scholarly journals Quick Setup of Force-Controlled Industrial Gluing Tasks Using Learning From Demonstration

2021 ◽  
Vol 8 ◽  
Author(s):  
Iñigo Iturrate ◽  
Aljaz Kramberger ◽  
Christoffer Sloth
Keyword(s):  
Position Control ◽  
Adaptive Controller ◽  
Contact Forces ◽  
Surface Geometry ◽  
Task Constraints ◽  
Impact Forces ◽  
Series Of Experiments ◽  
Contact Tasks ◽  
Collaborative Robot

This paper presents a framework for programming in-contact tasks using learning by demonstration. The framework is demonstrated on an industrial gluing task, showing that a high quality robot behavior can be programmed using a single demonstration. A unified controller structure is proposed for the demonstration and execution of in-contact tasks that eases the transition from admittance controller for demonstration to parallel force/position control for the execution. The proposed controller is adapted according to the geometry of the task constraints, which is estimated online during the demonstration. In addition, the controller gains are adapted to the human behavior during demonstration to improve the quality of the demonstration. The considered gluing task requires the robot to alternate between free motion and in-contact motion; hence, an approach for minimizing contact forces during the switching between the two situations is presented. We evaluate our proposed system in a series of experiments, where we show that we are able to estimate the geometry of a curved surface, that our adaptive controller for demonstration allows users to achieve higher accuracy in a shorter demonstration duration when compared to an off-the-shelf controller for teaching implemented on a collaborative robot, and that our execution controller is able to reduce impact forces and apply a constant process force while adapting to the surface geometry.

scholarly journals Small nanoparticles, surface geometry and contact forces

2018 ◽  
Vol 474 (2211) ◽  
pp. 20170723 ◽  
Author(s):  
Yoichi Takato ◽  
Michael E. Benson ◽  
Surajit Sen
Keyword(s):  
Contact Force ◽  
Contact Forces ◽  
Spherical Nanoparticles ◽  
Hertz Contact ◽  
Surface Geometry ◽  
Impact Forces ◽  
Nonlinear Contact ◽  
Amorphous Surface ◽  
The Impact ◽  
Small Nanoparticles

In this molecular dynamics study, we examine the local surface geometric effects of the normal impact force between two approximately spherical nanoparticles that collide in a vacuum. Three types of surface geometries—(i) crystal facets, (ii) sharp edges, and (iii) amorphous surfaces of small nanoparticles with radii R <10 nm—are considered. The impact forces are compared with their macroscopic counterparts described by nonlinear contact forces based on Hertz contact mechanics. In our simulations, edge and amorphous surface contacts with weak surface energy reveal that the average impact forces are in excellent agreement with the Hertz contact force. On the other hand, facet collisions show a linearly increasing force with increasing compression. Our results suggest that the nearly spherical nanoparticles are likely to enable some nonlinear dynamic phenomena, such as breathers and solitary waves observed in granular materials, both originating from the nonlinear contact force.

The Robotic Force Tracking Based on Single Input Fuzzy Adaptive Controller

2011 ◽  
Vol 328-330 ◽  
pp. 2117-2120
Author(s):  
Er Chao Li ◽  
Zhan Ming Li
Keyword(s):  
Position Control ◽  
Fuzzy Controller ◽  
Sliding Mode ◽  
Adaptive Controller ◽  
Mode Switching ◽  
Unknown Environment ◽  
Fuzzy Input ◽  
Force Tracking ◽  
The Difference ◽  
Contact Tasks

In order to realize precise contact tasks with an unknown environment, robotic force controllers have to adapt themselves to the unknown environment. In this paper, an effective controller for robot contact tasks is proposed using sliding mode switching surface as a sole fuzzy input variable for substituting the feedback controller in the model reference adaptive control (MRAC), the fuzzy force controller determines the adjustment to the position control loop, adjust the fuzzy controller parameters according to the difference between the actual and desired force responses. There is a sole fuzzy input variable in fuzzy logical controller(FLC), the total number of rules is greatly reduced compared to conventional FLC, hence generations and tuning of controller are more simple and easy. Simulations prove that the excellent adaptability as well as high precision is obtained by the proposed strategy.

Virtual Position Control During Contact Tasks

1999 ◽  
Author(s):  
Chad E. English ◽  
Donald L. Russell
Keyword(s):  
Mechanical Properties ◽  
Equilibrium Point ◽  
Position Control ◽  
Contact Forces ◽  
Control Technique ◽  
Prosthetic Limb ◽  
Virtual Paths ◽  
Reflex Action ◽  
Virtual Point ◽  
Contact Tasks

Abstract It is well known that human limbs have spring-like behaviour that has been attributed to both compliance of the muscles and reflex action [Hogan, 1985; Mussa-Ivaldi et al., 1985]. These mechanics are the premise behind the Equilibrium Point (EP) Hypothesis for motor control [Bizzi et al., 1992; Shadmehr et al., 1993; Won & Hogan, 1995]. Essentially, EP theory suggests that muscles are activated such that they define a unique position of the limb in the workspace, called the virtual point, where there are no net joint moments [Bizzi et al., 1992]. (Some refer to this as the equilibrium point, but that notation can be confused with a contact equilibrium.) If an amputee using a prosthetic limb uses the EP control technique naturally, it is necessary for the prosthesis to included mechanical properties sufficiently similar to those of a natural limb to produces similar results. The necessary mechanics, and limitations of not providing them, are the bulk of the authors’ current research. The current work examines the control of contact forces. A variety of studies have examined virtual paths for unconstrained motions [Flash, 1987; Lan & Crago, 1994]. Others have implemented a predetermined virtual point path in an artificial manipulator during contact tasks [Hogan, 1987]. However, little information is available on the virtual point path required to produce desired contact properties. This study analytically and numerical investigates the virtual point path required to vary contact force without producing a motion of the limb along the surface.

