Human–robot collaborative assembly in cyber-physical production: Classification framework and implementation ifi- the lue- sing ices tc.). ical po- tive d of dle ven trol ling and f a and ans, heir...

Write a page reflection on the applications, opportunities, and challenges of human-robot collaboration in manufacturing operations. Use this article as a base reference.


Human–robot collaborative assembly in cyber-physical production: Classification framework and implementation ifi- the lue- sing ices tc.). ical po- tive d of dle ven trol ling and f a and ans, heir and rely rd- ots ing CIRP Annals - Manufacturing Technology 66 (2017) 5–8 d on bly rks. obot tive d to IRP. Human–robot collaborative assembly in cyber-physical production: Classification framework and implementation Xi Vincent Wang a,*, Zsolt Kemény b, József Váncza (1)b,c, Lihui Wang (1)a aDepartment of Production Engineering, KTH Royal Institute of Technology, Sweden b Institute for Computer Science and Control, Hungarian Academy of Sciences, Hungary cDept. of Manufacturing Science and Engineering, Budapest University of Technology and Economics, Budapest, Hungary 1. Introduction Industrial production is nowadays experiencing changes that shift the emphasis of production—and related R&D work—towards increasing flexibility and responsiveness of production processes, facilities and entire production networks. Among the drivers of these changes, the key objectives are decreasing desired lead time and growing customisation, leading to higher diversity and more frequent changes of products, components and tasks to be handled within the same production unit. These trends are expected to affect the way both humans and machines are put to work—most importantly, the meaningful combination of human and robot skills is beginning to gain emphasis. The latter development aligns well with the shift towards more local autonomy in production processes: while certain routine tasks or specific skills can be efficiently supported by automation, local decisions or exceptional intervention often require a “human touch” due to the extraordinary characteristics of the given situation, the complexity, or the implicit nature of knowledge to be relied on in finding a viable solution in a limited time, with bounded resources at hand. The combination of human and artificial resources has not been part of mainstream automation practice where (1) robots and humans are generally kept away from each other, and (2) humans must adhere to work procedures as rigid as the rest of the automated production environment. Symbiotic Human–Robot Collaboration (HRC) steps beyond these limitations but requires Information and Communication Technology (ICT) has sign cantly changed assembly systems in the past years, partly due to massive connectivity of components and actors (LAN, Wi-Fi, B tooth, near field communication, etc.), and partly due to increa process observability and local computing capacity in smart dev (automatic identification, sensors, wearabledevices, smart tags, e The close and multi-directional interaction of virtual and phys entities forms a cyber-physical system where automated com nents and humans can be integrated in a cybernetic and collabora environment combining their complementary strengths instea mutual restriction of their potentials [4]. Robots exhibit high precision and repeatability, can han heavy loads and operate without performance deterioration e in difficult or dangerous environments. However, robot con systems quickly reach their limits in recognizing and hand unexpected situations, as reflected by the relatively rigid plans robot programs widespread in today’s automated systems. Humans tackle unexpected situations better, are aware o much larger part of the environment than formally declared, show more dexterity in complex or sensitive tasks. Hum however, are more prone to error, stress or fatigue [5], and t employment underlies strict health and safety regulations. With technologies able to bridge the gaps in skills operational characteristics, it is now becoming possible to on robots as collaborating partners instead of—potentially haza ous—tools [6]. The appearance of off-the-shelf industrial rob certified for operating alongside humans is a sign of HRC gain A R T I C L E I N F O Article history: Available online 26 April 2017 Keywords: Assembly Man–machine system Human–robot collaboration A B S T R A C T The production industry is moving towards the next generation of assembly, which is conducted base safe and reliable robots working in the same workplace alongside with humans. Focusing on assem tasks, this paper presents a review of human–robot collaboration research and its classification wo Aside from defining key terms and relations, the paper also proposes means of describing human–r collaboration that can be relied on during detailed elaboration of solutions. A human–robot collabora assembly system is developed with a novel and comprehensive structure, and a case study is presente validate the proposed framework. © 2017 Published by Elsevier Ltd on behalf of C Contents lists available at ScienceDirect CIRP Annals - Manufacturing Technology journal homepage: http: / /ees.elsevier.com/cirp/default .asp RC a more responsive, transparent and accessible environment backed by more computational intelligence [1–3]. ach and g to tion* Corresponding author. E-mail address: [email protected] (X.V. Wang). http://dx.doi.org/10.1016/j.cirp.2017.04.101 0007-8506/© 2017 Published by Elsevier Ltd on behalf of CIRP. acceptance and spreading in industrial production. Meanwhile the theoretical and technological supports for H are still undergoing notable development. A systematic appro to solutions involving HRC requires an efficient framework, methodologies for elaborating feasible solutions. Contributin these, the paper proposes a structured classification and solu framework, illustrated by practical examples. http://crossmark.crossref.org/dialog/?doi=10.1016/j.cirp.2017.04.101&domain=pdf http://dx.doi.org/10.1016/j.cirp.2017.04.101 mailto:[email protected] http://www.sciencedirect.com/science/journal/00078506 http://dx.doi.org/10.1016/j.cirp.2017.04.101 2. S H and subj min regu both