
Using a Mobile Robot to Test a Model of the Rat Hippocampus |
|
Connection Science,
Volume 10,
Issue 3-4,
1998,
Page 291-300
Neil Burgess,
James G Donnett,
John O'Keefe,
Preview
|
PDF (260KB)
|
|
摘要:
A model of how internal and external sensory information contribute to the firing of place cells in the rat hippocampus, and of how these cells contribute to the rat's spatial behavior, is tested on a miniature mobile robot. The experiments show that crude visual, odometric and short-range proximity information provided by sensors on the robot is sufficient to enable the formation of a robust spatial code within rectangular environments. They further show that the model of navigation can accurately return the robot to an unmarked goal location. Since the rat's perceptual systems are probably similarly crude, these results support our intuition that the model of hippocampal function is reasonable in not demanding too much of its inputs. The combined robotic and neuronal simulation can be used to make predictions regarding both electrophysiological and behavioral experiments. Finally, the model is applied to the neural basis of navigation in humans.
ISSN:0954-0091
DOI:10.1080/095400998116459
出版商:Taylor & Francis Group
年代:1998
数据来源: Taylor
|
9. |
Cerebellar Control of Robot Arms |
|
Connection Science,
Volume 10,
Issue 3-4,
1998,
Page 301-320
Patrick Van Der Smagt,
Preview
|
PDF (315KB)
|
|
摘要:
Decades of research into the structure and function of the cerebellum have led to a clear understanding of many of its cells, as well as how learning takes place. Furthermore, there are many theories on what signals the cerebellum operates on, and how it works in concert with other parts of the nervous system. Nevertheless, the application of computational cerebellar models to the control of robot dynamics remains in its infant state. To date, a few applications have been realized, but limited to the control of traditional robot structures which, strictly speaking, do not require adaptive control for the tasks that are performed since their dynamic structures are relatively simple. The currently emerging family of light-weight robots (Hirzinger, G. (1996) In Proceedings of the 2nd International Conference on Advanced Robotics, Intelligent Automation, and Active Systems, Vienna, Austria ) poses a new challenge to robot control: owing to their complex dynamics, traditional methods, depending on a full analysis of the dynamics of the system, are no longer applicable since the joints influence each other's dynamics during movement. Can artificial cerebellar models compete here? In this paper, we present a succinct introduction of the cerebellum, and discuss where it could be applied to tackle problems in robotics. Without conclusively answering the above question, an overview of several applications of cerebellar models to robot control is given.
ISSN:0954-0091
DOI:10.1080/095400998116468
出版商:Taylor & Francis Group
年代:1998
数据来源: Taylor
|
10. |
Shaping: The Link Between Rats and Robots |
|
Connection Science,
Volume 10,
Issue 3-4,
1998,
Page 321-340
Tony Savage,
Preview
|
PDF (266KB)
|
|
摘要:
The term shaping applies to a family of procedures which were originally developed in animal learning as methods of producing new and sometimes elaborate forms of behavior. The success of operant shaping methods in animal and abnormal psychology has attracted the attention of some roboticists who view these procedures as a means of generating adaptive behavior in both simulated and real robots. This paper outlines the two principal forms of shaping and looks at some of the associative and motivational influences which determine the course and outcome of shaping methods in animals. A number of recent robotic models which incorporate operant shaping concepts are then reviewed in terms of their eff ectiveness; the conclusion from this analysis is that, while some progress has been made, none of these models represent a satisfactory application of animal shaping procedures. It is clear that an effective implementation of shaping methods in behaviorbased robotics requires both a better understanding of the nature and course of shaping in real animals and the development, by roboticists and others, of more sophisticated models which can exploit this improved understanding.
ISSN:0954-0091
DOI:10.1080/095400998116477
出版商:Taylor & Francis Group
年代:1998
数据来源: Taylor
|
|