Virtual Verification of Human-Industrial Robot Collaboration in Truck Tyre Assembly

Transcription

1 Virtual Verification of Human-Industrial Robot Collaboration in Truck Tyre Assembly L. Hanson a, b, c, F. Ore a, d M. Wiktorsson d a Global Industrial Development, Scania CV AB, SE Södertälje, Sweden; b Product- and production development, Chalmers University of Technology, SE Göteborg, Sweden; c School of Engineering Science, University of Skövde, Skövde, SE , Sweden; d Mälardalen University, School of Innovation, Design and Engineering, SE Eskilstuna, Sweden; Human-industrial robot collaboration has been introduced as the ultimate combination for industry. The endurance and strength of a robot is combined with a human s flexibility, precision and quality skills. One challenge in the implementation of human-industrial robot collaboration is to create a safe working station for the operators, therefore most of the research focuses on these safety aspects. Industrial designers and engineers verify and optimise workstations in different simulation and visualisation tools in order to improve competitiveness, reduce late changes and reduce cost. Several robot tools and digital human modelling tools are available, but there are no or few simulation and visualisation tools that include both humans and robots. The aim of the proposed paper is to illustrate how unique software can be used to verify human-industrial robot collaboration. This software is a combination of the robot simulation tool IPS and the digital human modelling tool IMMA. The software demonstration is promising, covering the gap between digital human modelling tools and robot simulation tools. The simulation and visualisation tools generate pictures and animations, as well as quantified numbers to aid well-founded decision-making. The demonstration software was used to analyse a truck tyre assembly station. Fully manual, fully automated and human-industrial robot collaboration were compared. Practitioner Summary: The presented paper illustrates simulation and visualisation software for the virtual verification of Human - Industrial Robot collaboration. The software demonstration is a combination of the robot simulation tool IPS and the digital human modelling tool IMMA. The software demonstration is promising, covering the gap between digital human modelling tools and robot simulation tools. Keywords: ergonomics, digital human modelling, robot simulation, simulation and visualisation 1. Introduction Human - industrial robot collaboration has been introduced as the ultimate combination for industry (Tan et al. 2009). The endurance and strength of a robot is combined with a human s flexibility, precision and quality skills. Human-industrial robot collaboration is one possible solution for meeting increased global competition and upcoming demographic challenges. The truck industry has no in-depth experience of designing, verifying and optimising workstations where humans and robots collaborate. Traditionally, workstations have been either manual or automated by robots. To facilitate interaction between humans and industrial robots, the physical barriers surrounding the traditional industrial robot cells must be removed. The main prerequisite for doing this is solving personal safety issues in order to guarantee a safe working environment for the operators. Most human-industrial robot collaboration research focuses on these safety aspects, e.g. (Krüger et al., 2009), (Fischer and Henrich, 2009), (Morioka and Sakakibara, 2010), (Schmidt and Wand, 2012), (Zwicker and Reinhart, 2013), (Pedrocchi et al., 2013) and (Salmi et al. 2014). There is ongoing work at truck manufacturing companies to make human and robot workstations both ergonomic and productive. To be competitive, reduce late changes and reduce cost, these workstations are designed, verified and optimised using different simulation and visualisation tools, either a robot tool or a digital human modelling tool. Several robot tools exist; some are brand specific, such as ABB s RobotStudio, KUKA s KUKA.Sim (Vollmann, 2002) and Motoman s MotoSim, or there are generic software programs like Delmia (Brown, 2000), Robcad (Dong et al., 2007) and RoboSim (Lee and Elmaraghy, 1990). There are also 1

