Intelligent Scheduling of Robotic Flexible Assembly Cells

Multi-objective optimisation of dynamic scheduling in robotic flexible assembly cells via fuzzy-based Taguchi approach

Flexible Assembly Systems - Hardware Components and Features: Basic components of a FAS.- Classification of flexible assembly systems.- Examples of industrial installations.- Issues in Design, Planning and Scheduling of Flexible Assembly: FAS design issues.- Network design for material flow systems.- Design for assembly.- Assembly planning.- Planning and scheduling.- Loading and Routing Decisions in Flexible Assembly Systems: Description of a flexible assembly system.- Optimization of station workloads and product movements.- Design and balancing of flexible assembly lines.- Numerical examples.- Simultaneous loading and routing.- Sequential loading and routing.- Loading and Routing Decisions with Assembly Plan Selection: Simultaneous loading, routing and assembly plan selection in a general flexible assembly system.- Simultaneous loading, routing and assembly plan selection in a flexible assembly line.- Loading and Scheduling in Flexible Assembly Cells: Loading and scheduling in a robot assembly cell.- Loading and sequencing in printed circuit board assembly.- Production Scheduling in Flexible Assembly Lines: Flexible assembly line with limited intermediate buffers.- Flexible assembly line with no in-process buffers.- Global lower bounds for flexible flow lines with unlimited buffers.- Just-in-time scheduling of flexible assembly lines.- Multilevel scheduling of flexible assembly lines with limited intermediate buffers.- Machine and Vehicle Scheduling in Flexible Assembly Systems: Dispatching scheduling.- Machine and vehicle scheduling - a multi-level approach.- Machine and vehicle scheduling - a single-level approach.

This study describes the application of the develop ed model to the scheduling of the Robotic Flexible Assembly Cells (RFACs). The present study is a continuation of the paper titled “Development of a Fuzz ySimulation Model of Scheduling Robotic Flexible Assembly Cells”. This new model for multi-objective scheduling problems in RFACs was based on combining a fuzzy-based mathematical model with simulation software tools. In this study, a hypothetical case study of RFACs is presented to demonstrate the effectiveness of the developed model and then compare the results that are obtained by the new methodology with the common scheduling rules. The simulation results show that the performance of the proposed methodology outperforms the most popular scheduling rules from previous research.

This study describes the application of the developed approach to the scheduling of the Robotic Flexible Assembly Cells (RFACs). The present study is a continuation of the paper titled "development of fuzzy AHP and TOPSIS approach for multi-objective scheduling problems in robotic flexible assembly cells". In this paper, a realistic case study of RFACs is presented to examined and verified the developed approach via implement three phases: problem description; application of FAHP-FTOPSIS; and analysis of the results. The analysis showed the stability of the results for FAHP-FTOPSIS and also verified the robustness and the appropriateness of the proposed approach. This study also demonstrated that the proposed approach can be successfully used in the multi-objective optimisation of dynamic scheduling in RFAC.

Current developments in the field of automated assembly systems show an increasing interest in systems that are flexible in both CAD based product design and CAD based assembly. For the application addressed in this paper, coupling the vision system and a CAD database is of prime importance in order to achieve the required automatic reconfiguration of the assembly cell when new parts are defined. This paper presents a 3D CAD-based vision system for obtaining 3D data about the scene. After the images are acquired, edge detection is preformed and the detected edges are stored as chaincodes. Following that, a stereo vision algorithm is applied for finding the recognition features. The output are lists of features that are combined into a 3D wireframe representing the scene. The recognition algorithm takes the observed wireframe outputs from the stereo vision system, and compares them with a set of model wireframes derived from previous models, in order to select the 'best match', where the previous models used for recognition are derived from a product data model (PDM). The PDM is an interface between the CAD database and the recognition system, which allows the automatic generation of new models when new parts are introduced into the system. The vision system described in this paper is part of an intelligent robotic assembly cell, where the aim is to build a flexible intelligent robotic assembly cell, such that robots would be able to automatically assemble a random variety of small- batch products.

Companies are increasingly being forced by market demands to increase the competitiveness of their products in regard to price, quality and delivery time. Development, both of product and production, is the key word in achieving this goal. This paper focuses on the following related subject: cost-effective design, with applications to robotic assembly cells. Against this background, a total productivity model is introduced, which allows an opinion to be formed about the performance of robotic assembly cells during the conceptual stage of the design process.

To date, holonic manufacturing system (HMS), one of the novel concepts to meet the increasing globalization and customization of the manufacturing processes, has been conceptually constructed, and prototyping systems, holonic fixturing station (HOLOFIX), holonic autonomous guided vehicles (H-AGVs) etc. were developed with enhanced performances in their specific processes. This work presented an object oriented technology (OOT)-supported migration approach to holonic robot assembly cell. Based on the analysis of activities in the flexible assembly cell, an architecture of workstation for holonic robot assembly cell was proposed. The holonic robot workstation was established by incorporating a control computer that performs production planning, shop floor control and local operations into each of robot systems in the flexible assembly cell, workstations of this kind functions as the building blocks of holonic robot assembly cell. The holonic robot assembly cell employs a dynamic holarchy evolution mechanism to carry out assembly tasks. The entire implementation was conducted on the flexible assembly system in PMA/Leuven.

