Cognitive Mechatronic Devices for Reconfigurable Production

applied sciences

Article Cognitive Mechatronic Devices for Reconﬁgurable Production of Complex Parts

Panagiotis Karagiannis, George Michalos , Dionisis Andronas, Aleksandros-Stereos Matthaiakis, Christos Giannoulis and Sotiris Makris *

Laboratory for Manufacturing Systems and Automation, Department of Mechanical Engineering and Aeronautics, University of Patras, 26504 Patras, Greece; [email protected] (P.K.); [email protected] (G.M.); [email protected] (D.A.); [email protected] (A.-S.M.); [email protected] (C.G.) * Correspondence: [email protected]; Tel.: +30-2610-91-01-60

Featured Application: Robotic cell composed of cognitive mechatronic devices for complex part processing in the industrial lines.

Abstract: This paper discusses a robotic cell that handles geometrically complex products, exploiting cognitive control and actuation systems for the manipulation, assembly and packaging. The individual mechatronic components, namely a 6-DoF gripper and a ﬂexible assembly mechanism, have been designed via functional decomposition of the actual assembly and handling tasks. The ﬂexibility of

these mechanisms is exploited through control modules, performing different cognition functions at cell, resource and device level. The design approach can be generalized for tasks requiring dexterity

Citation: Karagiannis, P.; Michalos, and adaptation to products. A case study from the consumer goods sector, showcases the system’s G.; Andronas, D.; Matthaiakis, A.-S.; reconfigurability and efficiency. Giannoulis, C.; Makris, S. Cognitive Mechatronic Devices for Keywords: cognitive robotics; distributed control; mechatronic Reconfigurable Production of Complex Parts. Appl. Sci. 2021, 11, 5034. https://doi.org/10.3390/ app11115034 1. Introduction Flexibility and reconfigurability have been investigated at multiple production levels, Academic Editor: including production tools and resources using different technologies [1]. Particularly in Alessandro Gasparetto the robotics-automation fields [2] dexterous tooling and mechatronic systems [3], adaptable control strategies [4], reconfigurable and mobile robots [5] are key components used Received: 23 April 2021 to achieve flexibility and reconfigurability in production lines [6,7]. Although a lot of Accepted: 27 May 2021 progress has been made on the development of each component, the research community Published: 29 May 2021 has not focused its effort so much on their efficient integration and combination due to the high complexity of the endeavor. Thus, despite the high interest from the industry, Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in the researchers focused mainly on the development, testing and optimization of each published maps and institutional affil- individual technology, as described in more detail for each technology in the following iations. paragraphs. The flexibility of producing different product variants is manifested in multiple ways, starting with the physical/hardware reconfiguration of devices to accommodate variable product geometry. For instance, dexterous grippers with multiple fingers and degrees of freedom (DOFs) [8], modular finger structures [9], innovative gripping principles [10] or Copyright: © 2021 by the authors. actuation methods [11] and model-based control schemes [12] have been researched and Licensee MDPI, Basel, Switzerland. This article is an open access article developed. Such devices are partially able to reconfigure their structure or use modular distributed under the terms and contact points (interfaces) to handle different parts, but the key restriction is the difficulty conditions of the Creative Commons to achieve flexibility in a modular way at the operational and control levels at the same Attribution (CC BY) license (https:// time. These devices can be either: (a) part specific, with the disadvantage to have very creativecommons.org/licenses/by/ limited flexibility or (b) operation specific, able to handle a range of parts for a limited 4.0/). number of operations (e.g., simply for pick & place activities) or (c) control independent,

Appl. Sci. 2021, 11, 5034. https://doi.org/10.3390/app11115034 https://www.mdpi.com/journal/applsci Appl. Sci. 2021, 11, 5034 2 of 14

making them unable to be integrated in complex control architectures that support sensor fusion and cognitive functions. Another important element that is required to achieve flexibility is the adaptable sensing technology able to perceive the environment and help the system to adjust the operational parameters of the machine to specific cases. There are sensing technologies that focus on the process perception executed by the resources [13,14], while there are other algorithms able to detect human operators and their activities [15]. Apart from the sensing strategy, an important role plays the smart algorithm that processes the data and is able to apply different task execution strategies according to the vision results [16]. Typically, the above types of perception run independently and individually in a dedicated embedded PC, without having a holistic perception approach that can match the human capabilities. Referring to human capabilities, apart from perception capabilities, researchers are trying to replicate the human cognition capabilities. Thus, researchers’ vision is to create cognition enabled resources with high-level decision-making possibilities, adding another key element to implement flexible manufacturing lines. A lot of research activity has been focused on this technology in all production levels starting from the resource level up to the line level [17], in swarm robotics [18] even in social robotics [19]. Additionally, similar research activities have been focused in more innovative directions such as in human-robot collaborative cells [20], in logistics operations [21] and in cyber-physical systems [22], as well as in more traditional production lines [23] such as the automotive industry assembly lines [24]. This multi-level cognition approach may start from the high-level, namely task and motion planning to optimize the assembly operations, down to the differentiation of the local control strategy at tooling level. Additionally, the low-level control strategy takes into consideration perception data that provide information on the status of the resources and their surroundings. The key issue is that each of the suggested research focuses on a specific part of the cognition scale and does not offer a unique solution in all levels, adaptable to industrial needs. Thus, they are not modular and open control systems able to represent the reconfiguration potential required by the new production lines. Last but not least, inherently flexible and reconfigurable resources, namely robots, have been introduced in multiple forms within production lines. Especially mobile robots seem to be quite promising since there are several types, such as industrial mobile manipulators [25] and mobile platforms with few DoF for logistics applications [26]. Simulation tools and control models for such cases are already available and can be easily integrated in production applications [27,28]. Further types of robots such as collaborative manipulators have captured the interest of researchers over the last years [29–31] and provide further opportunities for integrating humans in the loop. Their potential of robots relies on the fact that when equipped with tool-changers, they can undertake different roles such as assembly, logistics, inspection & maintenance, quality control and others making them extremely reconfigurable when compared to stationary machinery. The key issue that is faced by such resources is the performance, due to having extra times for their transition to different stations as well as the computational time to implement sensing and safe manipulation in order not to affect the product quality. In this paper decentralize cognitive functions that are implemented on the station/resource/tool/device levels are used to overcome this limitation. The research community has addressed the above fields mainly by focusing on the development, testing and optimization of each technology individually. The efficient combination of these technologies under a common system leads to a complex result that is very difficult to deploy and maintain in an industrial application. However, the industrial world expresses high interest on this integration and dictates that these technologies should be combined to cover their requirements and form reconfigurable robotic systems (Figure1) . As it is shown in Figure1 different components should be integrated together sharing sensor information from the lower levels to the higher ones, while the top levels should provide instructions and other control-related commands based on sensor data input. At the same time, each component follows local sensing and control loops to update Appl. Sci. 2021, 11, 5034 3 of 14

Appl. Sci. 2021, 11, 5034 integrated together sharing sensor information from the lower levels to the higher 3ones, of 14 while the top levels should provide instructions and other control-related commands based on sensor data input. At the same time, each component follows local sensing and control loops to update its status and make low-level decisions on the fly. Thus, the future aimits status is to achieve and make better low-level production decisions performance on the ﬂy. as Thus, well theas to future achieve aim optimal is to achieve production better productioncontrol once performance the new technologies as well as to are achieve integrated optimal into production the respective control hardware once the new and softwaretechnologies. are integrated into the respective hardware and software.