2 several digital human modelling tools on the market, such as AnyBody (Rasmussen et al., 2002), Jack (Badler et al., 1993), Ramsis (Seidl, 1997), Safework/DelmiaV5 (Fortin et al., 1990) and Santos (Abdel- Malek et al., 2006). There are only a few research papers available on the simulation of human-industrial robot collaboration. Luh and Srioon (2000) simulated a human hand placed on an object that was carried by a robot. The hand was then controlled through keyboard commands, and the robotic movement was stored. Busch et al. (2013) visualised a welding cell where the human does the welding and the robot holds the objects that are to be welded together. The truck industry needs experience in designing, verifying and optimising human-industrial robot collaboration workstations and there is a need for an effective simulation and visualisation tool that facilitates such work. The aim of the proposed paper is to illustrate how a unique software for verifying human-industrial robot collaboration is used in a study of a workstation where truck tyres are assembled. 2. Method A prototype human-industrial robot collaboration verification software was used (Ore et al., 2014). The tool is a combination of the robot tool IPS (Segeborn et al., 2011) and the digital human modelling tool IMMA (Hanson et al., 2014). IPS is a highly automated tool for the optimisation of robot cells, taking into account robot reach, robot collision and cycle time. IMMA is a highly automated tool with a manikin-finding collision free motion that is ergonomically friendly. A truck tyre assembly workstation was analysed to illustrate the software. In the workstation, tyres are assembled with 10 nuts. Three scenarios were compared: fully manual, fully automated and human- robot collaborative workstations. The weight of a tyre is maximum 130 kg and it has a diameter of 1.1 metres. Two parameters were studied in the comparison: ergonomic load and productivity. These parameters were identified as important by Kruger et al. (2009) and Oberer-Treitz et al. (2013). The virtual verification was performed using the following five steps. Import product, resources and layout robot model and optimise the layout. A CAD model of the layout and the tyre was imported. Furthermore the robot model was imported, an ABB IRB6620 with the following specifications: six degrees of freedom, a load capacity of 150 kg, and a reach of 2.2 m. Joint information was provided from the robot supplier s website. Manual layout optimisation was performed to locate the best robot position, making it possible for the robot to reach all the handling positions. Create manikin. Stature was selected as key anthropometric variable and 50 th percentile Swedish male. A single 50% percentile manikin from a Swedish anthropometric database (Hanson et al., 2009) was created. The manikin was 1792 mm tall. Define human and robot tasks. The tyre assembly workstation consists of the following three tasks: get tyre, get 10 nuts and nut runner as well as assemble tyre. The tasks can get categorised as: Get tyre - heavy work, get nuts and nut runner high precision, assemble tyre high precision. In the human-industrial robot collaboration, using human and robot strength, get tyre was performed by the robot and get nuts and get nut runner, as well as assemble tyre was done by the human. A hand over and a hand back task were introduced to the human-industrial robot collaboration. Due to the high load, a lifting device was used also in the fully manual workstation. Create simulations and export parameters. Collision-free paths were created for the human and the robot tasks in each of the three assembly scenarios. These human and robotic movements were then linked together to provide a complete assembly simulation. Ergonomic and productivity parameters were exported from the simulation. Analyse ergonomic and productivity parameters. Ergonomic load was quantified in time weighted RULA (McAtamney and Corlett, 1993) scores and productivity in operation time. The manual operation time was calculated by using SAM (Laring et al., 2002), a standard time measurement system based on MTM (Maynard et al., 1948). The robot time was extracted from the simulation. The robot speed used in the evaluation was reduced to 70% of the maximum speed. This reduction takes into consideration acceleration, retardation and lower speeds close to assembly positions. 2