inte 2.1. I hum asse [7]. wor hold dem also asse have heav prop feed safe [10] cont both the depl train 2.2. A men appe its s mer quan prop � Te (a va [1 (p tim ph pi op Fig. 1 proce X.V. Wang et al. / CIRP Annals - Manufacturing Technology 66 (2017) 5–86 tate of the art in HRC and its classification RC research has been ongoing for decades, with service robots vehicles operating in unstructured environment being the ect of most intense interest. Industrial production has been a ority field in this regard, partly due to health and safety lations limiting HRC in practice. Yet, recent results indicate the revision of established views and regulations, and nsified research of HRC in production scenarios. Human–robot collaboration for assembly ndustrial robots had been expected to work as the assistant of an workers for a long time, comprising a fast and automatic mbly system and collaborative manufacturing environment Different robot and gripper structures are developed to assist kerson the assembly line. Ingeneral, mostof the tasks focusedon ing an object for the person, laying it aside or retrieving it on and [7,8]. In recent years, the production engineering society gave considerable attention to the collaborative systems in mbly lines. Human–machine and human–robot interactions been identified as a feasible solution especially suitable for y and bulk component handling [9]. Morioka and Sakakibara [2] osed an assembly system based on HRC. The cooperative parts ing station is established based on information support and ty management mechanisms. At the control level, Krüger et al. proposed a framework design for stable and robust interaction rol. Intuitive programming mechanisms were developed for onlineand offline programming via gestures and voices [11]. For satisfaction of human needs, augmented reality was also oyed in the factory for virtual assembly, assembly guidance, ing, maintenance, etc. [12]. Classification of human–robot collaboration nswering the need for a systematic analysis of HRC require- ts and adequate solutions, several considerations have ared to classify or characterise individual cases [8]. Due to pecific constraints, industrial production usually occupies a e subset of possibilities. While some characteristics are titative, most of the studies highlighted a small number of erties that define distinct classes of HRC instances: mporal and spatial relation of collaborating humans and robots gents in a more generic sense)—while this shows wide riation in cases like teleoperation or assisted vehicle steering 3], it is assumed in industrial production that the agents artially) share the same space, even though their activity over e does not necessarily overlap [7]. Close collaboration with ysical contact [14]—e.g., common handling of large work- eces—does, naturally, require co-location and simultaneous eration [15]. � Agent multiplicity can be covered to its full diversity by industrial HRC applications. Literature commonly distinguishes between single, multiple, and team (Fig.1 left), the latter being a group acting by consensus or coordination, and interacting with the environ- ment and other agents in a specified way (e.g., via a “spokesman”). Multiple agents can compete for resources and other agents’ services (e.g., one robot serving several manned workstations). � Agent autonomy and closely related leader–follower relationships express how much of robot action is directly determined by human agents, or which agent takes the lead in the given task. Partitioning along autonomyor initiative canvarydependingonthe application field [16]. In an industrial context, inactive (resting), active (leading), and supportive (following) behaviour can be distin- guished, and many of the current considerations assume that these roles are assigned before task execution (Fig. 1 middle). In some cases, adaptive agents are also contemplated that assign leader/ follower roles on-the-fly—these have gained little practical significance so far, and will be omitted in this paper. Other aspects, such as modes of sensing, interaction, (mutual) awareness are typically treated as independent (orthogonal) characteristics that rarely form distinct classes. 3. Requirements of symbiotic HRC in assembly 3.1. Symbiotic HRC structure in assembly Symbiotic collaboration is set aside from conventional HRC by several key characteristics: � intuitive and multimodal programming environment: workers do not need prior in-depth knowledge of the system, � zero-programming: ideally, the workers can work with the robots via gestures, voice commands, and other forms of natural inputs without the need of coding, � immersive collaboration: with the help of different devices, e.g. screens, goggles, wearable displays, the workers can collaborate with the robots with actively engaged senses, and � context/situation dependency: the system should be capable of interleaving autonomous human with robot decisions based on trustworthy inputs from on-site sensors and monitors inspecting both humans and robots. The systematic elaboration of solutions for such collaborative cases requires an analysis and synthesis framework containing (1) means to classify and characterise the problem, and (2) solution templates and guidelines for elaboration of a solution seamlessly fitting into existing production premises. Fundamental elements in characterising an HRC scenario are: � The actors, i.e., robots and humans either actively taking part in the production process, or occupying a part of the available space, . Classification schemes of human–robot collaboration with regard to agent multiplicity (left), initiative (middle), and alignment of human actions with the nominal ss definition (right). of n a sely tual the nal s, or ent, an- tion t of led tion ng- rce ring tem nal and ess and s. ned be ted ces. ach .) is els, ice, be e.g., ut- the ge/ ned also s it rface obot X.V. Wang et al. / CIRP Annals - Manufacturing Technology 66 (2017) 5–8 7 along with multiplicities and roles that outline where prioritising or conflict
Oct 28, 2021
SOLUTION.PDF

Get Answer To This Question

Related Questions & Answers

More Questions »

Submit New Assignment

Copy and Paste Your Assignment Here