Figure 1. Key components that compose a reconﬁgurablereconfigurable production system.system.

In thisthis paper,paper, thethe implementationimplementation of of a a robotic robotic assembly assembly system system is is described, described similar, similar to tothe the one one illustrated illustrated in the in ﬁgurethe figure above, above focusing, focusing on the on integration the integration of mechatronic of mechatronic devices whichdevices possess which theirpossess own their cognition own cognition capabilities capabilities on a local on level a local using level their using local their sensors. local Additionally,sensors. Additionally, a high-level a controlhigh-level framework control(station framework cognition) (station is included, cognition) which is included monitors, whichthe operations monitors to the be performed operations by to the be local performed control by systems the local of the control individual systems resources. of the Theindividual innovative resources. part of The this innovative study is the part integration of this study of different is the components,integration of including different components,both the hardware including (robot, both tools the andhardware mechatronic (robot, devices),tools and andmechatronic the software devices (control), and and the softwaresensing algorithms) (control and under sensing a common algorithms architecture) under that a can co implementmmon architecture cognitive functions that can implementat different cognitive levels. Data functions from sensors at different integrated levels into. Data the from assembly sensors station integrated are used into both the assemblylocally by station the resources are used to both execute locally the by processes the resources assigned to execute to them the but processes also escalated assigned to tohigher them control but also levels escalated so that to they higher can control be used levels to orchestrate so that they the can operation be used of to all orchestrate devices. the operationThe sections of all of devices. this paper are organized as follows: Section2 provides the description of theThe industrial sections problem of this paper addressed are organized in this paper. as follows: In Section Section3 the 2 provides different the modules description that ofhave the been industrial developed problem and addressed integrated in within this paper. the actual In Section cell are 3 discussed. the different The modules execution that of havethe assembly been developed scenario, and with integrated the use ofwithin the developed the actual modulescell are discussed. and the evaluation The executi ofon the of theachieved assembly results, scenario, is outlined with inthe Sections use of the4 and developed5, respectively. modules Finally, and thethe conclusionsevaluation of and the achievedsome areas results, for future is workoutlined are presented in Section in 4 Section and 6 Section. 5, respectively. Finally, the conclusions2. Industrial and Problem some areas Description for future work are presented in Section 6.

2. InduThestrial industrial Problem problem Description examined in this study involves the manipulation and assembly of geometrically complex products, namely, shaver handles and razor heads, including severalThe products industrial of their problem variants. examined The challenge in this presented study involves by this the type manipulation of products is and the assemblyabsence of of symmetry geometrically along morecomplex than products, one axes andnamely, the non-uniform shaver handles geometrical and razor transition heads, includingbetween the several different products section of of theirthe product. variants. The challenge presented by this type of productsFigure is2 theshows absence two variants of symmetry of shaver along handles more and than the one possible axes and states the of non equilibrium-uniform geometricalthat can be found transition when between they are the freely different placed section on a surface. of the Theproduct. proposed assembly system needs to be dexterous enough to accurately orient and position each variant from any of these states, exhibiting robustness and repeatability. Appl.Appl. Sci. Sci. 2021 2021, 11, 11, 5034, 5034 4 4of of 14 14

FigureFigure 2 2shows shows two two variants variants of of shaver shaver handle handles sand and the the possible possible sta statestes of of equilibrium equilibrium Appl. Sci. 2021, 11, 5034 thatthat can can be be found found when when they they are are freely freely placed placed on on a asurface. surface. The The proposed proposed assembly assembly system system4 of 14 needsneeds to to be be dexterous dexterous enough enough to to accurately accurately orient orient and and position position each each variant variant from from any any of of thesethese states, states, exhibiting exhibiting robustness robustness and and repeatability. repeatability.

FigFigureFigureure 2. 2.2. States StatesStates of ofof stable stablestable equilibrium equilibriumequilibrium (poses) (poses)(poses) for forfor two twotwo products products.products. .

TheTheThe assembly assemblyassembly of ofof each eacheach variant variantvariant requires requiresrequires further furtherfurther dexterity dexteritydexterity and andand control controlcontrol since sincesince the thethe process processprocess requiresrequiresrequires a a adifferent differentdifferent sequence sequencesequence of ofof complex complexcomplex motions motionsmotions (both (both(both rotational rotationalrotational and andand translational translationaltranslational along alongalong multiplemultiplemultiple axes) axes)axes) bet betweenbetweenween the thethe two twotwo components. components.components. As AsAs demonstrated demonstrateddemonstrated in inin Figure FigureFigure 3 ,3 the , the first ﬁrst first variant variant hashashas a aspring spring spring-loaded-loaded-loaded pivoting pivoting pivoting razor razor razor head, head, head, whereas whereas whereas the the second the second second one one comesone comes comes with with a with afixed fixed a razor ﬁxedrazor head.razorhead. head.

FigureFigureFigure 3. 3.3. Assembly AssemblyAssembly steps stepssteps of ofof the thethe two twotwo product productproduct variants variants.variants. . Both assembly operations involve the elastic deformation of parts, for the creation BothBoth assembly assembly operations operations involve involve the the elastic elastic deformation deformation of of parts, parts, for for the the creation creation of of of the assembly joints. For the implementation of these motions, four degrees of freedom thethe assembly assembly joints. joints. For For the the implementation implementation of of these these motions, motions, four four degrees degrees of of freedom freedom (DoFs) are required. Finally, for the packaging of the final product, picking, manipulation (DoFs)(DoFs) are are required. required. Finally, Finally, for for the the packaging packaging of of the the final final product, product, picking, picking, manipulation manipulation (90 degrees rotation along the longitudinal axis) and placement in packaging trays are (90(90 degrees degrees rotation rotation along along the the longitudinal longitudinal axis) axis) and and placement placement in in packaging packaging trays trays are are again required. againagain required. required. The human hand can very easily grasp, manipulate and assemble such objects thanks The human hand can very easily grasp, manipulate and assemble such objects thanks to theThe analogue human sensorial hand can feedback very easily from grasp, the nerve manipulate endings, and complemented assemble such with objects the thanks visual to the analogue sensorial feedback from the nerve endings, complemented with the visual perceptionto the analogue provided sensorial bythe feedback eyes and from the the real nerve time endings, processing complemented of this info with by thethe brain.visual perception provided by the eyes and the real time processing of this info by the brain. However,perception when provided discussing by the the eyes automation and the real of suchtime process,processing a flexible of this system info by is the required brain. However, when discussing the automation of such process, a flexible system is required thatHowever, would when be able discussing to adapt tothe different automation products of such geometries process, anda flexible assembly system steps. is required Current thatapproachesthat would would be be rely able able on to to hard adapt adapt automation to to different different devices products products which geometries geometries are custom-made and and assembly assembly and steps. dedicatedsteps. Curr Current theent approachesproduct,approaches making rely rely on iton obsoletehard hard automation automation once a new devices devices product which which is launched are are custom custom in market.-made-made and and dedicated dedicated the the product,product, making making it it obsolete obsolete once once a anew new product product is is launched launched in in market. market. 3. Approach 3.3. Approach ApproachThe scope of this paper is to investigate a complete hardware and control system that canThe performThe scope scope equally of of this this paper efficiently paper is is to to investigate the investigate in-hand a manipulationacomplete complete hardware hardware and assembly and and control control of suchsystem system objects. that that cByancan incorporatingperform perform equally equally flexible efficiently efficiently resources the the in suchin-hand-hand as manipulation robots,manipulation dexterous and and grippersassembly assembly and of of such mechatronicsuch objects. objects. ByassemblyBy incorporating incorporating devices flexible andflexible integrating resource resource thems ssuch such under as as robots, robots, a multilevel dexterous dexterous control grippers grippers architecture, and and mechatronic mechatronic a modular assemblyandassembly reconfigurable devices devices and and solution integrating integrating can bethem them achieved. under under a These amultilevel multilevel elements control control are architecture described architecture in, a detailed, amo modulardular in the following paragraphs and then validated in a case study as presented in Section4.

