Exoskeleton-centered Process Optimization in ... - ScienceDirect

Available online at www.sciencedirect.com

ScienceDirect Procedia CIRP 41 (2016) 740 – 745

48th CIRP Conference on MANUFACTURING SYSTEMS - CIRP CMS 2015

Exoskeleton-centered process optimization in advanced factory environments Carmen Constantinescu a,b,*, Daniela Popescub, Paul-Cristian Muresana,b, , Sebastian-Ioan Stanaa,b, a Fraunhofer Institute for Industrial Engineering (IAO), Nobelstr. 12, Stuttgart 70569, Germany Faculty of Machine Building, Technical University of Cluj-Napoca (TUC-N), Memorandumului 28, Cluj-Napoca 400114,Romania

b

* Corresponding author. Tel.: +49-175-575-1155 ; E-mail address: [email protected]

Abstract The European manufacturing sectors have to face the challenge of workplace safety and security representing costs up to 4% of GNP. A userfriendly and cooperative human-robotic exoskeleton for manual handling work can represent a possible innovative solution. For highlighting and revealing the benefits of human-robotic exoskeleton, represented by increased employment and productivity and reduced back injuries, the validation of this concept in digital and virtual environment has been performed, in two relevant industry sectors, car disassembly and automotive suppliers. The digital models of the test cases and the simulation of ergonomic workplaces were realized by using state-of-the-art Digital Manufacturing technologies and systems. This simulation technology allowed the realization of two states of factory environment, the “as-is” state without an exoskeleton and the to-be state after employing the use of an exoskeleton. The challenges of coupling the exoskeleton to the human operators in Digital Manufacturing Systems, the first results and the next steps, are presented as well.

© © 2015 2015 The The Authors. Authors. Published Published by by Elsevier Elsevier B.V. B.V.This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the Scientific Committee of 48th CIRP Conference on MANUFACTURING SYSTEMS - CIRP CMS Peer-review under responsibility of the scientific committee of 48th CIRP Conference on MANUFACTURING SYSTEMS - CIRP CMS 2015 2015. Keywords: Modelling; Simulation; Robotics; Workplace; Ergonomics.

1. Motivation The manufacturing strategy in the automotive industry focuses on employing innovative and effective procedures and work processes, ranging from depollution to targeted dismantling and then recovering the components. The industry of the future has to represent a source of sustainable employment and to form an integral part of the growth of industrial ecology [1]. The automotive industry faces the shift of the focus from car assembly to dismantling and car disassembly. The required flexibility here represents one of the main challenges for manufacturing [2, 3]. Once dismantled, the reusable components represent a profitable supply source both for the repairs industry and the end-user [4]. And even further down the line, the steel, metal and plastics are recycled, and can be

re-used again for manufacturing new vehicle, thus further extending the life of the materials [5]. In this new industry as well as in car assembly, many workers are affected by health problems especially of the lower back because of repetitive lifting heavy weights like seats and batteries. The costs for health insurance in these factories are very high; in order to avoid such difficulties the Exoskeleton concept and philosophy will support the workers in lifting and manipulating weights thus relevant improving their health. The motivation of the work presented in this paper is to support the decision of employing the Exoskeleton in such type of factories by showing in digital world, through modelling and simulation, which are the expected benefits. The digital models of the identified test cases in car assembly and disassembly and the simulation of ergonomic workplaces are developed by using state-of-the-art Digital Manufacturing technologies and systems, which allowed the

2212-8271 © 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the scientific committee of 48th CIRP Conference on MANUFACTURING SYSTEMS - CIRP CMS 2015 doi:10.1016/j.procir.2015.12.051

Carmen Constantinescu et al. / Procedia CIRP 41 (2016) 740 – 745

realization of two states of factory environment, before employment of the exoskeleton and after. The market of commercial simulation system is very mature and the modelling and simulation of work places with digital humanoid already proved its benefits. As the Exoskeleton itself is a new and innovative concept, the modelling and simulation systems have to-be enhanced with a new type of digital worker enforced by the Exoskeleton. This represents the contribution of the work presented in this paper. The challenge of coupling the Exoskeleton to the human operators in Digital Manufacturing Systems, the first results and the next steps, represent the content of this paper.

741

recognition will be included in the final design, to ease the work and also to increase the level of safety and security of the worker. The Exoskeleton compound parts are clustered in four main systems, as shown in Fig. 2: structural parts, assisting actuators, intelligent device and actuators, and passive structures for assisted movement [1, 2, 6].

