Gearing up and accelerating cross‐fertilization between academic and industrial robotics research in Europe:

This paper presents the application of a novel projection-based safety system for ensuring hard safety in human-robot collaboration. We adapted the proposed sensor system to incorporate the joint positions of a collaborative robot, thus offering the opportunity to establish minimal and well-shaped safety spaces around the robot at any time. In this contribution we explain in detail main challenges and their solutions for generating and monitoring such safety spaces. Furthermore, we build up a collaborative workplace and evaluate the sensor system concerning its behavior and detection capabilities under operational conditions.

This paper takes an important step in characterizing a novel field of architectural research where a robotic system moves on a construction site and positions building components in-situ. Developed by the research group of Gramazio & Kohler at ETH Zurich, this approach offers unique advantages over traditional building technology: it fosters non-standard building processes, it can be directly applied on the construction site and it is easily scalable and it offers digital integration and informational oversight across the entire design and building process. Featuring a comprehensive new approach to architecture and technology, this paper considers 1) research parameters and components of in-situ robotic fabrication (such as tolerance handling, man-machine cooperation and localisation), 2) experimentation and building prototypes at full architectural scale, and 3) the architectural implications of integrating these findings into a systemic, unifying process at the earliest stages of design. As a result, in-situ robotic fabrication opens up entirely new possibilities of automated construction that are not limited by the constraints of prefabrication; the most evident and radical consequences of in-situ robotic fabrication are the ability to digitally oversee and control a large number of aspects of design and fabrication within an efficient and flexible building process.

In this chapter we illustrate the main results of the TRAFCON experiment. We consider a real AGV based automatic warehouse system and we model the traffic control problem exploiting coordination diagrams and taking into account all the constraints holding in real plants. We propose a novel traffic manager that, besides efficiently controlling the coordinated motion of the AGVs, can dynamically change the paths the robots are following. The TRAFCON traffic manager is experimentally validated on simulated real plants and on a small-scale automatic warehouse.

The presented work approaches programming of redundant robots such as the KUKA Lightweight Robot IV in a co-worker scenario from a user-centered point of view. It specifically asks, how the user’s implicit knowledge about the scene and the task can be transferred effectively to the robot through kinesthetic teaching. It proposes a new method to visualize the implicit scene model conveyed by the user when teaching a respective inverse kinematics and measures generalization by the robot. Based on these insights and empirical results from a previously performed user study, the present study argues that physical guidance of a task in confined spaces with static obstacles is too difficult to achieve in a single interaction. Summarizing earlier results and putting them into context, it is shown how to assist users to remedy this issue. The key is to divide the process in an explicit configuration phase for teaching the implicit scene model and a subsequent already assisted programming phase to teach the task based on a particular assisted gravity compensation mode. Further results from the user study confirm that this renders kinesthetic teaching in confined spaces feasible and enables a flexible and fast reconfiguration of the robot.

Grasping individual objects from an unordered pile in a box has been investigated in stationary scenarios so far. In this work, we present a complete system including active object perception and grasp planning for bin picking with a mobile robot. At the core of our approach is an efficient representation of objects as compounds of simple shape and contour primitives. This representation is used for both robust object perception and efficient grasp planning. For being able to manipulate previously unknown objects, we learn object models from single scans in an offline phase. During operation, objects are detected in the scene using a particularly robust probabilistic graph matching. To cope with severe occlusions we employ active perception considering not only previously unseen volume but also outcomes of primitive and object detection. The combination of shape and contour primitives makes our object perception approach particularly robust even in the presence of noise, occlusions, and missing information. For grasp planning, we efficiently pre-compute possible grasps directly on the learned object models. During operation, grasps and arm motions are planned in an efficient local multiresolution height map. All components are integrated and evaluated in a bin picking and part delivery task.

This paper presents work on automatic grasp generation and grasp learning for reducing the manual setup time and increase grasp success rates within bin-picking applications. We propose an approach that is able to generate good grasps automatically using a dynamic grasp simulator, a newly developed robust grasp quality measure and post-processing methods. In addition we present an offline learning approach that is able to adjust grasp priorities based on prior performance. We show, on two real world platforms, that one can replace manual grasp selection by our automatic grasp selection process and achieve comparable results and that our learning approach can improve system performance significantly. Automatic bin-picking is an important industrial process that can lead to significant savings and potentially keep production in countries with high labour cost rather than outsourcing it. The presented work allows to minimize cycle time as well as setup cost, which are essential factors in automatic bin-picking. It therefore leads to a wider applicability of bin-picking in industry.

In this paper we introduce an online object manipulation system for the NAO robot that is able to detect and grasp an object out of a human hand and then give it back in real-time. Known objects are rendered from 3D models and detected stereo contour-based by using a new stereo vision head for NAO. In order to grasp objects, motion trajectories are generated by an A* planner while avoiding obstacles. In order to safely release objects back into a human hand, a combination of tactile and force sensors of the carrying arm is used to detect whether someone touched the grasped object. We performed quantitative experiments in order to evaluate the quality of the detector, the time to grasp an object, as well as the number of successful grasps. We demonstrated the whole system on the real robot.