3.1. 6-DoF Electromechanical Gripper Aiming at avoiding the use of stationary devices, for the intermediate manipulation and the consequent cycle time increase, an electromechanical gripper with in-hand manipulation capabilities has been developed to serve both feeding and packaging operations. Appl. Sci. 2021, 11, 5034 5 of 14

and reconfigurable solution can be achieved. These elements are described in detailed in the following paragraphs and then validated in a case study as presented in Section 4.

3.1. 6-DoF Electromechanical Gripper Appl. Sci. 2021, 11, 5034 Aiming at avoiding the use of stationary devices, for the intermediate manipulation5 of 14 and the consequent cycle time increase, an electromechanical gripper with in-hand manipulation capabilities has been developed to serve both feeding and packaging Aoperations decomposition. A decomposition of handling tasks,of handling as performed tasks, as by performed human operators by human has operators been used has to definebeen used the requiredto define manipulationthe required manipulation degrees of freedom. degrees of Results freedom. will Results drive thewill design drive the of end-effector’sdesign of end- configurationeffector’s configuration space and space the enlistment and the enlistment of actuation of actuation components. components. Integral designIntegral constraint design constraint is the achievement is the achievement of multiple-products of multiple manipulation-products manipulation without hardware without modifications.hardware modifications. FocusingFocusing onon thethe decompositiondecomposition ofof handlinghandling procedure,procedure, allall variantsvariants entail entail a a sequence sequence ofof eighteight steps,steps, namely:namely: approach,approach, grasp,grasp, align,align, regrasp,regrasp, rotate,rotate, translate,translate, regrasp, regrasp, place place and and release.release. ByBy usingusing robot’srobot’s degreesdegrees ofof freedom,freedom, “approach”“approach” andand “place”“place” cancan bebe exceptedexcepted fromfrom designdesign methodology.methodology. Referring Referring to to Figure Figure4 ,4 the, the required required mechanisms mechanisms and and respective respective degreesdegrees of of freedom, freedom, for for feeding feeding and and packaging packaging operations operations of of all all product product variants, variants, are are summarizedsummarized inin TableTable1 .1.

Figure 4. Degrees of freedom for indicative product’s manipulation process steps: (a) grasp and Figure 4. Degrees of freedom for indicative product’s manipulation process steps: (a) grasp and align,align, ( b(b)) release release and and regrasp, regrasp, (c) manipulate(c) manipulate phase phase 1, (d ) 1, manipulate (d) manipulate phase 2, phase (e) release 2, (e) and release regrasp, and (fregrasp,) release ( afterf) release placement. after placement (d4: Prismatic. (d4: jointPrismatic stroke joint of ﬁnger stroke 4; d5:of finger Prismatic 4; d5: joint Prismatic stroke ofjoint ﬁnger stroke 5). of finger 5.) Table 1. Summary of mechanisms and degrees of freedom per handling process. Table 1. Summary of mechanisms and degrees of freedom per handling process. Operation Resource Mechanism DoF Operation Resource Mechanism DoF Approach Robot Not Available Not Available Approach Robot Not Available Not Available Grasp Gripper Gripper Grasping Grasping mechanism mechanism TransY TransY Align Gripper Gripper Grasping Grasping mechanism mechanism TransY TransY Regrasp Gripper Gripper ManipulationManipulation mechanism mechanism TransZ,TransZ, TransX TransX Manipulation Gripper Manipulation mechanism RotY, RotX, TransZ, TransX RotY, RotX, Manipulation Gripper Manipulation mechanism Release Gripper Manipulation mechanism TransZ,TransZ, TransX TransX Place Robot N/A N/A Release Gripper Manipulation mechanism TransZ, TransX Release after placement Gripper Grasping mechanism TransY Place Robot N/A N/A Release after placement Gripper Grasping mechanism TransY

The end-effector includes two mechanisms, namely “Grasping mechanism” and “Manipulation Mechanism”. Considering the grasping fingers as independent kinematic chains, the overall gripper consists of one closed and two open kinematic chains, leading to a total of six DoFs. The two open kinematic chains comprise of one active prismatic joint (P), actuated by a linear actuator. These chains correspond to the fingers of a parallel gripper that is used for the parts’ grasping and re-grasping. This gripper configuration (Figure5) Appl. Sci. 2021, 11, 5034 6 of 14

The end-effector includes two mechanisms, namely “Grasping mechanism” and “Manipulation Mechanism”. Considering the grasping fingers as independent kinematic chains, the overall gripper consists of one closed and two open kinematic chains, leading

Appl. Sci. 2021, 11, 5034 to a total of six DoFs. The two open kinematic chains comprise of one active prismatic6 of 14 joint (P), actuated by a linear actuator. These chains correspond to the fingers of a parallel gripper that is used for the parts’ grasping and re-grasping. This gripper configuration (Figure 5) offers the capability of grasping compensation (due to different object geometries),offers the capability as well ofas graspingalignment compensation correction before (due toand different after the object in-hand geometries), manipulatio as welln processas alignment. The same correction result before is achieved and after by the an in-handarticulated manipulation open chain process. including The sametwo rotary result andis achieved two prismatic by an articulatedjoints with position open chain control. including However, two rotarythis configuration and two prismatic was avoided joints sincewith positionopen chains control. require However, heavier this duty conﬁguration supports and was bearings avoided increasing since open impleme chainsntation require costs.heavier duty supports and bearings increasing implementation costs.