2. Design of an Exoskeleton in manufacturing industry: Robo-Mate The innovation of Robo-Mate * approach consists of the first initiative to employ the Exoskeleton concept in production areas having as main goal to keep the sites of production in Europe and to maintain and increase the employment in European factories. The initiative is very ambitious “as-is” the first time when this type of robotics development is thought for manufacturing industry. The Robo-Mate idea was embraced with enthusiasm by key players in automotive industry and suppliers, e.g. Renault, Fiat, and Compa. Starting with a very deep requirements analysis of industry needs to the final modelling and simulation of the identified use-cases, the RoboMate is represented by a light to medium-weight exoskeleton with actuators that assists workers in manipulating heavy parts in many various fields such as automotive industry. The benefits showed by the exoskeletons in medical and military areas have to-be migrated into the manufacturing industries for manipulating and handling heavy goods in discrete manufacturing and assembly/disassembly operations [5, 6]. The concept of how the Robo-Mate Exoskeleton will look like is presented in Fig. 1.a and 1.b.

Fig. 2. Main systems of the Robo-Mate Exoskeleton (©Robo-Mate)

The Exoskeleton controller is based on a complex sensor system which enables to consider the force and pressure applied by the operator, to recognize his intention and properly manipulate the parts of the Exoskeleton. As the main benefit of the Exoskeleton is to reduce the workrelated injuries, the ergonomics aspects represent one of the main focus points. Modelling and simulation of an ergonomic work place represent a crucial dimension of Robo-Mate. 3. Approach for Modelling and Simulating an Exoskeleton in the Manufacturing Industry In order to investigate the use of Exoskeleton in manufacturing industries, the modelling and simulation of workplace conditions and ergonomics analysis represent, play a critical role [4, 7]. 3.1. The challenge of a new resource for digital manufacturing: Human Operator and Exoskeleton

Fig. 1a. Front View Fig. 1b. Back View Fig. 1. First image and idea representation of Robo-Mate Exoskeleton (©Robo-Mate)

The Robo-Mate Exoskeleton consists of an actuated upperbody, a supportive lower body and electrical batteries. Tag readers, intelligent grippers, head-up display and video

The work presented in this paper is founded by the European Commission under the framework of the project “Robo-Mate: Intelligent exoskeleton

As previously explained Digital Manufacturing play a crucial role in factory planning towards achieving competitiveness goals set by the European Industry. Keeping a high employment rate and still having work efficiency from the state-of-the-art of robotics field has not achieved the success in manufacturing industries. In order to maintain the employment rate as high as possible, the human workforce should be involved in a more efficient way by increasing their motivation and commitment. The Exoskeleton approach is a promising

based on human-robot interaction for manipulation of heavy goods in Europe's factories of the future”. Available at: http://www.robo-mate.eu/

742

Carmen Constantinescu et al. / Procedia CIRP 41 (2016) 740 – 745

solution which answers these challenges by engaging the human workforce to use at maximum their main capabilities skills and its tacit knowledge. The main objective here is to keep the operator safe, secure and healthy [8]. The methodology followed in this work starts with a deeper analysis of the production area and of the corresponding work places where the Exoskeleton seems to-be suitable. After the selection of the workplace where from the point of view of ergonomics specialist the Exoskeleton could bring benefits, the modeling and simulation of the “as-is” situation is performed. The “to-be” status of the workplace, where the Human Operator and the Exoskeleton act as a single and new type of manufacturing resource, should be implemented [6]. Having such a new resource (Human and Exoskeleton) in the library of a Simulation Software brings the big challenge of coupling the digital humanoid, e.g. Jack in the case of Siemens Process Simulate, and the Exoskeleton. In Fig. 3 is envisioned a fist humanoid digital model of an Exoskeleton [9]. The complexity of this coupling is generated by the fact that the digital humanoid kinematics is very complex, having 360 Joints and an inverse kinematic model [10, 11]. The Exoskeleton has a less complex kinematics system which should work in perfect collaboration with the human kinematics. Fig. 4 a better gives understanding of the joints and their new constrains. Here are presented the joints of the humanoid arm, the joints marked in red represents the constraints which are Fig. 3.First model of a digital coming from the humanoid and Exoskeleton Exoskeleton designer’s (©Robo-Mate) team. In the case of the wrist, the joint connecting the hand with the forearm, the red joint is blocked by the Exoskeleton constrains, and the joints represented by green are modified by decreasing the amplitude from 180° down to approximately 150°. The challenge here was to identify how the Exoskeleton modifies and constrains the human joints. The human operator`s joints are mostly defined by the parameters: torque, velocity, dimension, position and weight [12, 13]. The joints of each human body are characterized by different properties. A general equation to describe all humans joints properties could not be yet defined; there for a standard humanoid was created in order to-be used in all modelling and simulation tolls and systems. The Exoskeleton follows a modular concept, consisting of a mandatory trunk (as in Fig. 1) and various combinations of passive and active arms, as presented in Fig. 5. Each module will affect the humanoid in a

different way, giving a different gravity center, another kinematics and other modified parameters. In this paper the main focus resides on the Exoskeleton´s arms and their coupling to the digital humanoid. The extension with the trunk coupling will represent the topic in other research paper [11, 14].

