Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: May 10, 2026

A Flexible Platform for Monitoring Cerebellum-Dependent Sensory Associative Learning

A Flexible Platform for Monitoring Cerebellum-Dependent Sensory Associative Learning

Published on: January 19, 2022

Filtering sensory information with XCSF: improving learning robustness and robot arm control performance.

Jan Kneissler¹, Patrick O Stalph, Jan Drugowitsch

¹Department of Computer Science, University of Tübingen, Sand 14, 72076 Tübingen, Germany jan.kneissler@uni-tuebingen.de.

Evolutionary Computation

|June 11, 2013

Summary

This study enhances robot arm control by integrating Kalman filtering with XCSF (eXternal Classifier System) for improved sensor noise tolerance. The combined approach significantly boosts learning and control performance in noisy environments.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Distributed control circuits across a brain-and-cord connectome.

Nature·2026

Same author

Competitive integration of time and reward explains value-sensitive foraging decisions and frontal cortex ramping dynamics.

Neuron·2025

Same author

Distributed control circuits across a brain-and-cord connectome.

bioRxiv : the preprint server for biology·2025

Same author

A unifying theory of receptive field heterogeneity predicts hippocampal spatial tuning.

bioRxiv : the preprint server for biology·2025

Same author

Stochastic Optimal Control and Estimation with Multiplicative and Internal Noise.

Advances in neural information processing systems·2025

Same author

Multimodal cue integration and learning in a neural representation of head direction.

Nature neuroscience·2025

Area of Science:

Robotics
Machine Learning
Control Systems

Background:

Robot arm control traditionally uses learning classifier systems like XCSF.
Exploiting predictive knowledge of motor activity for state changes remains underexplored.
Noisy sensors often degrade the performance of robot control systems.

Purpose of the Study:

To integrate forward velocity kinematics from XCSF with Kalman filtering for robust robot arm control.
To mitigate the impact of sensor noise on learning and control performance.
To enhance trajectory planning and kinematic learning using filtered sensory data.

Main Methods:

Utilized XCSF's forward velocity kinematics for predictions.
Incorporated Kalman filtering to estimate arm positions by combining sensory data and XCSF predictions.

More Related Videos

SSVEP-based Experimental Procedure for Brain-Robot Interaction with Humanoid Robots

SSVEP-based Experimental Procedure for Brain-Robot Interaction with Humanoid Robots

Published on: November 24, 2015

An Experimental Platform to Study the Closed-loop Performance of Brain-machine Interfaces

An Experimental Platform to Study the Closed-loop Performance of Brain-machine Interfaces

Published on: March 10, 2011

Related Experiment Videos

Last Updated: May 10, 2026

A Flexible Platform for Monitoring Cerebellum-Dependent Sensory Associative Learning

A Flexible Platform for Monitoring Cerebellum-Dependent Sensory Associative Learning

Published on: January 19, 2022

SSVEP-based Experimental Procedure for Brain-Robot Interaction with Humanoid Robots

SSVEP-based Experimental Procedure for Brain-Robot Interaction with Humanoid Robots

Published on: November 24, 2015

An Experimental Platform to Study the Closed-loop Performance of Brain-machine Interfaces

An Experimental Platform to Study the Closed-loop Performance of Brain-machine Interfaces

Published on: March 10, 2011

Tested the combined approach on a simulated kinematic robot arm model.

Introduced a heuristic parameter to limit self-prediction influence on learning.

Main Results:

The integration of Kalman filtering and XCSF significantly improved learning and control performance.
The system demonstrated enhanced noise tolerance, handling over 10 times higher noise levels.
Underestimated variance in XCSF predictions could lead to self-delusional spiraling effects, hindering learning.

Conclusions:

Combining Kalman filtering with XCSF offers a robust solution for robot arm control in noisy environments.
The heuristic parameter is crucial for preventing learning degradation due to prediction inaccuracies.
This approach substantially increases the system's resilience to sensor noise.