FigureFigure 5. 5. 66-DoF-DoF electromechanical electromechanical gripper gripper.. The in-hand manipulation process is performed by a re-grasping component, attached ontoThe the end-effectorin-hand manipulation of the closed-loop process kinematic is performed chain. This by chain a re-grasping consists of component, three active attachedprismatic onto (P), onethe activeend-effector rotational of the (R), closed two passive-loop kinematic rotational (R)chain. and This one passivechain consists prismatic of three(P) joints active leading prismatic to four (P), DoF. one This active conﬁguration rotational (R), space two enables passive the rotational manipulated (R) and object one to passivebe planarly prismatic moved (P) and joint rotateds leading around to twofour axes,DoF. thus This achieving configuration the desired space pose, enables before the manipulatedregrasping. object to be planarly moved and rotated around two axes, thus achieving the desiredBased pose, on before the conﬁguration regrasping.space illustrated in Figure4, the position and orientation of the manipulationBased on the mechanism’sconfiguration end-effector space illustrated can be in calculated Figure 4, based the position on the closedand orientation kinematic ofchain the forward manipulation kinematic mechanism’s Equation (1). end-effector can be calculated based on the closed kinematic chain forward kinematic Equation (1).

z = d1 + d3cos ϕ1 ϕ1 = atan[L12/(d2 − d1)] z = d1 + d3cosφ1 φ1 = atan[L12/(d2 − d1)] x = d3sin ϕ1 θ1 = rotary motor angle (1) x = d3sinφ1 θy1 ==rotary0 motor angle (1) y = 0 where di is the stroke of “prismatic joint” of finger i, ϕ1 is the rotation caused by the two θ wverticalhere diprismatic is the stroke joints, of “prismatic1 is the rotation joint” of of finger the rotational i, φ1 is the joint rotation and L 12causedis the by distance the two of verticalbetween prismatic the two verticaljoints, θ motors.1 is the rotation of the rotational joint and L12 is the distance of betweenTwo the layers two havevertical been motors. developed for the control of the specific device. Controllers that regulateTwo layers the position have been of each developed actuator for according the control to proportional–integral–derivative of the specific device. Controllers (PID) that regulatecontroller, the acceleration position of each and velocityactuator parameters.according to Thoseproportional parameters–integral form–derivative the lower-level (PID) controllercontrol that, acceleration is run locally and on velocity the device. parameters This layer. Those communicates parameters with form the the intermediate lower-level controllayer, which that is is run responsible locally on for the the device generation. This oflayer trajectories communicates and synchronized with the intermediate motions, in layer,respect which of a device’s is responsible forward for and the inverse generation kinematics. of trajectories Control and layer synchronized communication motions, involves in respectthe exchanging of a device’s of messages forward through and inverse Controller kinematics. Area Network Control bus layer (CAN-bus), communication related to involvestrajectory the goals exchanging execution of and messages the actuator through status Controller monitoring. Area The Network grasping, bus release (CAN and-bus the), in-hand manipulation operations are triggered and synchronized by the station controller, related to trajectory goals execution and the actuator status monitoring. The grasping, according to sensor data. For object identification, those data are summarized as object release and the in-hand manipulation operations are triggered and synchronized by the type, pose, 2D position and orientation.

3.2. Parallel Manipulator for Assembly Operations In current practice, hard-automated cam-actuated assembly machines are used for every variant. These machines have great repeatability and production rates; however, any change in the product geometry or material, condenses them obsolete. For this restriction to be overcome, a mechatronic device has been designed and developed. The designed manipulator needs to be capable of high-speed, accurate and delicate motions Appl. Sci. 2021, 11, 5034 7 of 14

station controller, according to sensor data. For object identification, those data are summarized as object type, pose, 2D position and orientation.

3.2. Parallel Manipulator for Assembly Operations In current practice, hard-automated cam-actuated assembly machines are used for every variant. These machines have great repeatability and production rates; however, Appl. Sci. 2021, 11, 5034 every variant. These machines have great repeatability and production rates; however,7 of 14 any change in the product geometry or material, condenses them obsolete. For this restriction to be overcome, a mechatronic device has been designed and developed. The designed manipulator needs to be capable of high-speed, accurate and delicate motions for assembling the parts of interest. Elaborating on on the the assembly assembly procedures procedures of of Figure 33,, a series primitive translational and rotational relative movements between the parts need to take place. Having Having a a device able to assemble both variants ( (FiguresFigures 66 and and7 )7 without) without hahardwarerdware modiﬁcations modifications prerequires prerequires that that the manipulator the manipulator comprises comprises all degrees all degrees of freedom of freedomas listed inas Tablelisted2 .in Table 2.

Figure 6. Degrees of freedom for ﬁxed head razor: (a) Initial state, (b) Translation for engagement Figure 6. Degrees of freedom for fixed head razor: (a) Initial state, (b) Translation for engagement (c) rotation forfor clipping.clipping.

Figure 7. DegreDegreeses of freedom for pivot head razor: ( a) Initial state, ( b) Translation andand rotationrotation forfor engagement ( c) rotation forfor ﬁrstfirst clippingclipping ((dd)) rotationrotation forfor secondsecond clipping.clipping.

Table 2.2. SummarySummary ofof degreesdegrees ofof freedomfreedom forfor assemblyassembly operationsoperations perper productproduct variant.variant.

OperationOperation ProductProduct Variant variant DoFDoF Translation for endearment Fixed TransZ, TransX Translation for endearment Fixed TransZ, TransX Clipping Fixed RotY Clipping Fixed RotY Translation for endearment Pivot TransZ, TransX TranslationRotation for for endearment endearment PivotPivot TransZ,RotY TransX RotationRotation for endearment of clipping PivotPivot RotYRotX Rotation of clipping Pivot RotX

Aiming for accuracy and robustness, a mini parallel robot configuration has been set up. The necessity of applying symmetrical forces and torques made platform manipulators more appealing that open kinematic chain configurations. The majority of parallel manipulators with multiple rotational and translational degrees of freedom are inspired by the Stewart platform [32]. However, this configuration comes with six degrees of freedom and a high number of actuators. For keeping control and implementation costs low, the parallel robotic device of this study has been designed having only the required degrees of freedom. The proposed robot is composed of a platform supported by four legs, which Appl. Sci. 2021, 11, 5034 8 of 14