Fig. 4. Limited arm joints by the Exoskeleton design (©Robo-Mate)

The main output of coupling of arms to the digital humanoid is further on called JackEx and can be realized in three instances [12, 6, 14]. The first instance consists of two active arms (powered by motors) Fig. 5a, the second instance consists of two passive (powered by springs) arms, Fig. 5b, and the third instance is composed of a passive and an active arm Fig. 5c.

Fig. 5a. Active arms .

Fig. 5b. Passive arms

Fig. 5c.Passive arm and active arm

Fig. 5: The Exoskeleton Instances (©Fraunhofer IAO)

Typical human modelling and simulation tools such as Delmia and Jack used for production optimization have a library of manufacturing resources, where the human operators or the machines, devices, robots are integrated in the software. When simulating and optimizing the production area, complex mathematical algorithms are used [9]. The parameters of the human models are pre-defined and they can be modified in order to detect the collisions between human and machines. At the same time, the ergonomics analysis works under a different

743

Carmen Constantinescu et al. / Procedia CIRP 41 (2016) 740 – 745

mathematical model. After all mentioned steps were performed the next challenge consists of integrating JackEx (Jack and Exoskeleton) in one of the commercial simulation tool. In this paper the work is realized by using Siemens Process Simulate Software. The JackEx was added to the software libraries as a new resource and an ergonomics analysis with JackEx has to-be realized. Industry end users interested to see the efficiency of the Exoskeleton in their factories will run a simulation by replacing the human operator with JackEx. In Fig. 6 a first envisioned JackEx is presented. Based on the simulation results the Fig. 6. JackEx first envision planner workplace (©Fraunhofer IAO) could adjust the layout accordingly and re-run the simulation, having a holistic overview of the benefits induced by the JackEx [6, 10]. 3.2. Mathematical Modelling and Roadmap The starting base of this work, realized as mentioned above in Siemens Software Process Simulate, is the kinematics of the human Jack. Rediscovering a new human model was not an option so it was decided to remodel the already existing human joints accordingly to the calculus results of the new parameters for the human joints [3, 15]. The guiding idea of calculating the new parameters of the human joints is presented in Fig. 7. Here is showed schematically showed the arm of the human body, composed from: shoulder joint (A), length of the upper arm (segment AB), elbow joint (B), length of the lower arm (segment BC), wrist joint (C), middle of segments AB and BC (points E and D). The actuated parts which support the arm weight are elbow and shoulder.

The two motors which move and sustain the Exoskeleton arm (as shown in Fig. 6) are placed in the positions A and B from Fig. 7. The power of these motors has to-be set up as they should sustain the whole weight of the human arm and Exoskeleton itself. The calculation method presented in the following paragraph, Equation 1 is an empirical one representing the required power for the shoulder motor. An exemplary calculus is presented in the following paragraph as well. A main assumption of the calculus is that on average the human arm weights about 5.3% of the total body weight [16, 17], reason for calculus parameters presented in Table 1.

P

MaxShould

M

Max

xω

(1)

The variables of the equation are: P = Power; M = Torque;

Z

= Angular velocity.

As demonstrated in the graphic from Fig. 8 the torque depends on the angle ( M ). The maximum torque is when M reaches 90°. The required torque for shoulder motor is presented in Equation 2. The velocity is predefined to Z = 1.5 rad/s. M

F x l x 9,8 x sinM

(2)

By substituting in this equation the variables from Fig. 7, the following formula is obtained: M

^ {[(m_AB) x 9,80 ] x l_AE} 

{[(m_B) x9.80]xl_AB}

(3)