The scientific goal of HANDS.DVI consists of developing a common framework to programming robotic hands independently from their kinematics, mechanical construction, and sensor equipment complexity. Recent results on the organization of the human hand in grasping and manipulation are the inspiration for this experiment. The reduced set of parameters that we effectively use to control our hands is known in the literature as the set of synergies. The synergistic organization of the human hand is the theoretical foundation of the innovative approach to design a unified framework for robotic hands control. Theoretical tools have been studied to design a suitable mapping function of the control action (decomposed in its elemental action) from a human hand model domain onto the articulated robotic hand co-domain. The developed control framework has been applied on an experimental set up consisting of two robotic hands with dissimilar kinematics grasping an object instrumented with force sensors.

This paper presents for the first time an application study of using dexterous robotic hands for deboning operation so as to establish a human-robot co-working platform for cutting, deboning and muscle extraction operation in meat industry. By setting up a test rig consisting of a support and a customized knife integrated with force sensors and utilizing a modified data glove, manual ham deboning operations are carried out providing essential information and background for the robotic hand design, appropriate force/torque and position sensors identification, and human-robot co-working platform trajectory planning. Principle component analysis method is then employed for trajectory mapping and planning associated with the knife peak coordinates, and concept of force cone is introduced leading to an efficient algorithm for trajectory planning. Further, design and kinematics of a metamorphic hand are investigated laying a background for measuring manipulation and grasp quality of the proposed robotic hand. The above experimental, theoretical, hardware and software preparations finally lead to the applications of using two dexterous robotic hands, i.e. one Shadow C6M left hand and one KCL G4 metamorphic hand to replace human left hand in deboning operation. The experiment thus laid background work for the robotization of meat industry and gave insight into the benchmarking of utilizing dexterous hand in deboning operation constructing a human-robot co-working hyper-flexible cell.

Body Extenders (BE) are an emerging class of wearable robots, aiming at physically supporting humans during the complex handling of materials in un-structured environments. In the framework of the TESBE experiment of the European RTD project ECHORD, PERCRO and Telerobot successfully developed three core technologies deemed as enabling to make safer and more efficient the use of BE in the envisaged application scenarios. In particular the newly developed force control has allowed to reduce by a factor of 5-7 times the resistance forces exerted by the device on the operator’s body, during the tracking of its movements, the collaborative control of the BE posture has demonstrated its capability to prevent the overturning of the system under the action of gravity, when approaching its equilibrium boundary, and the new haptically enhanced gripper allowed to speed up the grasping of objects having different shapes and sizes, thanks to its original highly under-actuated mechanism that automatically adapts the orientations/positions of its multiple grasping surfaces.

This work describes the ECHORD Experiment ASTROMOBILE, a project aimed to design, develop and test a system for favourable independent living, improved quality of life and efficiency of care for senior citizens in domestic environments. The system, composed of a mobile robotic platform (called ASTRO) and an Ambient Intelligent Infrastructure that actively cooperated between them and with the end-user, was designed and implemented with a user-centred design approach, involving different stakeholders. The system was designed to deliver services to users, like drug delivery, stand support, reminding, info-entertainment. The design took advantages of the integration of robotic platforms with smart environments, to provide to users higher quality and localization based services. Senior end-users were involved in the experimentation of the system in the DomoCasa Living Lab and feedbacks were gathered for the technology assessment. Particularly, this paper demonstrates the general feasibility of the ASTROMOBILE system and thanks to users feedbacks its acceptability and usability.

This chapter presents an intuitive laser-based teleoperation scheme to enable the safe operation of a multirotor UAV by an untrained user in a cluttered environment using a haptic joystick. An obstacle avoidance strategy is designed and implemented to autonomously modify the position setpoint of the UAV if necessary. This scheme includes a novel force-feedback algorithm to enable the user to feel surrounding environment of the UAV as well as the disturbances acting on it. The stability analysis of the whole teleoperation loop, including the nonlinear dynamics of both UAV and joystick, is provided. The implementation of the teleoperation scheme on the Flybox hexacopter platform by the company Skybotix is described. Finally, experimental results and videos are reported to demonstrate the successful implementation and the performance of the overall system.

This chapter intends to provide a description of the MAAT experiment in the framework of the ECHORD European project. The experiment is aimed at developing a novel robotic system for upper-limb rehabilitation, capable of maximizing patient motivation and involvement in the therapy and performing a continuous assessment of the progress of the patient recovery in a multimodal way. The key-issue of the MAAT approach is to include the patient in the control loop by means of multimodal patient data (biomechanical as well as physiological data) and an immersive virtual reality system. To this purpose, a bio-cooperative controller is developed that incorporates multimodal data and adaptively and dynamically change the complexity of the therapy and of the virtual environment in accordance with specific patient requirements and abilities. Two MAAT robotic platforms have been developed for the experimental validation of the proposed approach. They consist of the same multimodal interface and differ in the used robotic arm in charge of delivering the therapy. Preliminary experimental data on healthy subjects are reported in this chapter. The application to stroke patients is envisaged.