Aiming for accuracy and robustness, a mini parallel robot configuration has been set up. The necessity of applying symmetrical forces and torques made platform manipulators more appealing that open kinematic chain configurations. The majority of parallel manipulators with multiple rotational and translational degrees of freedom are Appl. Sci. 2021, 11, 5034 inspired by the Stewart platform [32]. However, this configuration comes with six degrees8 of 14 of freedom and a high number of actuators. For keeping control and implementation costs low, the parallel robotic device of this study has been designed having only the required degrees of freedom. The proposed robot is composed of a platform supported by four legs,are characterized which are characterized as d1, d2, d3 as, d d4 1in, d Figures2, d3, d48 in and Figure9, whilsts 8 and its 9 components, whilst its components can secure the can securerazor heads the razor via pneumatic heads via clamping, pneumatic without clamping, hardware without modifications. hardware Themodifications. same approach The sameis followed approach for theis followed positioning for the of the positioning shavers’ handles.of the shavers’ Each of handles. the legs, Each supporting of the legs, the supportingplatform, is the a distinct platform, kinematic is a distinct chain kinematic consisting chain of aconsisting series of passiveof a series and of activepassive joints. and activeThe first joints. kinematic The first chain kinematic is RPRS chain and is is RP theRS only and oneis the with only an one active with rotationalan active rotational joint (the jointunderlined (the underlined joints are jointsthe active are theones, active while ones, the rest while are the passive rest are joints). passive The joints) remaining. The remainingkinematic chainskinematic have chains RPS, R havePSP, RRPPSPS, R configurationsPSP, RPSP configurations (the underlined (the joints underlined are the activejoints areones, the while active the ones, rest while are passive the rest joints). are passive The platform joints). The (Figure platform8) has ( fourFigure DoF, 8) has enabling four DoF, the enablingexecution theof elegant execution high-speed of elegant trajectories high-speed for the trajectories assembly of forboth the products. assembly The ofnumber both products.of DoF was The confirmed number by ofGruebler’s DoF was equation confirmed as well by as Gruebler’simulationss equationusing a CAD as tool well (i.e., as simulationsDELMIA V5 using by Dassault a CAD Systems,tool (i.e., Paris,DELMIA France). V5 by Dassault Systems, Paris, France).

Appl. Sci. 2021, 11, 5034 9 of 14

Figure 8. Robotic device for assembly operations. Figure 8. Robotic device for assembly operations.

In terms of control, there are two layers, which communicate through a CAN-bus for sending trajectories and monitoring the actuator status and actual position. The intermediate layer the forward and inverse kinematics require a series of computations through vector analysis and analytical methods, based on geometrical constraints. As shown in Figure 9, given the position p and orientation R of the platform, based on vector analysis, the solution of every chain di can be expressed as:

di = p + bi − ai (2) {S}: di = p + Rbi − ai

FigureFigure 9. 9. PlatformPlatform vector vector analysis analysis..

TheIn terms synchronization of control, there of are parallel two layers, manipulator which communicate trajectories, through the PLC a CAN-bus-controlled for pneumaticsending trajectories clamps and and the monitoring part feeding the resources actuator are status performed and actual by position.the station The controller, interme- presenteddiate layer in the the forward next section. and inverse kinematics require a series of computations through vector analysis and analytical methods, based on geometrical constraints. 3.3. ControlAs shown Architecture in Figure of9 ,Mechatronic given the position Devices p and orientation R of the platform, based on vectorA analysis,multi-layer the control solution system of every with chain cognitive di can be capabilities, expressed as: distributed at different levels, has been employed for the subsystems’ exploitation of flexibility and dexterity. d = p + b − a Figure 10 presents the control hierarchyi and cognitioni i functionalities for each level. (2) {S} : di = p + Rbi − ai

Figure 10. Overall system hierarchical architecture.

The lower level includes the hardware components that employ their own local controllers for sensing, actuation and networking operations, either autonomously or in coordination with the higher levels. A typical example, being the actuators of a mechatronic device, such as a robot or a gripper, which can adapt its operation (motion range, applied forces, speed, trajectory etc.) using input from the higher levels. The local control system can be instructed on the fly for the execution of different command sequences, customized to each individual operation. Additionally, this level includes the sensors installed in the production cell that provide real time data on the status of the production and the process execution, namely, part existence/position or the status of assembly. This is used either by the lower-level resources/devices or it is escalated to the higher control level explained hereafter. The intermediate level is implemented in an Information and Communications Technology (ICT)/software context, including the services and topics, playing the role of middleman between the top and bottom layers of the hierarchy discussed. Its operation relies on Robot Operating System (ROS) topics, services and actions that are adapted for each hardware component of the lower level and have a double role. On the one hand, they are translating and dispatching the instructions, coming from the higher control level Appl. Sci. 2021, 11, 5034 9 of 14

Appl. Sci. 2021, 11, 5034 9 of 14 Figure 9. Platform vector analysis.

The synchronization synchronization of of parallel parallel manipulator manipulator trajectories, trajectories, the PLC-controlled the PLC-controlled pneu- maticpneumatic clamps clamps and and the the part part feeding feeding resources resources are are performed performed by by thethe stationstation controller, presented in the next section.

3.3. Control Architecture of Mechatronic Devices A multi-layer multi-layer control control system system with with cognitive cognitive capabilities, capabilities, distributed distributed at different at different levels, haslevels, been has employed been employed for the subsystems’for the subsystems’ exploitation exploitation of ﬂexibility of flexibility and dexterity. and dexterity. Figure 10 presentsFigure 10 the presents control the hierarchy control andhierarchy cognition and functionalitiescognition functionalities for each level. for each level.

Figure 10. Overall system hierarchical architecture. Figure 10. Overall system hierarchical architecture. The lower level includes the hardware components that employ their own local controllersThe lower for sensing, level includes actuation the and hardware networking components operations, that either employ autonomously their own orlocal in coordinationcontrollers for with sensing, the higher actuation levels. and A typical networking example, operations, being the either actuators autonomously of a mechatronic or in device,coordination such as with a robot the or higher a gripper, leve whichls. A can typical adapt example, its operation being (motion the actuators range, applied of a forces,mechatronic speed, device, trajectory such etc.) as a using robot input or a fromgripper, the which higher can levels. adapt The its local operation control (motion system canrange, be instructedapplied forces, on the speed, fly for trajectory the execution etc.) using of different input commandfrom the higher sequences, levels. customized The local tocontrol each individual system can operation. be instructed Additionally, on the fly this for level the includes execution the sensorsof different installed command in the productionsequences, customized cell that provide to each real individual time data operation. on the status Additionally, of the production this level and includes the process the execution,sensors installed namely, in part the existence/positionproduction cell that or theprovide status real of assembly.time data This on the is used status either of the by theproduction lower-level and resources/devices the process execution, or it is namely, escalated part to existence/position the higher control or level the explained status of hereafter.assembly. This is used either by the lower-level resources/devices or it is escalated to the higherThe control intermediate level explained level is implementedhereafter. in an Information and Communications Tech- nologyThe (ICT)/software intermediate level context, is implemented including the in services an Informatio and topics,n and playing Communications the role of middlemanTechnology between(ICT)/software the top context, and bottom including layers the of services the hierarchy and topics, discussed. playing Its the operation role of reliesmiddleman on Robot between Operating the top System and bottom (ROS) topics,layers servicesof the hierarchy and actions discussed. that are Its adapted operation for eachrelies hardware on Robot component Operating ofSystem the lower (ROS level) topics, and haveservices a double and actions role. On that the are one adapted hand, they for areeach translating hardware andcomponent dispatching of the the lower instructions, level and coming have froma double the higherrole. On control the one level hand, and onthey the are other translating hand, they and aredispatching responsible the for instructions, publishing coming sensor datafrom messagesthe higher on control the respec- level tive topics. This provides a more standardized configuration method for the programmer, who can use a common formatting to configure and program all devices and modules. Through this modular approach, multiple hardware devices, sensors or mechatronic machines, can be added, simply by introducing the respective ROS service/topic/action in the intermediate layer that will be able to parse, transform and transfer the information. The central control system that receives sensor data, performs online decision-making and orchestrates the execution of different operations by the resources resides at the top level. The programmer can change the application structure and the connection between sensor input and task execution, through the update of an extensible markup language (XML-formatted) document that contains information on all the resources and their operations. This file is also used to generating the task orchestration sequence. At first, all resources are declared, and a list of tasks is loaded. Based on this list, each task is assigned to a resource, while the sequence of the tasks is generated via the keywords “barrierPre”, which is the enabling condition for the execution of the specific task and the Appl. Sci. 2021, 11, 5034 10 of 14