{[(m_BC) x9,80]*(l_BC l_AB)} `xsinM

Finally, by replacing the variables with their value as in the table below is presented a final value for the maximum torque is obtained. Table 1. Values for variables of Equation 3 Values for variables of Equation 3 1kg = 9,80N m = mass M = angle between ߮_‫=ܤܣ‬0°….180° shoulder and arm ߮_‫=ܥܤ‬0° D = angular acceleration ݈_‫=ܤܣ‬0.3 ݉ ݈_‫=ܧܣ‬0.15 ݉ ݈_‫=ܥܤ‬0.33 ݉ P= Power ݈_‫=ܥܤ‬0.33 ݉ M=Torque ݈_‫=ܦܤ‬0.1 m Z = angular velocity ݉_‫=ܤܣ‬7 ݇݃ ؄70N ݉_‫=ܤ‬2 ݇݃ l = length ݉_‫=ܥܤ‬5 ݇݃ 

Fig. 7. Schematic representation of the human arm (©Robo-Mate)

After solving equation 2, with the above parameters the final obtained result for maximum torque is 33.71 Nm (Mmax=33,71Nm).

744

Carmen Constantinescu et al. / Procedia CIRP 41 (2016) 740 – 745

The final result of maximum power required calculated conform Equation 1 is as follow:

P MaxShould

33,71(Nm) x 1.5(rad/s)

50.56W

(4)

comprises only a part of the required features; for the rest of them in-depth work should be performed in the future. The current status of this work represents the definition of the new center of gravity of the JackEx, and the identification on how every different Exoskeleton module influence the position of gravity center. The current work is performed with preliminary versions of the Exoskeleton components. In the final modelling and simulation of JackEx in real manufacturing environment the work will be done with final version of the Exoskeleton. Fig. 9 shows the first results of the coupling, for a briefly understanding of how JackEx will look like.

Fig. 8. Generated moment based on angle between arm joints (©Robo-Mate)

First step was to determine where the Exoskeleton has a direct impact on the human mechanics and the corresponding constrains. Based on this the rescale, remodeling and redefining the mechanical parameters are realized [15, 18]. The first two steps are addressing directly the creation of an emulated exoskeleton in humanoid Jack, followed by attaching/coupling the exoskeleton modules to the body of human Jack. In this phase, Jack should be able to move in a perfect synchronization with the attached Exoskeleton modules and execute production tasks. Practically, the Exoskeleton is intimated coupled to the human Jack. The innovation in this step consists of creating a virtual point as the yellow ball in Fig. 9, giving the possibility to attach the Exoskeleton on to the human Jack [12]. The Classic Jack Software was used to work on detailed to the coupling activities. It offers few possibilities to enter into details when decomposing human Jack in its compound parts; working in deep details of subassemblies is not possible. The Exoskeleton weight has to-be distributed over the human body therefore new forces are applied over t-he humanoid. Forces have a variation depending on the human posture and lifted weight, which works under a mathematical algorithm. A big challenge regarding the coupling is to define the new gravity center for each JackEx instances, as presented in Fig. 5. 4. Conclusion and future work In order to embed JackEx to the simulation software e.g. Siemens Process Simulate, a suitable export format of the JackEx from Classic Jack Software should be found. The export format should be able to pack the human, the new algorithm and parameters [11]. Siemens Software portfolio works under universal import/export format called .jt. It

Fig. 9. First coupling results of Exoskeleton and Jack (©IAO)

The JackEx resource will play a crucial role in factories of the future, by keeping a high human employment rate and thus helping to reduce the company’s total cost for producing or disassembling complex products thus bringing a high level of flexibility in to the shop floor. Through the employment of 3D Model and virtual simulation, a suitable reconfiguration for the working space “as-is” situation could be validated. The workers’ health would be significantly improved thus reducing costs of the injuries at the workplace. The work presented in this paper is based on the modeling and simulation of the automotive industry assembly and disassembly use cases, where the ”as it should be” state of the production area by employing the Exoskeleton was realized by emulating the Exoskeleton.[4,19] The development of JackEx development is a work in progress, taking up the challenges of defining new algorithms to transform JackEx into a capable resource to analyze the human operator during different tasks and giving the right feedback for ergonomics analyses [9]. 5. Acknowledgements The authors acknowledge the teams of the European companies [1] from automotive assembly and disassembly sectors for their engagement in all phases of the performed work: from requirements analysis, to delivering required information and data, for giving valuable advice and feedback

Carmen Constantinescu et al. / Procedia CIRP 41 (2016) 740 – 745

and for approving correctness and appropriateness of the results. In the same time, the designers of the Exoskeleton delivered us a first idea on how the Exoskeleton modules, which will be finalized in the next periods. Last, but not least, many warmly thanks to Dr. Bernd Brinkmeier from Siemens for guiding and supporting us in developing very accurate digital models by using all complex functionalities of Siemens Process Simulate.

