Learning the motion map of a robot arm with neural networks /
Abstract: "This paper discusses the integration of neural self- organization and circular reaction into a robot guidance system (Neurobot). A two degree of freedom robot arm 'learns' a cognitive map which contains both the visual workspace and also the robot's joint angles in a...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Book |
| Language: | English |
| Published: |
College Station, Tex. :
Texas A & M University, Computer Science Dept.,
[1990]
|
| Series: | Technical report (Texas A & M University. Computer Science Department) ;
90-008. |
| Subjects: |
| Summary: | Abstract: "This paper discusses the integration of neural self- organization and circular reaction into a robot guidance system (Neurobot). A two degree of freedom robot arm 'learns' a cognitive map which contains both the visual workspace and also the robot's joint angles in a biologically inspired neural network. The range of positions of the robot's end effector is called the workspace, the corresponding joint angle space is called the configuration space. In other words, the workspace is the physical world of the robot while the configuration space is an abstract space necessary for controlling the arm motions. Neurobot creates an association between its visual position and its joint position by training a self-organizing neural network using both spaces as inputs. This position-map is thus a hypersurface represented by a sheet of neurons in which each neuron represents a point in both the workspace and the configuration space. The position-map can be considered sensory-motor schema in the sense that it can now be used for robot guidance without using computationally expensive transformation equations. Rather, a visual input target would cause a spreading activation to the present position. Further, it can also perform efficient obstacle avoidance where visual input obstacles inhibit or gate the correpsonding joint angles. Thus, it can be used for minimal computation robot guidance, obstacle avoidance, and most importantly, is learned by a self-organization process governed by a circular feedback loop." |
|---|---|
| Item Description: | "January 1990." |
| Physical Description: | 6 leaves : illustrations ; 28 cm. |
| Bibliography: | Includes bibliographical references. |