and on the other hand, they are responsible for publishing sensor data messages on the respective topics. This provides a more standardized configuration method for the programmer, who can use a common formatting to configure and program all devices and modules. Through this modular approach, multiple hardware devices, sensors or mechatronic machines, can be added, simply by introducing the respective ROS service/topic/action in the intermediate layer that will be able to parse, transform and transfer the information. The central control system that receives sensor data, performs online decision- making and orchestrates the execution of different operations by the resources resides at the top level. The programmer can change the application structure and the connection between sensor input and task execution, through the update of an extensible markup language (XML-formatted) document that contains information on all the resources and their operations. This file is also used to generating the task orchestration sequence. At Appl. Sci. 2021, 11, 5034 10 of 14 first, all resources are declared, and a list of tasks is loaded. Based on this list, each task is assigned to a resource, while the sequence of the tasks is generated via the keywords “barrierPre”, which is the enabling condition for the execution of the specific task and the “barrierPost”“barrierPost“ which which is is the the condition condition that that will will be be truetrue after after thethe execution.execution. Additionally,Additionally, in in somesome tasks,tasks, speciﬁcspecific input is is expected, expected, (e.g., (e.g., a asignal signal provided provided by by a sensor), a sensor), defined deﬁned by bythe thekeyword keyword “input”, “input”, while while the the tasks tasks providing providing this this information, information, contain the the keyword keyword “output”.“output”. AnAn exampleexample sequencesequence ofof differentdifferent taskstasks thatthat areare assignedassigned toto differentdifferent resourcesresources andand thethe wayway theythey areare orchestratedorchestrated byby usingusing thethe aforementionedaforementioned conditions,conditions, isis presentedpresented inin FigureFigure 11 11..

FigureFigure 11.11.Task Task orchestrationorchestration based based on on XML XML conﬁguration configuration ﬁle. file. (Ti: (Ti: Task Task number.) number)

TheThe specific specific architecture architecture aims aims at at exploiting exploiting and and combining combining the the operation operation of of local local controlcontrol hubs,hubs, whichwhich havehave specificspecific rolesroles andand cancan bebe easily easily reconfigured. reconfigured. TheThe systemsystem isis anan agnosticagnostic devicedevice andand cancan bebe expandedexpanded withwith thethe additionaddition ofof eithereither newnew instructionsinstructions oror newnew modules;modules; thus,thus,the the timetime andandthe theeffort effort of of system system reconfiguration reconfiguration is is reduced. reduced. 4. Implementation in Consumer Goods Case Study 4. Implementation in Consumer Goods Case Study A case study from the consumer goods industry has been carried out in order to assess A case study from the consumer goods industry has been carried out in order to the performance of the approach. The uniqueness of the scenario has originated from the factassess that the in theperformance same robotic of cell,the approach. a combination The ofuniqueness mechatronic of devices,the scenario orchestrated has originated under thisfrom multi-layer the fact that control in thescheme, same was robotic able to cell, randomly a combination assemble of any mechatronic components devices, with complicateorchestrated and under variable this geometry.multi-layer The control parts scheme, that were was assembled able to randomly are the ones assemble shown any in Figurescomponents2 and 3wi. th complicate and variable geometry. The parts that were assembled are the onesThe designshown in as Figure well ass 2 a and view 3. of the actual setup are shown in Figure 12. This cell comprisesThe design a Racer as 7 well robot as (COMAU, a view of Turin,the actual Italy) setup for the are pick shown and in place Figure operations, 12. This cell an activecomprises reconfigurable a Racer 7 griperrobot for(COMAU, the grasping Turin, and Italy manipulation) for the pick operations, and place a operations, conveyor belt an foractive the transferreconfigurable of the handles,griper for an the assembly grasping machine and manipulation to connect the operations, heads to the a conveyor handles and a vision system for part detection. The handles are randomly placed on the moving conveyor belt, while a controlled lighting area is used for acquiring images by the camera. Under the camera’s field of view, three parameters for each of the handles are determined: • Variant type—fixed or pivoting • Position—handles’ 2D coordinates on the conveyor surface • Orientation—detects which side the handle is laying on. As the conveyor moves to the right, the handle goes through the vision-controlled area towards the robot, which moves above the handle. The gripper controller is then notified through the control architecture of the part’s detected orientation and is actuated accordingly (by its local control) to grasp the handle and perform the necessary manipulation. Appl. Sci. 2021, 11, 5034 11 of 14

belt for the transfer of the handles, an assembly machine to connect the heads to the handles and a vision system for part detection. The handles are randomly placed on the moving conveyor belt, while a controlled lighting area is used for acquiring images by the camera. Under the camera’s field of view, three parameters for each of the handles are determined:  Variant type—fixed or pivoting  Position—handles’ 2D coordinates on the conveyor surface  Orientation—detects which side the handle is laying on. As the conveyor moves to the right, the handle goes through the vision-controlled area towards the robot, which moves above the handle. The gripper controller is then Appl. Sci. 2021, 11, 5034 notified through the control architecture of the part’s detected orientation and is actuated11 of 14 accordingly (by its local control) to grasp the handle and perform the necessary manipulation.

FigureFigure 12.12.( (AA)) CADCAD designdesign and and ( B(B)) Real Real setupsetup ofof thethe assembly assembly cell. cell.

