scholarly journals Open Loop Position Control of Soft Continuum Arm Using Deep Reinforcement Learning

Author(s):  
Sreeshankar Satheeshbabu ◽  
Naveen Kumar Uppalapati ◽  
Girish Chowdhary ◽  
Girish Krishnan
2018 ◽  
Author(s):  
Sreeshankar Satheeshbabu ◽  
Naveen Kumar Uppalapati ◽  
Girish Chowdhary ◽  
Girish Krishnan

Soft robots undergo large nonlinear spatial deformations due to both inherent actuation and external loading. The physics underlying these deformations is complex, and often requires intricate analytical and numerical models. The complexity of these models may render traditional model based control difficult and unsuitable. Model-free methods offer an alternative for analyzing the behavior of such complex systems without the need for elaborate modeling techniques.In this paper, we present a model-free approach for open loop position control of a soft spatial continuum arm, based on deep reinforcement learning. The continuum arm is pneumatically actuated and attains a spatial workspace by a combination ofunidirectional bending and bidirectional torsional deformation. We use Deep-Q Learning with experience replay to train the system in simulation. The efficacy and robustness of the control policy obtained from the system is validated both in simulation and on the continuum arm prototype for varying external loading conditions


Author(s):  
Yogesh V. Hote ◽  
A. N. Jha ◽  
J. R. P. Gupta

In this paper, simple approach is proposed to determine reduced order model of a unstable open-loop position control system. This approach is based on Krishnamurthy’s approach on Routh criterion on reduced order modelling. The results are simulated in Matlab environment.


2017 ◽  
Vol 9 (12) ◽  
pp. 168781401773719 ◽  
Author(s):  
Jinlong Piao ◽  
XueJun Jin ◽  
Jinwoo Jung ◽  
Eunpyo Choi ◽  
Jong-Oh Park ◽  
...  

1981 ◽  
Vol 25 (1) ◽  
pp. 571-575
Author(s):  
R. A. Miller ◽  
R. J. Jagacinski ◽  
R. B. Nalavade ◽  
W. W. Johnson

Subjects manipulated a position control stick with one hand and a velocity control stick with the other hand in order to capture a moving target displayed on an oscilloscope screen. The two control sticks were additively coupled. In order to understand the coordination of the two control sticks, event-based first-order markov “activity sequence generators” were constructed for individual subjects. These discrete probabilistic structures are closely related to each subject's overall plan or general strategy for the capture task. Striking individual differences and strategic errors in performance were revealed. When combined with additional time-conditioned (open-loop) and error-conditioned (closed-loop) details, the activity sequence generators provide a basis for a hierarchic description of this perceptual-motor skill.


Author(s):  
J. M. Pottinger ◽  
N. D. Ring

The application of position control to externally powered prosthetic arms leads to increased function and versatility when compared to a prosthetic system with open-loop control. However, various compromises are necessary owing to the limited number of available control sites. An investigation is being conducted into the possibility of combining the advantages of position control during the dynamic phase of movement with the locking facility of velocity–time control during the static phase, which leads to a larger number of available control sites and thus a greater number of degrees of freedom.


2014 ◽  
Vol 602-605 ◽  
pp. 1157-1160
Author(s):  
Mei Yu ◽  
Guo Wei Liu ◽  
Bing Kong

In view of the present mining spraying manipulator operation is not flexible ,spraying effect is poor, susceptible to interference and other issues, this paper studies and realizes a kind of low cost, strong practicability of spraying manipulator control system. Using S7-200PLC and 2MA860H drive to control the 86BYG250A stepper motor open-loop control system, and the precise position control is realized. By controlling of the x-y axis mine spraying manipulator. Validate the system operation is simple, highly efficient and stable, energy conservation and environmental protection, strong anti-jamming capability, it can be widely used in all kinds of mining equipment.


Author(s):  
Won Young Kim ◽  
Kishore Pochiraju

As 3D printing technology becomes more ubiquitous and pervasive, printed robot structures are likely to become popular. However, these 3D printed structures have inherent flexibility which causes link deflections and vibrations that lead to difficulties in maintaining gait and produce instabilities during motion. In this paper a biped robot, realized with 3D printing techniques, was modeled and its gait was simulated under position control. Spring and damper elements are used within powered joints to model the link flexibility. Servos were placed at the joints and an open-loop gait trajectory was executed by posing the robot through a series of servo angles. The gait of the printed robot was designed using the model. The printed robot’s stability and accelerations during the motion were characterized with 3-axis accelerometers and gyroscopes mounted on the robot. The acceleration measurements from the printed robot are then compared with the model behavior.


Sign in / Sign up

Export Citation Format

Share Document