Next Article in Journal
Research on Game-Playing Agents Based on Deep Reinforcement Learning
Next Article in Special Issue
Design and Scaling of Exoskeleton Power Units Considering Load Cycles of Humans
Previous Article in Journal
Gait Analysis for a Tiltrotor: The Dynamic Invertible Gait
Previous Article in Special Issue
Gait Transition from Pacing by a Quadrupedal Simulated Model and Robot with Phase Modulation by Vestibular Feedback
 
 
Article
Peer-Review Record

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

by Konrad Johan Jensen *, Morten Kjeld Ebbesen and Michael Rygaard Hansen
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Reviewer 3: Anonymous
Submission received: 21 February 2022 / Revised: 14 March 2022 / Accepted: 16 March 2022 / Published: 17 March 2022
(This article belongs to the Special Issue Mechatronics Systems and Robots)

Round 1

Reviewer 1 Report

This paper is well written and can be accepted after minor revisions. Comments are listed following: The crane considered in this work has a similar mechanism to robot arms and works on using neural networks for robot arm control may be of reference values to this paper, e.g., distributed recurrent neural networks for cooperative control of manipulators: a game-theoretic perspective,kinematic control of redundant manipulators using neural networks,A novel recurrent neural network for manipulator control with improved noise tolerance,A dynamic neural network approach for efficient control of manipulators. Please discuss the possibility to use above dynamic neural networks for the control of the crane and possible pros and cons in comparison with the presented approach in the introduction part. Please provide detailed data for the experimental results. Also, preferably, a demonstration video will help reviewers to see the real performance. 

Author Response

See PDF file

Author Response File: Author Response.pdf

Reviewer 2 Report

This manuscript proposes a novel method for deflection compensations using based on a shallow neural network.

Minor comments, suggestions:

  • Abstract and conclusion should contain quantitative results: show how the proposed method performs to other possible methods;
  • The proposed neural network for deflection estimation is a classical architecture, the thorough description is unnecessary;
  • Have you considered other machine learning based methods for estimation? Did you compare to the presented MLP?
  • Did you experiment with hyperparameter tuning, e.g. number of hidden layer neurons, activation, different regularization methods and parameters?

In my opinion, it is really good to see that the simulated results were verified in a real life experiment, the manuscript is a lot more confident.

Author Response

See PDF file

Author Response File: Author Response.pdf

Reviewer 3 Report

1. This well-written paper presents an interesting deflection compensation scheme for a hydraulic loader crane. The authors offer thorough kinematics and actuator analysis and show the neural network deflection estimator. 

2. It will be nice to see more experimental results conducted on the hydraulic loader crane.

Author Response

See PDF file

Author Response File: Author Response.pdf

Back to TopTop