DuringDuring thisthis manipulation,manipulation, thethe robotrobot bringsbrings thethe grippergripper toto thethe feeding feeding position position of of the the assemblyassembly machine,machine, whichwhich hashas alsoalso beenbeen instructedinstructed toto assumeassume itsits feedingfeeding configuration.configuration. AfterAfter the the handle handle has has been been released released by theby the gripper, gripper, the assemblerthe assembler secures secures it and it uses and its uses DoFs its toDoFs insert to theinsert head the into head the into handle, the handle, following following the necessary the necessary movements movements as shown as in shownFigure3 in. UponFigure completion, 3. Upon completion, the assembler the assembler publishes anpublishes operation an completionoperation completion message, enablingmessage, theenabling robot/gripper the robot/gripper controller controller to work together to work in together removing in the removing part from the the part feeder from and the proceedingfeeder and withproceeding the next with cycle. the next cycle. 5. Results and Discussion 5. Results and Discussion The proposed solution has been set up and run in a laboratory environment, present- The proposed solution has been set up and run in a laboratory environment, ing a high repeatability rate at low production rates. The combination of technologies has presenting a high repeatability rate at low production rates. The combination of managed to assemble more than 500 handles of both types that were randomly fed to the technologies has managed to assemble more than 500 handles of both types that were system, having demonstrated a 98% success rate. The robotic cell’s perception system was randomly fed to the system, having demonstrated a 98% success rate. The robotic cell’s able to identify the variant of the product as well as its original pose with 100% accuracy. perception system was able to identify the variant of the product as well as its original This contributed to the reconfiguration of mechatronics that with no hardware modifi- pose with 100% accuracy. This contributed to the reconfiguration of mechatronics that cations were able accomplish all assembly operations. This shows that the system has with no hardware modifications were able accomplish all assembly operations. This achieved its flexibility goal, being able to assemble products of different geometries, follow- shows that the system has achieved its flexibility goal, being able to assemble products of ing different processes and to accommodate the uncertainty attributed to the unconstrained motiondifferent of geometries, the complex following parts on different a conveyor processes surface. and The to experimental accommodate results the uncertainty (Table3) indicateattributed that to in-handthe unconstrained manipulation motion and of assembly the complex are theparts most on a susceptible conveyor surface operations,. The whichexperimental result from results their (Table complexity. 3) indicate The performancethat in-hand evaluationmanipulation aimed and on assembly the assessment are the ofmost the system susceptible as whole operations, rather the which performance result from of the their devices complexity. and modules The as performance individual entities.evaluation Thus, aimed the evaluationon the assessment of operations of the towardssystem as the whole end ofrather the assemblythe performance process of (e.g., the transferringdevices and tomodules packaging as indivi area) consistdual entities. of several Thus, tests the that evaluation all preceding of operations operations towards were successful. Compared to the current practice, although it was not the primary target, the production rate is considered lower, since for the current product, the custom-made machines are optimized for large volumes and their repeatability is also closer to 100%. The implemented robotic cell demonstrated a mean cycle time over 11.2s whereas traditional assembly lines reach up to 2s per part. The performance of the robotic cell can be further increased, however, the robustness is affected, in terms of successful assembled products. Future work involves optimization of product-oriented components for ensuring more effective handling allowing greater velocities. Nevertheless, more robots and devices can be added to the system to increase the system’s output without being constrained to a single product. The use of easily exchangeable parts on the grippers and assemblers allows them to be used over multiple product generations when compared to hard automation. Unlike the case of hard automation, in the case of a machine break-down or a scheduled maintenance, the production can be transferred to another line thanks to the modular hardware and control software. Appl. Sci. 2021, 11, 5034 12 of 14

Table 3. Experimental results.

Total Successful Operation Variant Resource Percentage Tests Operations Detection of pose on the conveyor Vision 500 500 100.00% Grasping Gripper 1 500 500 100.00% In-hand manipulation Gripper 1 500 496 99.20% Transferring to assembly station Robot 1 496 496 100.00% AssemblyFixed Parallelmanipulator 496 492 99.19% Grasping Gripper 2 492 492 100.00% In-hand manipulation for packaging Gripper 2 492 491 99.80% Transferring to packaging area Robot 2 491 491 100.00% Release in tray Gripper 2 491 491 100.00% Total 98.20% Detection of pose on the conveyor Vision 504 504 100.00% Grasping Gripper 1 504 504 100.00% In-hand manipulation Gripper 1 504 498 98.81% Transferring to assembly station Robot 1 498 498 100.00% AssemblyPivot Parallelmanipulator 498 489 98.19% Grasping Gripper 2 489 489 100.00% In-hand manipulation for packaging Gripper 2 489 488 99.80% Transferring to packaging area Robot 2 488 488 100.00% Release in tray Gripper 2 488 488 100.00% Total 96.83%

6. Conclusions This paper introduced a flexible assembly cell composed by cognitive mechatronic devices able to handle complex parts, integrated in a hierarchical, multi-level control architecture for the implementation of flexible coordination and control schemes. The utilized devices performed manipulation, assembly and packaging operations of different parts and variants, proving that they can be easily updated and reconfigured both from software and hardware side. Additionally, the multi-level control approach allows the system configuration and enhancement with similar devices based on production needs, leading to a higher overall flexibility potential. In the future, effort will be made on improving the production rate, with the use of several faster resources. For example, SCARA-type robots will be introduced to perform faster feeding, while stronger motors will be installed in the mechatronic devices. Apart from this, multiple assembly devices will be implemented to run in parallel and the break- even point for an efficient return of investments should be estimated. Optimization of the algorithms that coordinate all these resources is required.

Author Contributions: Conceptualization, P.K. and G.M.; Investigation, P.K., D.A. and A.-S.M.; Methodology, P.K. and G.M.; Software, D.A., A.-S.M. and C.G.; Supervision, S.M.; Writing—original draft, P.K. and D.A.; Writing—review & editing, P.K., G.M. and S.M. All authors have read and agreed to the published version of the manuscript. Funding: This work has been partially funded by the EC research project “ESMERA—European SMEs Robotics Applications” (Grant Agreement: 780265). Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Appl. Sci. 2021, 11, 5034 13 of 14

Data Availability Statement: Not applicable. Conﬂicts of Interest: The authors declare no conﬂict of interest.

