Pose-dependent tool tip dynamics prediction using transfer learning

Chen, Gengxiang; Li, Yingguang; Liu, Xu

doi:10.1016/j.ijmachtools.2018.10.003

Cited by 64 publications

(21 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this work, the original ANN with no payload (source) is adapted to the case with 3 kg payload (target) using training data at only 45 strategic poses. As illustrated in Figure 7, these strategic points have been selected at the local and global extremums of the distribution map (Chen et al, 2019). To be able to adapt the model with only 45 strategic poses means a 90% reduction in training data compared to training the ANN from scratch at 3 kg.…”

Section: Learning-based Framework For Prediction Of Natural Frequencymentioning

confidence: 99%

“…The new ANN can then be tuned with much less training data. This technique, called transfer learning (Chen et al, 2019), is used in this research to adapt a robot model to new situations. Alternatively, transfer learning can be used to map an ANN trained based on theoretical models to the case of real robots using only a few experimental tests (Tercan et al, 2018).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Suppression of robot vibrations using input shaping and learning-based structural models

Newman

Khoshdarregi

2020

Journal of Intelligent Material Systems and Structures

View full text Add to dashboard Cite

Industrial robots used in manufacturing processes such as drilling of aerospace structures undergo many rapid positioning motions during each operation. Such aggressive motions excite the structural modes of the robot and cause inertial vibrations at the end-effector, which may damage the part and violate the tolerance requirements. This article presents a vibration avoidance technique based on input shaping combined with a learning-based structural dynamic model. A theoretical dynamic model is first developed for commonly used robotic arms considering the flexibilities of the first three joints of the robot. An artificial neural network is developed and used in conjunction with the dynamic model to predict the natural frequency of the system at any pose in the workspace. Transfer learning techniques are used to extend the trained artificial neural network to account for the mass of the payload with minimal data collection. To reduce the residual vibrations of the robot in rapid motions, zero-vibration derivative shapers are designed and implemented. The effectiveness of the presented methodology has been validated experimentally on a Staubli RX90CR robot with an open-architecture control system developed fully in-house. Experimental results show more than 85% reduction in residual vibrations during aggressive motions of the robot.

show abstract

Section: Learning-based Framework For Prediction Of Natural Frequencymentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Suppression of robot vibrations using input shaping and learning-based structural models

Newman

Khoshdarregi

2020

Journal of Intelligent Material Systems and Structures

View full text Add to dashboard Cite

show abstract

“…The recent emergence of transfer learning appears to overcome this problem in manufacturing. Transfer-learning applications are increasing and include fault detection and condition causality in product quality management, fault diagnosis and condition-based maintenance in machine maintenance, and tool tip dynamics prediction in machine chatter [42,43].…”

Section: Learning-based Analyticsmentioning

confidence: 99%

A Holonic-Based Self-Learning Mechanism for Energy-Predictive Planning in Machining Processes

2019

View full text Add to dashboard Cite

The present work proposes a holonic-based mechanism for self-learning factories based on a hybrid learning approach. The self-learning factory is a manufacturing system that gains predictive capability by machine self-learning, and thus automatically anticipates the performance results during the process planning phase through learning from past experience. The system mechanism, including a modeling method, architecture, and operational procedure, is structured to agentize machines and manufacturing objects under the paradigm of Holonic Manufacturing Systems. This mechanism allows machines and manufacturing objects to acquire their data and model interconnection and to perform model-driven autonomous and collaborative behaviors. The hybrid learning approach is designed to obtain predictive modeling ability in both data-existent and even data-absent environments via accommodating machine learning (which extracts knowledge from data) and transfer learning (which extracts knowledge from existing knowledge). The present work also implements a prototype system to demonstrate automatic predictive modeling and autonomous process planning for energy reduction in milling processes. The prototype generates energy-predictive models via hybrid learning and seeks the minimum energy-using machine tool through the contract net protocol combined with energy prediction. As a result, the prototype could achieve a reduction of 9.70% with respect to energy consumption as compared with the maximum energy-using machine tool.

show abstract

“…This method can not only predict the performance of the machine tool, but also be more convenient for the performance analysis of the machine tool with the tool spindle combination changed. Chen et al [ 23 ] and Liu et al [ 24 ] combined substructure coupling with deep learning and used the transfer learning method to predict tool tip dynamic information at different positions. The existing modal [ 25 ] analysis methods are limited to the static conditions of machine tools, and most of them are applied to the transition of machining state, and few of them are used to analyze the influence of cutting force.…”

Section: Introductionmentioning

confidence: 99%

Cutting Force Transition Model Considering the Influence of Tool System by Using Standard Test Table

Chen

Zhang

Wang

2021

Sensors

View full text Add to dashboard Cite

The cutting force prediction model usually uses the classical oblique transformation method, which introduces the orthogonal cutting parameters into the oblique milling edge shape, and combines the geometric parameters of the tool to convert the orthogonal cutting force into the actual cutting force, thereby predicting the cutting force. However, this cutting force prediction method ignores the impact of tool vibration in actual machining, resulting in a large difference between the prediction model and the actual measurement. This paper proposes a cutting force conversion model considering the influence of the tool system. The proposed model fully considers the impact of tool vibration on the cutting force. On the basis of the orthogonal model, superimposing the additional cutting force generated by tool vibration makes the predicted value of the model closer to the actual cutting force. The results of milling experiments show that the conversion model can obtain higher prediction accuracy. Moreover, compared with the original conversion model, the accuracy of the proposed model is significantly improved.

show abstract

Pose-dependent tool tip dynamics prediction using transfer learning

Cited by 64 publications

References 24 publications

Suppression of robot vibrations using input shaping and learning-based structural models

Suppression of robot vibrations using input shaping and learning-based structural models

A Holonic-Based Self-Learning Mechanism for Energy-Predictive Planning in Machining Processes

Cutting Force Transition Model Considering the Influence of Tool System by Using Standard Test Table

Contact Info

Product

Resources

About