Our biggest event in the last two months was the preparation and realization of our exhibition an the Hannover fair (08.04. - 12.04.2013). We presented our enhanced FlexIRob system as it was evaluated in our user study with workers at Harting. In particular, we presented the fast configuration of the Kuka LWR IV to a confined workspace and our Assisted Gravition Compensation mode which allows to teach a trajectory assisted by the robot in avoiding collisions with the environment. Our booth was part of the trade-show presentation of the Spitzencluster "Intelligente Technische Systems OstWestfalenLippe (it's owl)" where we represented the activities of Bielefeld University, especially of the CoR-Lab and the CITEC, in this area.

The fair was a big success. Our system ran well through the whole week and we gathered a lot of positive feedback from the visitors of our booth. We thank Harting for sponsoring the nice obstacles for our presentation.  Several members of the local governments like the mayor of Bielefeld, Pit Clausen, Svenja Schulze, the minister for research and education of NRW, state secretary Helmut Dockter, and district president Marianne Thomann-Stahl have visited our exhibition. The picture below shows Svenja Schulze and Helmut Dockter at our booth. On Friday, we had the opportunity to present our system to three groups of pupils which visited us in the context of an information day for pupils interested in engineering oriented careers. I have attached a report in the local newspaper as deliverable about our dessimination activity at the Hannover fair.

In the last two months we have concentrated on the integration of a 3D Vision sensor into the FlexIRob scenario and the implementation of a detection of dynamic obstacles in the workspace of the robot arm.

First, a calibration of the sensor (here, Kinect camera) and the robot arm is realized by using the end effector position as a source for corresponding points between the two coordinate systems from which the transformation from the camera's coordinate system to the robot's coordinate system can be computed using Singular Value Decomposition. By simply ensuring through manual maneuvering of the robot arm that the end-effector is the closest point in the camera, the detection of the end-effector in the 3D sensor can be realized as simple nearest point detection. The advantage of this approach is that it is independent from knowing the appearance of the concrete end-effector.

Second, the FlexIRob system detects the robot arm in the visual sensor by estimating the arm through four cylinders which closely envelope the segments of the arm and the end-effector, transforming these cylinders into the camera's coordinate system, and excluding all 3D points within these four cylinders from the raw scene.

The third component of our perception subsystem focuses on the separation of the scene layer into static obstacles which are already encoded in the redundancy resolution of the robot arm and dynamic obstacles which appear spontaneously and should be avoided through an appropriate strategy during normal task completion. We realized this separation by integrating the Articulated Scene Model approach.

Last, the 3D points being part of the dynamic layer are clustered into connected components and represented as 3D boxes.

The intermediate results of all steps described above are depicted in the picture below.



Further, we are pleased to announce that our paper "Assisted Gravity Compensation to Cope with the Complexity of Kinesthetic Teaching on Redundant Robots" was accepted for publication at the International Conference on Robotics and Automation 2013. And Sebastian Wrede will give on the 5th of March a talk at the HRI2013 in Tokyo on our journal publication "A User Study on Kinesthetic Teaching of Redundant Robots in Task and Configuration Space". Interested people can also visit us at our stand on the Hannovermesse (08.04.-12.04.2013).

The activities of the last two months are two-folded. We further consolidated our results from the user study at Harting and proceeded with the integration of perception in the FlexIRob-Setup.

Our submission "A User Study on Kinesthetic Teaching and Learning for Efficient Reconfiguration of Redundant Robot" is now finally accepted for publication in the Journal of Human Robot Interaction. With this publication we also get the opportunity to present our results at the HRI 2013 conference in Tokyo. To summarize the achieved results, with this publication we evaluated with workers from our industrial partner the esisting FlexIRob setup and an enhanced version of it where we introduced a new controller which supports the user in a teach-in task by keeping track of the task-independent environmental constraints learned in a previous configuration step.

As dissemination activities, we have presented the results of the user study to the workers which have participated in the study. Further, the idea of the FlexIRob system and the conducted user study have been introduced to Harting's costumers through an article in the tec.News (issue 23, p.16).

In addition to the above points, we have implemented in the last two months the perception chain which applies the Articulated Scene Model approach to co-worker scenarios. We have defined middleware specific data formats (here Robotic Service Bus and Robotics Systems Types) to provide the captured Kinect data to all components of our system (e.g., the simulation or the robot arm) and implemented the new components within the system framework. Currently, we are working on the integration of our dynamic obstacle avoidance concept on the FlexIRob demonstrator, the Kuka LWR IV, aiming at a first calibration of the robot arm and the camera(s) and the visualization of dynamic obstacles in the simulation of the robot arm and its environment as a first test.