References 1. Chryssolouris, G. Manufacturing Systems: Theory and Practice; Springer: New York, NY, USA, 2006. 2. Reinhart, G. Assembly Automation. In CIRP Encyclopedia of Production Engineering; Metzler, J.B., Ed.; Springer: Berlin/Heidelberg, Germany, 2014; pp. 52–54. 3. Zaeh, M.; Gao, R. Mechatronics. In CIRP Encyclopaedia of Production Engineering; Springer: Berlin/Heidelberg, Germany, 2014; pp. 861–866. 4. Altintas, Y. Adaptive Control. In CIRP Encyclopedia of Production Engineering; Metzler, J.B., Ed.; Springer: Berlin/Heidelberg, Germany, 2014; pp. 17–19. 5. Lien, T.K. Robot. In CIRP Encyclopaedia of Production Engineering; Springer: Berlin/Heidelberg, Germany, 2014; pp. 1068–1076. 6. Gania, I.; Stachowiak, A.; Ole´sków-Szłapka, J. Flexible manufacturing systems: Industry 4.0 solution. DEStech Trans. Eng. Technol. Res. 2018.[CrossRef] 7. Florescu, A.; Baraba¸s,S.; Sârbu, F. Operational parameters estimation for a flexible manufacturing system. A case study. In Proceedings of the MATEC Web of Conferences; EDP Sciences: Les Ulis, France, 2017; Volume 112, p. 05008. [CrossRef] 8. Spiliotopoulos, J.; Michalos, G.; Makris, S. A Reconfigurable Gripper for Dexterous Manipulation in Flexible Assembly. Inventions 2018, 3, 4. [CrossRef] 9. Rahman, N.; Canali, C.; Caldwell, D.G.; Cannella, F. Dexterous Gripper Synthesis from Modular Finger Approach. In Proceedings of the 2017 ASME International Design Engineering Technical Conferences and Computers and Information in Engineering Conference (13th ASME), Cleveland, OH, USA, 6–9 August 2017. 10. Fantoni, G.; Hansen, H.N.; Santochi, M. A new capillary gripper for mini and micro parts. CIRP Ann. 2013, 62, 17–20. [CrossRef] 11. Fantoni, G.; Santochi, M.; Dini, G.; Tracht, K.; Scholz-Reiter, B.; Fleischer, J.; Lien, T.K.; Seliger, G.; Reinhart, G.; Franke, J.; et al. Grasping devices and methods in automated production processes. CIRP Ann. 2014, 63, 679–701. [CrossRef] 12. Gabriel, F.; Fahning, M.; Meiners, J.; Dietrich, F.; Dröder, K. Modeling of vacuum grippers for the design of energy efficient vacuum-based handling processes. Prod. Eng. 2020, 14, 545–554. [CrossRef] 13. Vongbunyong, S.; Kara, S.; Pagnucco, M. Application of cognitive robotics in disassembly of products. CIRP Ann. 2013, 62, 31–34. [CrossRef] 14. Manolescu, A.; Fota, A.; Oancea, G. Engineering Systems Design and Analysis. In Proceedings of the ASME 2012 11th Biennial Conference on Engineering Systems Design and Analysis, Nantes, France, 2–4 July 2012; ASME: New York, NY, USA, 2012; Volume 4, pp. 457–466. 15. Bogue, R. Sensors for robotic perception. Part one: Human interaction and intentions. Ind. Robot. Int. J. 2015, 42, 386–391. [CrossRef] 16. Pfeiffer, M.; Schaeuble, M.; Nieto, J.; Siegwart, R.; Cadena, C. From perception to decision: A data-driven approach to end-to-end motion planning for autonomous ground robots. In Proceedings of the 2017 IEEE International Conference on Robotics and Automation (ICRA), Singapore, 29 May–3 June 2017; pp. 1527–1533. 17. Argyrou, A.; Giannoulis, C.; Sardelis, A.; Karagiannis, P.; Michalos, G.; Makris, S. A data fusion system for controlling the execution status in human-robot collaborative cells. Procedia CIRP 2018, 76, 193–198. [CrossRef] 18. Valentini, G.; Hamann, H.; Dorigo, M. Efficient Decision-Making in a Self-Organizing Robot Swarm: On the Speed versus Accuracy Trade-Off. In Proceedings of the 2015 International Conference on Autonomous Agents and Multiagent Systems (AAMAS ’15), Istanbul, Turkey, 4–8 May 2015; pp. 1305–1314. 19. Alaieri, F.; Vellino, A. Ethical Decision Making in Robots: Autonomy, Trust and Responsibility. In International Conference on Social Robotics; Springer: Cham, Switzerland, 2016; pp. 159–168. 20. Glogowski, P.; Lemmerz, K.; Hypki, A.; Kuhlenkötter, B. ROS-Based Robot Simulation in Human-Robot Collaboration. In Developing Support Technologies, Biosystems & Biorobotics; Karafillidis, A., Weidner, R., Eds.; Springer International Publishing: Cham, Switzerland, 2018; pp. 237–246. [CrossRef] 21. Scholz-Reiter, B.; Görges, M.; Philipp, T. Autonomously controlled production systems—Influence of autonomous control level on logistic performance. CIRP Ann. 2009, 58, 395–398. [CrossRef] 22. Vrabiˇc,R.; Kozjek, D.; Malus, A.; Zaletelj, V.; Butala, P. Distributed control with rationally bounded agents in cyber-physical production systems. CIRP Ann. 2018, 67, 507–510. [CrossRef] 23. Krüger, J.; Wang, L.; Verl, A.; Bauernhansl, T.; Carpanzano, E.; Makris, S.; Fleischer, J.; Reinhart, G.; Franke, J.; Pellegrinelli, S. Innovative control of assembly systems and lines. CIRP Ann. 2017, 66, 707–730. [CrossRef] 24. Makris, S.; Chryssolouris, G. Web-services-based supply-chain-control logic: An automotive case study. Int. J. Prod. Res. 2013, 51, 2077–2091. [CrossRef] 25. Michalos, G.; Kaltsoukalas, K.; Aivaliotis, P.; Sipsas, P.; Sardelis, A.; Chryssolouris, G. Design and simulation of assembly systems with mobile robots. CIRP Ann. 2014, 63, 181–184. [CrossRef] 26. Michalos, G.; Kousi, N.; Makris, S.; Chryssolouris, G. Performance Assessment of Production Systems with Mobile Robots. Procedia CIRP 2016, 41, 195–200. [CrossRef] Appl. Sci. 2021, 11, 5034 14 of 14

27. Kousi, N.; Michalos, G.; Makris, S.; Chryssolouris, G. Short-term Planning for Part Supply in Assembly Lines Using Mobile Robots. Procedia CIRP 2016, 44, 371–376. [CrossRef] 28. Fragapane, G.; Ivanov, D.; Peron, M.; Sgarbossa, F.; Strandhagen, J.O. Increasing ﬂexibility and productivity in Industry 4.0 production networks with autonomous mobile robots and smart intralogistics. Ann. Oper. Res. 2020.[CrossRef] 29. Kildal, J.; Ipiña, I.; Martín, M.; Maurtua, I. Collaborative assembly of electrical cabinets through multimodal interaction between a robot and a human worker with cognitive disability. Procedia CIRP 2021, 97, 184–189. [CrossRef] 30. Papanastasiou, S.; Kousi, N.; Karagiannis, P.; Gkournelos, C.; Papavasileiou, A.; Dimoulas, K.; Baris, K.; Koukas, S.; Michalos, G.; Makris, S. Towards seamless human robot collaboration: Integrating multimodal interaction. Int. J. Adv. Manuf. Technol. 2019, 105, 3881–3897. [CrossRef] 31. Michalos, G.; Kousi, N.; Karagiannis, P.; Gkournelos, C.; Dimoulas, K.; Koukas, S.; Mparis, K.; Papavasileiou, A.; Makris, S. Seamless human robot collaborative assembly—An automotive case study. Mechatronics 2018, 55, 194–211. [CrossRef] 32. Stewart, D. A Platform with Six Degrees of Freedom. Proc. Inst. Mech. Eng. 1965, 180, 371–386. [CrossRef]