ADAPTIVE AND PREDICTIVE CONTROL OF A SIMULATED ROBOT ARM

Silvia Tolu, Mauricio Vanegas, Jesus A. Garrido, Niceto R. Luque, Eduardo Ros

Research output: Contribution to journalJournal articleResearchpeer-review

Abstract

In this work, a basic cerebellar neural layer and a machine learning engine are embedded in a recurrent loop which avoids dealing with the motor error or distal error problem. The presented approach learns the motor control based on available sensor error estimates (position, velocity, and acceleration) without explicitly knowing the motor errors. The paper focuses on how to decompose the input into different components in order to facilitate the learning process using an automatic incremental learning model (locally weighted projection regression (LWPR) algorithm). LWPR incrementally learns the forward model of the robot arm and provides the cerebellar module with optimal pre-processed signals. We present a recurrent adaptive control architecture in which an adaptive feedback (AF) controller guarantees a precise, compliant, and stable control during the manipulation of objects. Therefore, this approach efficiently integrates a bio-inspired module (cerebellar circuitry) with a machine learning component (LWPR). The cerebellar-LWPR synergy makes the robot adaptable to changing conditions. We evaluate how this scheme scales for robot-arms of a high number of degrees of freedom (DOFs) using a simulated model of a robot arm of the new generation of light weight robots (LWRs).
Original languageEnglish
Article number1350010
JournalInternational Journal of Neural Systems
Volume23
Issue number3
Number of pages15
ISSN0129-0657
DOIs
Publication statusPublished - 2013
Externally publishedYes

Keywords

  • Adaptation, Physiological
  • Algorithms
  • Arm
  • Artificial Intelligence
  • Cerebellum
  • Computer Simulation
  • Feedback
  • Humans
  • Models, Neurological
  • Movement
  • Neurons
  • Predictive Value of Tests
  • Robotics
  • COMPUTER
  • MOTOR CONTROL
  • INTERNAL-MODELS
  • CEREBELLAR CONTROL
  • PURKINJE-CELLS
  • REGRESSION
  • CORTEX
  • Light weight robot
  • recurrent control architecture
  • locally weighted projection regression
  • adaptive learning
  • Computer Networks and Communications
  • Adaptive control architecture
  • Adaptive learning
  • Control architecture
  • Incremental learning
  • Light weight robots
  • Locally weighted projection regressions
  • Number of degrees of freedom
  • Predictive control
  • Learning algorithms
  • Learning systems
  • Robotic arms
  • Computer simulation
  • adaptation
  • algorithm
  • arm
  • article
  • artificial intelligence
  • biological model
  • cerebellum
  • computer simulation
  • cytology
  • feedback system
  • human
  • movement (physiology)
  • nerve cell
  • physiology
  • predictive value
  • robotics
  • ROBOTS

Cite this