3 3. Results The results showed that it was possible to simulate the tyre workstation as fully manual, fully automated and as a human-industrial robot collaboration scenario using the tool combination of IPS and IMMA, Figure 1. Figure 1. Illustration of the layout and the three scenarios tested: fully manual, fully automated and human-industrial robot collaboration workstations. Table 1 shows the productivity measured in time and the ergonomic load, the RULA score. Both workstations including humans, i.e. the fully manual and the human-industrial robot collaboration, are human friendly according to their RULA score. The human-industrial robot collaboration workstation is slightly more ergonomic than the manual workstation. The time measures show that the automated workstation is the fastest, the fully manual is the slowest and human-industrial robot collaboration is in between. Table 1. Ergonomic load (RULA score) and productivity, measured as time to perform an activity in three scenarios, fully manual (human), human-industrial robot collaboration (HIRC), and fully automated (robot). Human HIRC Robot Time [s] RULA [-] n.a. 4. Discussion The demonstration software presented in this paper is able to simulate and visualise fully manual, fully automated and human-robot collaborative workstations. The presented demonstration software is based on IPS, IMMA and matlab code, including several import/export change steps. This software has great potential. When the individual software programs are fully merged, the resulting demonstration software will probably be both effective and user friendly, because each included software program is considered to be effective (Tran, 2013) and user friendly (Hanson et al., 2014). The demonstration software covers a gap between digital human modelling tools and robot simulation tools. The demonstration software used is one of the first available human-industrial robot collaboration software programs for simulation and visualisation. There is a need for such tools, as both research and industry require optimisation, simulation and visualisation tools. The advantage of such a tool is that production concepts can be tested early in the design phase at a low cost. The simulation and visualisation tools generate pictures and animation, as well as quantified numbers that facilitate well-founded decision-making. All three workstation scenarios were human friendly and had good ergonomic scores; the RULA scores were 3.3 and 3 for fully manual and human-industrial robot collaboration respectively. The reduction was small, but shows the potential for decreasing stress on the human operator through designing humanindustrial robot collaboration assembly stations. Not surprisingly, the productivity measure showed that the full automated scenario was fastest, and the pure manual scenario was the slowest. The reason for this is simply that robot motions are faster than human motions. The human-industrial robot collaboration was in between. The reduction in operation time can, in practice, result in great improvements to production systems through-put; more than 50% higher using human-industrial robot collaboration instead of fully manual. It is important to emphasise that the aim of this paper is not necessarily to promote human-industrial 3

4 robot collaboration systems, but to show the evaluation potential provided by the software. This paper illustrates that both the ongoing software development for virtual human-industrial robot collaboration and its application are promising areas. The software will allow concepts to be evaluated early in the design phase. There are a number of open questions in a new production concept, for instance who should do what. In this tyre assembly task the robot did the heavy lifting and the human task required high precision. The simulated tyre work sequence was short and had few tasks, so in longer sequences there may be a need for support regarding how to allocate tasks. There may also be a need for optimising hand over and hand back control between the human and robot. Further research in virtual human-industrial robot collaboration is needed to keep the same rate of development. Tan et al, 2009 are convinced that human-industrial robot collaboration is a dream combination of human flexibility and machine efficiency. Acknowledgements The research has been conducted as a collaboration between two Swedish research projects: Collaborative Team of Man and Machine and Creation of Muscle Manikins. Both projects are funded by Sweden s innovation agency, Vinnova. References Abdel-Malek, K., Yang, J., Marler, T., Beck, S., Mathai, A., Zhou, X., Patrick, A. and Arora, J Towards a new generation of virtual humans. International Journal of Human Factors Modelling and Simulation, 1, Badler, N. I., Phillips, C. B. and Webber, B. L Simulating Humans: Computer Graphics, Animation, and Control. Oxford Univ. Press. Brown R. G Driving digital manufacturing to reality. Proceedings of IEEE Winter Simulation Conference, pp Busch, F., Wischniewski, S. and Deuse, J. Application of a character animation SDK to design ergonomic human-robotcollaboration. 