References: [1] Robo-Mate Project Factsheet. Avaible at: http://www.robomate.eu/sites/robo-mate.eu/files/downloads/factsheet_ robomate_projectdesign_neu_2013-11-25.pdf ; 2013. [2] Op De Beeck R, Hermans V. Research on work-related low back disorders. In: European Agency for Safety and Health at Work. Available at: https://osha.europa.eu/en/publications/reports/204; 2000. [3] Forge S, Blackman C. A Helping Hand for Europe: The Competitive Outlook for the EU Robotics Industry, Luxembourg. Available at: http://www.eurosfaire.prd.fr/7pc/doc/1290673085_eu_robotics_industry _jrc61539.pdf; 2010. [4] Linda C, Cagan S, Cagan J. Recursive Anneling: A Computational Model for Machine Design at: Research in Engineering Design (1995). Springer-Verlag London Limited; 1995. p. 102 – 125. [5] Constantinescu C, Landherr M, Neumann M. Aligning the Product, Factory and ICT Life Cycles: Challenges and Opportunities. In: Teti, Roberto (Chairman); Westkämper, Engelbert (International Program Committee). Innovative and Cognitive Production Technology and Systems 2012, 4 S. - URN urn:nbn:de:0011-n-2112752. [6] C. Constantinescu, P. C. Mureșan, S. M. Gînța, O. Todorovic, Modelling and simulation of advanced factory environments integrating intelligent exoskeleton, Proceedings of the 2014 International Conference on Production Research - Regional Conference Africa, Europe and the Middle East and 3rd International Conference on Quality and Innovation in Engineering and Management, Cluj-Napoca, 1-5 July 2014, ISBN 978-973-662-978-5, pp. 100-105 [7] Westkämper E. Digital Manufacturing In The Global Era. In: Digital Enterprise Technologies; 2007. p. 3–14.

[8] del Ama AJ, Koutsou AD, Moreno JC et. al. Review of hybrid exoskeletons to restore gait following spinal cord injury. In: Journal of Rehabilitation Research and Development, 49(4); 2012. p. 497–514. [9] PLM Automation Siemens, Process Simulate. Available at: http://www.plm.automation.siemens.com/en_us/products/tecnomatix/ass embly_planning/process_simulate/index.shtml; 2014. [10] Zoss A, Kazerooni H, Chu A. On the Mechanical Design of the Berkeley Lower Extremity Exoskeleton (BLEEX). In: IEEE/RSJ International Conference on Intelligent Robots and Systems; 2005. [11] De Sapio V, Warren J, Khatib O. Predicting reaching postures using a kinematically constrained shoulder model. In: Lenarcic J, Roth B. (Eds.), Advances in Robot Kinematics. Springer; 2006. p. 209-218. [12] PLM Automation Siemens, Classic Jack. Available at: http://www.plm.automation.siemens.com/en_us/products/tecnomatix/ass embly_planning/jack/classic_jack.shtml; 2014. [13] Khatib, O., Le Maitre, J.F., 1978. Dynamic control of manipulators operating in a complex environment. In: Proceedings RoManSy’78, 3rd CISM-IFToMM Symp. Theory and Practice of Robots and Manipulators, Udine, Italy, pp. 267–282. [14] Majstorovic V, Trajanovic M, Vitkovic N, Stojkovic M. Reverse engineering of human bones by using method of anotomic features. In: CIRP-ANNALS – Manufacturing Technology 62; 2014. p. 167-70. [15] Khatib O, Le Maitre JF. Dynamic control of manipulators operating in a complex environment. In: Proceedings RoManSy’78, 3rd CISMIFToMM Symp. Theory and Practice of Robots and Manipulators, Udine, Italy; 1978. p. 267–282. [16] Bogue R. Exoskeletons and robotic prosthetics: A review of recent developments. In: Industrial Robot: An International Journal, 36(5); 2009. p. 421–427 [17] Delp SL, Anderson FC, Arnold AS, Loan P, Habib A, John CT, Guendelman E, Thelan DG. OpenSim: open source software to create and analyse dynamic simulations of movement. In: IEEE Transactions on Biomedical Engineering 55; 2007. p. 1940–1950. [18] Dollar AM, Herr H. Lower Extremity Exoskeletons and Active Orthoses: Challenges and State-of-the-Art. In: IEEE Transactions on robotics, vol. 24, no. 1, February; 2008. [19] Delp, S.L., Anderson, F.C., Arnold, A.S., Loan, P., Habib, A., John, C.T., Guendelman, E., Thelan, D.G., 2007. OpenSim: open source software to create and analyse dynamic simulations of movement. IEEE Transactions on Biomedical Engineering 55, 1940–1950.

745