Visual feedback has traditionally been used in the assembly process to a very limited extent. By incorporating high bandwidth visual feedback into the manipulator feedback loop, more flexible and adaptable robotic assembly systems can be developed. The authors present experimentally tested algorithms and propose solutions for the problem of dynamic sensor placement in assembly tasks.

During the early 1980s, the concept of a flexible assembly machine was first suggested by Hounsfield (1983). Following this, a variety of research projects have been conducted in an attempt to develop a flexible assembly system that is functional and economically viable. The most significant contribution has been made by the EURICA EU 321 - FAMOS - INFACT* project. However, despite the developments made during these projects, no industrial based system exists today. This paper presents a simulation model of a novel concept for a multi-station flexible automatic assembly machine, and examines the application of a rule based control strategy for the control of a materials handling system used in such a system. * http://www.3.eureka.be/home/

The concept of robotic flexible assembly cell (RFAC) has the potential to introduce significant improvements in system performance, as highlighted in Chap. 1. To achieve these improvements, the problems of RFAC at scheduling level need to be critically addressed. To date, researchers have paid little attention to handling the scheduling problems relevant to RFAC.

Dynamics and control for in-space assembly robots with large translational and rotational maneuvers

Most of industrial robotic assembly tasks today require fixed initial conditions for successful assembly. These constraints induce high production costs and low adaptability to new tasks. In this work we aim towards flexible and adaptable robotic assembly by using 3D CAD models for all parts to be assembled. We focus on a generic assembly task - the Siemens Innovation Challenge - in which a robot needs to assemble a gear-like mechanism with high precision into an operating system. To obtain the millimeter-accuracy required for this task and industrial settings alike, we use a depth camera mounted near the robot’s end-effector. We present a high-accuracy two-stage pose estimation procedure based on deep convolutional neural networks, which includes detection, pose estimation, refinement, and handling of near- and full symmetries of parts. The networks are trained on simulated depth images with means to ensure successful transfer to the real robot. We obtain an average pose estimation error of 2.16 millimeters and 0.64 degree leading to 91% success rate for robotic assembly of randomly distributed parts. To the best of our knowledge, this is the first time that the Siemens Innovation Challenge is fully addressed, with all the parts assembled with high success rates.

As manufacturing trends shift towards customized production, the demand for agile automation systems capable of efficient adaptation to rapidly changing task requirements is rising. This work presents a flexible robotic assembly system that combines CAD based component localization, compliance control, and a multi-modal gripper to enable robust and efficient programming of complex tasks. The process can be easily configured for novel assemblies through a dedicated Graphical User Interface (GUI), which facilitates component identification and task sequencing. Component poses are extracted from 3D CAD models in reference to the assembly origin, while the active compliance scheme compensates for minor positioning errors. The gripper incorporates a parallel jaw element, a rotating module, and an electromagnet to minimize retooling delays. The first iteration of the proposed system placed first in the manufacturing track of the Robotic Grasping and Manipulation Competition of the 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), experimentally validating its efficiency.

Recently a major trend in flexible and precision assembly automation, and in automation in general, has been in robots that can safely work next to people. Increasingly, as part of collaborative automation, such robots are also designed to collaborate as teams with other robots. These intelligent machines are called cobots, or collaborative robots, and are reducing the cost and time to install robots in many applications. Flexible and precision assembly refers to an assembly system that can build multiple similar products with little or no reconfiguration of the assembly system. It can serve as a model for some of the emerging applications in flexible and precision automation. Precision assembly implies assembly with precise motions and operations. Recently, machine learning interacting with the robots’ sensors enables robots to assemble parts and components with precision that exceeds their own motion precision. A truly flexible assembly system would enable precision assembly, include flexible part feeding, grasping, and fixturing, as well as a variety of mating and fastening processes that can be quickly added or deleted without costly engineering. There is a limited science base for how to design flexible assembly systems in a manner that will yield predictable and reliable throughputs. The emergence of geometric modeling systems (as part of CAD) has enabled work in geometric reasoning in the past few years. Geometric models have been applied in areas such as machine vision for object recognition, design, and throughput analysis of flexible part feeders, dynamic simulation of assembly stations and assembly lines. Still lacking are useful techniques for automatic model generation, planning, error representation, error prevention (avoidance), and error recovery. Future software architectures for flexible and precision automation should include geometric modeling and reasoning capabilities to support autonomous, sensor-driven systems.

A simulation model for use in designing robotic assembly cells