2nd International Digital Human Modeling Symposium, 2013 Ann Arbor, United States of America. Chryssolouris, G., Hryssolouris, G., Mavrikios, D., Papakostas, N., Mourtzis, D., Michalos, G. and Georgoulias, K Digital manufacturing: history, perspectives, and outlook. Proceedings of the Institution of Mechanical Engineers, pp Dong, W., Hui, L. and Xiaoting, T Off-line programming of Spot-weld Robot for Car-body in White Based on Robcad. International Conference on Mechatronics & Automation, p Fischer, M. and Henrich, D D collision detection for industrial robots and unknown obstacles using multiple depth images. Advances in robotics research. Springer. Fortin, C., Gilbert, R., Beuter, A., Laurent, F., Schiettekatte, J., Carrier, R. and Decamplain, B SAFEWORK: a microcomputer-aided workstation design and analysis. New advances and future developments. Computer-aided ergonomics, Hanson, L., Högberg, D., Carlson, J. S., Bohlin, R., Brolin, E., Delfs, N., Mårdberg, P., Gustafsson, S., Keyvani, A., and Rhen, I.-M IMMA Intelligently moving manikins in automotive applications. Third International Summit on Human Simulation. Tokyo, Japan. Hanson, L., Sperling, L., Gard, G., Ipsen, S. and Vergara, C. O. (2009). Swedish anthropometrics for product and workplace design. Applied Ergonomics, 40 (4), pp Kruger, J., Lien, T. K. and Verl, A Cooperation of human and machines in assembly lines. CIRP Annals - Manufacturing Technology, 58, pp Laring, J., Kadefors, R., Örtegren, R. and Forsman, M MTM-based ergonomic workload analysis. International Journal of Industrial Ergonomics, 30, pp Lee, D. and Elmaraghy, W ROBOSIM: a CAD-based off-line programming and analysis system for robotic manipulators. Computer-aided engineering journal, 7, pp Maynard, H. B Methods-time measurement, McGraw-Hill. McAtamney, L. and Corlett, E. N RULA: A survey method for the investigation of work-related upper limb disorders. Applied Ergonomics, 24, pp Morioka, M. and Sakakibara, S A new cell production assembly system with human-robot cooperation. CIRP Annals - Manufacturing Technology, 59, pp Mourtzis, D., Papakostas, N., Mavrikios, D., Markris, S. and Alexopoulos, K The role of simulation in digital manufacturing: applications and outlook. International Journal of Computer Integrated Manufacturing. Oberer-Treitz, S., Dietz, T. and Verl, A Safety in industrial applications: From fixed fences to direct interaction. 44th International Symposium on Robotics. Ore, F., Hanson, L., Delfs, N. and Wiktorsson, M Virtual Evaluation and Optimisation of Industrial Human-Robot Cooperation: An Automotive Case Study. 3rd International Digital Human Modeling Symposium Pedrocchi, N., Vicentini, F., Matteo, M. and Tosatti, L. M Safe Human-Robot Cooperation in an Industrial 4

5 Environment. Int. J of Advanced Robotics System, Rasmussen, J., VonDrak, V., Damsgaard, M., DeZee, M., Christensen, S. T. and Dostal, Z The Anybody project - Computer analysis of the human body. Biomechanics of Man, pp Salmi, T., Väätäinen, O., Malm, T., Montonen, J. and Marstio, I Meeting New Challenges and Possibilities with Modern Robot Safety Technologies. Enabling Manufacturing Competitiveness and Economic Sustainability. Springer. Schmidt, B and Wand, L Active collision avoidance for human-robot collaborative manufacturing. Proceedings of the SPS12 conference, The Swedish Production Academy, pp Seidl, A., 1997 RAMSIS - A New CAD-Tool for Ergonomic Analysis of Vehicles Developed for the German Automotive Industry, SAE Technical Paper Segeborn, J., Segerdahl, D., Ekstedt, F., Carlson, J.S., Carlsson, A., Söderberg, R., A generlized method for weld load balancing in multi station sheet metal assembly lines. Proceedings of the ASME 2011 International Mechanical Engineering Congress & Exposition IMECE2011, November 11-17, Denver, Colorado, USA Tan, J. T. C., Duan, F., Zhan, Y., Watanabe, K., Kato R. & Arai, T Human-robot collaboration in cellular manufacturing: Design and development. International Conference on Intelligent Robots and Systems, pp Tran, D Optimizing welding cycle with the utilization of the IPS tool. Master thesis, Department of Engineering Sciences and Mathematics, Luleå University of Technology. Vollmann, K A new approach to robot simulation tools with parametric components. Industrial Technology, IEEE International Conference on, pp, Zwicker, C. and Reinhart, G Human-robot-collaboration system for a universal packaging cell for heavy electronic consumer goods. 5th International conference on Changeable, Agile, Reconfigurable and Virtual Production, pp