Abstract
The use of locally weighted regression in memory-based robot learning is explored. A local model is formed to answer each query, using a weighted regression in which close points (similar experiences) are weighted more than distant points (less relevant experiences). This approach implements a philosophy of modeling a complex function with many simple local models. The author explains how an appropriate distance metric or measure of similarity can be found, and how the distance metric is used. How irrelevant input variables and terms in the local model are detected is also explained. An example from the control of a robot arm is used to compare this approach with other robot control and learning techniques.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">></ETX>
Keywords
Metric (unit)Similarity (geometry)Artificial intelligenceRobotComputer scienceRegressionFunction (biology)Regression analysisSimple (philosophy)Machine learningMathematicsStatisticsEngineering
Affiliated Institutions
Related Publications
Path loss and delay spread models as functions of antenna height for microcellular system design
Results of wideband path loss and delay spread measurements for two representative microcellular environments in the San Francisco Bay area in the 1900 MHz band are presented. T...
Perfect metrics
The authors describe an experiment in the construction of perfect metrics for minimum-distance classification of character images. A perfect metric is one that, with high probab...
A numerical solution to the generalized mapmaker's problem: flattening nonconvex polyhedral surfaces
Methods are described to unfold and flatten the curved, convoluted surfaces of the brain in order to study the functional architectures and neural maps embedded in them. In orde...
Near-optimal nonholonomic motion planning for a system of coupled rigid bodies
How does a falling cat change her orientation in midair without violating angular momentum constraint? This has become an interesting problem to both control engineers and robot...
Stabilizing I-O receding horizon control of CARMA plants
Stabilizing input-output receding horizon control (SIORHC) yields dynamic feedback compensators capable of stabilizing any stabilizable linear plant under sharp conditions. The ...
Publication Info
- Year
- 2002
- Type
- article
- Pages
- 958-963
- Citations
- 108
- Access
- Closed
External Links
Social Impact
Altmetric
PlumX Metrics
Social media, news, blog, policy document mentions
Citation Metrics
108
OpenAlex
Cite This
Christopher G. Atkeson
(2002).
Using locally weighted regression for robot learning.
, 958-963.
https://doi.org/10.1109/robot.1991.131713
Identifiers
- DOI
- 10.1109/robot.1991.131713