Resonance-suppression Control for Electro-hydrostatic Actuator as Two-inertia System

Tsuda, Kazuki; Sakuma, Tomoki; Umeda, Kodai; Sakaino, Sho; Tsuji, Toshiaki

doi:10.1541/ieejjia.6.320

Cited by 16 publications

(4 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the contrary, by using autoregressive models such as SM2SM and S2SM, we can obtain both true and estimated values of the slave states, and thus can calculate the estimation errors. Notably, in bilateral control-based imitation learning, the master and slave estimates are strongly related because they are naturally controlled to satisfy the following control goals (1) and (2). Therefore, in this study, we assumed that the estimation error of the slave is equal to the estimation error of the master, as follows:…”

Section: ) Command Feedback In Autoregressive Models (The Proposed Me...mentioning

confidence: 99%

“…To realize robot automation, they are required to perform tasks in diverse environment and while treating various objects. Moreover, motion control is difficult but its demand is rising [1], [2]. Therefore, machine learning is being used to study motion planning of robots to generate various motions and facilitate easier programming for motion planning.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A New Autoregressive Neural Network Model with Command Compensation for Imitation Learning Based on Bilateral Control

Hayashi,

Sasagawa,

Sakaino

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

In the near future, robots are expected to work with humans or operate alone and may replace human workers in various fields such as homes and factories. In a previous study, we proposed bilateral control-based imitation learning that enables robots to utilize force information and operate almost simultaneously with an expert's demonstration. In addition, we recently proposed an autoregressive neural network model (SM2SM) for bilateral control-based imitation learning to obtain long-term inferences. In the SM2SM model, both master and slave states must be input, but the master states are obtained from the previous outputs of the SM2SM model, resulting in destabilized estimation under large environmental variations. Hence, a new autoregressive neural network model (S2SM) is proposed in this study. This model requires only the slave state as input and its outputs are the next slave and master states, thereby improving the task success rates. In addition, a new feedback controller that utilizes the error between the responses and estimates of the slave is proposed, which shows better reproducibility.

show abstract

Section: ) Command Feedback In Autoregressive Models (The Proposed Me...mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A New Autoregressive Neural Network Model with Command Compensation for Imitation Learning Based on Bilateral Control

Hayashi,

Sasagawa,

Sakaino

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Typical examples are timingbelt systems [1], series elastic actuators [2], [3], hydraulic actuators [4], and many of robot joints with gears [5]. Therefore, improving the control performance of two-inertia systems has been the subject of many researches [6] [7] [8].…”

Section: Introductionmentioning

confidence: 99%

Design of Resonance Ratio Control with Relative Position Information for Two-inertia System

Araake

Sakaino

Tsuji

2019

2019 IEEE/ASME International Conference on Advanced Intelligent Mechatronics (AIM)

Self Cite

View full text Add to dashboard Cite

Two-inertia systems are prone to resonance vibrations that degrade their control performances. These unwanted vibrations can be effectively suppressed by control methods based on a disturbance observer (DOB). Vibration suppression control methods using the information of both the motor and load sides have been widely researched in recent years. Methods that exploit the spring deflection or torsional force of two-inertia systems have delivered promising performances. However, few conventional methods have exploited the relative position information, and the discussion of position control is currently insufficient. Focusing on the relative position, this study proposes a new resonance ratio control (RRC) based on the relative acceleration and state feedback. The structure of the proposed RRC is derived theoretically and the proposed method is experimentally validated.

show abstract

“…Bobrow et al used an EHA to develop a multi-axis robot (7) . In recent years, robots utilizing the superior backdrivability of EHAs have been actively researched (8)- (10) . The hydraulic circuit of an EHA is shown in Fig.…”

Section: Introductionmentioning

confidence: 99%

Reaction Force Estimation of Electro-hydrostatic Actuator Using Reaction Force Observer

et al. 2018

Self Cite

View full text Add to dashboard Cite

Electro-hydrostatic actuators (EHAs) are hydraulic actuators that are flexible and exhibit high backdrivability. Flexible operation and accurate detection of reaction forces are required for robots to be able to perform in environments in which they will be cohabiting with humans. However, nonlinear elements that degrade detection accuracy, such as static friction, backlash, and oil leakage, are present in hydraulic systems. In addition, pressure sensors in hydraulic systems are not very accurate at estimating reaction forces, because they cannot estimate internal forces and viscous friction. In this study, we propose a combination of control algorithms for accurately estimating reaction forces. Static friction is compensated by using feedback modulators. In addition, we use a backlash and oil leakage compensator, which do not require any models, to suppress the relative velocity between the motor-side and load-side. Then, the use of a reaction force observer (RFOB) that exploits both pressure sensors and encoders is proposed. The RFOB can be implemented because disturbances are linearized by the compensators. Experimental results show that reaction forces can be estimated with very high accuracy using the proposed RFOB. In addition, we implemented force control using the RFOB and evaluate the force tracking performances by improving the estimation accuracy.

show abstract

Resonance-suppression Control for Electro-hydrostatic Actuator as Two-inertia System

Cited by 16 publications

References 26 publications

A New Autoregressive Neural Network Model with Command Compensation for Imitation Learning Based on Bilateral Control

A New Autoregressive Neural Network Model with Command Compensation for Imitation Learning Based on Bilateral Control

Design of Resonance Ratio Control with Relative Position Information for Two-inertia System

Reaction Force Estimation of Electro-hydrostatic Actuator Using Reaction Force Observer

Contact Info

Product

Resources

About