6533b830fe1ef96bd129659a

RESEARCH PRODUCT

Online Deflection Compensation of a Flexible Hydraulic Loader Crane Using Neural Networks and Pressure Feedback

Konrad Johan JensenMorten Kjeld EbbesenMichael Rygaard Hansen

subject

VDP::Teknologi: 500Control and OptimizationArtificial IntelligenceMechanical EngineeringPhysics::Space Physicsdeflection compensation; kinematics; loader crane; hydraulics; neural network

description

The deflection compensation of a hydraulically actuated loader crane is presented. Measurement data from the laboratory are used to design a neural network deflection estimator. Kinematic expressions are derived and used with the deflection estimator in a feedforward topology to compensate for the static deflection. A dynamic deflection compensator is implemented, using pressure feedback and an adaptive bandpass filter. Simulations are conducted to verify the performance of the control system. Experimental results showcase the effectiveness of both the static and dynamic deflection compensator while running closed-loop motion control, with a 90% decrease in static deflection.

yearjournalcountryeditionlanguage
2022-03-17Robotics
https://doi.org/10.3390/robotics11020034