As part of our dissemination activities we organized the Workshop “Technology Monitoring: Robotics today and tomorrow – Technology towards Assistive Automation”. It was part of the OWL MASCHINENBAU Academy. OWL MASCHINENBAU is a network of innovation targeting to strengthen the economic and technological power of the regional industry in production technology. It is an association of local industry, SME, and research institutions of the region East-Westphalia. It targets to transfer research to application in industry, to initiate cooperative networks, and to organize advanced training.

Our workshop focused on challenges arising from the dramatic changes of technology in automation and particular robotics especially with regard to close collaboration of humans and machines. This collaboration has big potential to improve manufacturing work flow strengthening the global competitiveness and sustaining the local value chain. The workshop approached the topic “interactive robotics” from two sides: it introduced future technologies (redundancies, force control, innovative controls) enabling human-robot interaction and showed methods for interactive configuration and safe operation of such systems. The goal of the workshop was to give an outlook on technology which might be introduced in the next years into the automation and production field and will influence future production systems. Questions concerning safety, acceptance, and the role of the Spitzencluster it’s owl2 (Intelligente Technische Systeme OstWestfalenLippe) for the local area completed the workshop. The main presenters were Dr. P. Pfaff from KUKA Laboratories, Dr. M. Ruskowski from Carl Cloos Schweißtechnik, Dr. F. R ̈thling from Robert Bosch GmbH, Prof. Dr. A. Schneider from FH Bielefeld, and Dr. A. Swadzba and Prof. Dr. J. Steil from Research Institute for Cognition and Robotics (CoR-Lab). The workshop ended with hand-on demonstrations in the CoR-Lab.

Study results

In the last two months our main focus was on analyzing the data that we collected with our FlexIRob system during the user study on physical human robot interaction at Harting. We were able to evaluate the questionnaire and got first insights about the perception of our system by potential co-workers.

Descriptive evaluation

We had 45 users, 16 male and 29 female, belonging to the working units 'cage assembly', 'mechanical processing', 'die casting', 'internal part production', 'machine assembly', and 'prefabrication'. Most of them were working for 1 to 33 years at Harting (M = 11.36 years, SD = 9.78 years, Min = 1, Max = 33). The participants' mother tongue distributes over German (77,8%), Russian (11,1%), Turkish (6,7%) and Greek(2,2 %). Concerning their highest level of education, almost a half of the participants declared the receipt of a certificate of secondary education (CSE) or a general CSE (Haupt- oder Realschulabschluss), ca. 13% received a vocational or university-entrance diploma, and the remaining participants finished a vocational education (Berufsausbildung).

Answering the questions raised

As reported earlier, we conducted the study to answer questions concerning the perception and manageability of the FlexIRob system. In particular: Is such a physical MMI actually intuitive? For all users? Is it comfortable? How exhausting is the work with the robot arm?  Is the reconguration of the robot practicable? How well and how fast can it be done?  What are the characteristics of the training data provided by the naive users of the system?

The results are very convincing. First of all, the users felt not threatened by the robot and perceived it to be reliable, which is an important fact regarding the applicability in future industrial co-worker scenarios. Concerning the physical interaction, they stated that the robot was easy and comfortable to handle as well as the operation of the robot to be self-explanatory. Additionally, they felt the system to give helpful feedback during the collaboration. 

Evaluation concerning assisted vs. not-assisted wire-loop game

As reported earlier, we divided the participants into two groups during the wire-loop game: One group (group A) was assisted by the robot that respects the constraints of the environment and the other one (group N) not. Each joint of the robot arm needs to be controlled manually. Our motivation was to evaluate the new control mode, i.e. whether solving the wire-loop game was easier for participants of group A or not. Again, the results are promising. First of all, we can report that there is a significant difference in the perceived simplicity of operating the robot arm. The participants of group A appraised the operation easier than the ones from group N. Second, group A declared the settings of the robot to be significant more appropriate than group N. Finally, we could detect a marginal significant effect on the handling of the robot arm during the interaction. Participants of group A felt the handling more comfortable than the ones from group N.

A first analysis of the trajectories taught-in during the wire-loop game shows differences in the duration and the smoothness of the trajectories between the two groups. The assisted group is significant faster in teaching a trajectory than the not-assisted group. Group A also produces much smoother and more comparable trajectories than group N.