The fate of formaldehyde in sugar manufacture and in products

2015 ◽  
pp. 692-696
Author(s):  
Remi Aubry ◽  
Laurence Gasnot
Keyword(s):  
Sugar Industry ◽  
Common Knowledge ◽  
Microbial Populations ◽  
Optimal Dosage ◽  
Consumer Health ◽  
Microbiological Activity ◽  
Specific Process ◽  
Beet Sugar ◽  
Series Of Experiments

A study was carried out in six beet sugar factories in France during the 2012/13 sugar campaign. The objective was to assess the optimal dosage of formaldehyde solutions at specific process stages and in different existing factory set-ups in order to obtain the desired effect on microbial populations, without interference with the quality of the products. In addition harmlessness regarding consumer health was to be demonstrated. A series of experiments was conducted resulting in new data allowing refreshment of common knowledge and references existing regarding the use of formaldehyde solutions in the sugar industry. The effectiveness and convenience for controlling microbiological activity in beet sugar manufacture was assessed. Formaldehyde reduces sugar losses and protects in-process products without harming their further use, such as for ethanol production.

THE RULE OF COMBINING A PRIORI AND EXPERIMENTAL INFORMATION IN THE TASKS OF MANAGING TESTS OF COMPLEX TECHNICAL PRODUCTS ACCORDING TO THE BERNOULLI SCHEME

2020 ◽  
pp. 3-13
Author(s):  
Yu. I. Buryak ◽  
A. A. Skrynnikov
Keyword(s):  
Field Experiments ◽  
A Priori ◽  
Field Tests ◽  
Experimental Information ◽  
Full Scale ◽  
Testing Complex ◽  
Series Of Experiments ◽  
Full Scale Tests ◽  
Technical Products

The article is devoted to the substantiation of the procedure for testing complex technical systems to assess the probability of performing the task, taking into account a priori data obtained from the results of modeling, field tests of components and prototypes, operation of analogues, etc. The conditions for the formation of a combined sample consisting of field experiments and experiments counted on the results of modeling are justified. Data uniformity is checked using the Student's criterion. The minimum volume of full-scale tests is determined by the requirement of equality of the amount of Fischer information about the estimated parameter obtained during full-scale tests and at the expense of a priori data A strategy for conducting field experiments is proposed, in which the required quality of evaluating the probability of completing the task is achieved with the minimum possible number of field experiments. At the first stage, a series of experiments with a volume equal to half of the required sample size is performed. At the second stage, the experiments are conducted sequentially with an assessment after each experiment of the requirements for the amount of information about the evaluated parameter and for the uniformity of data. Experiments are terminated when the specified requirements are met, and then a combined sample is formed, which is used to evaluate the probability of the system performing the task. A model example is considered. The estimation of the gain in the number of experiments performed at different probability values was carried out.

An Approach towards Survey and Analysis of Cloud Robotics

2017 ◽  
pp. 208-231 ◽  
Author(s):  
Akash Chowdhury ◽  
Swastik Mukherjee ◽  
Sourav Banerjee
Keyword(s):  
Robot Learning ◽  
Robotic System ◽  
System Quality ◽  
Storage Unit ◽  
Cloud Robotics ◽  
Processing Power ◽  
Cloud Datacenter ◽  
Total Structure ◽  
Collaborative Robot

This chapter highlights the total structure and capabilities of robotic systems. This chapter then discusses the invocation of cloud technology in robotics technology empowering the whole system with higher processing power and bigger storage unit which was not possible earlier in the conventional robotic system being restricted in on-board manipulation. The flexibility of handling big data, ability to perform cloud computing, crowed sourcing and collaborative robot learning using the cloud robotics technology has been discussed briefly. This chapter describes concepts of Cloud Enabled Standalone Robotic System (CeSRS), Cloud Enabled Networked Robotic System (CeNRS), Cloud Robotic Networking System (CRNS), Standalone Robotic System (SRS), Common Networked Robotic (CNRS), Infrastructure As A Service (IAAS), Multi Robot System, R/R and R/C Network, ROS, Tele Operated Robotic System, Quality of Service (QoS), Virtual Machine (VM) and Cloud Datacenter. The existing applications of the cloud robotics technology are also described. However, the chapter focuses on the problems either inherited from the parent technology or appeared in the child technology. This chapter further recommends some solutions, new future directions and research aspects of the cloud robotics technology depending on the applications.

Sign in / Sign up

Export Citation Format

Share Document