In this paper, we present a mechanical hand-tracking system with tactile feedback designed for fine manipulation in teleoperation scenarios. Alternative tracking methods based on artificial vision and data gloves have become an asset for virtual reality interaction. Yet, occlusions, lack of precision, and the absence of effective haptic feedback beyond vibrotactile still appear as a limit for teleoperation applications. In this work, we propose a methodology to design a linkage mechanism for hand pose tracking purposes, preserving complete finger mobility. Presentation of the method is followed by design and implementation of a working prototype, and by evaluation of the tracking accuracy using optical markers. Moreover, a teleoperation experiment involving a dexterous robotic arm and hand was proposed to ten participants. It investigated the effectiveness and repeatability of the hand tracking with combined haptic feedback during a proposed pick and place manipulation tasks.

Robotic avatar technology has the potential to impact the future of human connectivity, transporting the sense of a human’s presence to a remote location anywhere and in real-time. In this regard, the recent ANA Avatar XPRIZE challenge fostered the development of the ultimate generation of non-autonomous robotic avatar designs. This paper is devoted to introducing our system proposal, allowing intuitive motion-based control and multi-modal feedback. The teleoperated robot endorses an anthropomorphic upper body with dual arms and dexterous hands for fine manipulation of even small objects, as well as an omnidirectional platform for improved and safe locomotion. Special focus is pointed at the crucial challenge of providing self-body perception to the operator, in particular regarding control of the arms and rendering of haptic sensations. To this end, we propose an upper-limb exoskeleton and a teleoperation architecture allowing retargeting of the operator’s skills and receiving tactile and kinesthetic force feedback with realistic and informative perception of the own’s arms. Lastly, the platform was validated during laboratory and challenge missions mimicking several social, cooperative, and cultural scenarios from which we report on the lesson learned and future improvements.

Wearable haptic devices are used to render sense of touch in different virtual reality, simulated or teleoperated environments. Design of these devices has to comply with dimensional constraints imposed by ergonomics and usability, introducing compromises between size of the actuators and features of the rendered haptic stimuli, such as peak forces and bandwidth. We propose a new design approach for wearable haptic devices resembling the micro-macro actuation concept: it is based on two distinct actuators with different mechanical reduction, coupled by an elastic element. We show this better matches the output range of the actuators to features of the signals used in typical haptic rendering. We investigated the approach through an analytical model, a numerical model, and physical experiments conducted after design and development of a working prototype. The theoretical and simulated models allow to better understand dynamic interaction of the system parts, and to define design guidelines for the development of the real device. The adopted design solutions were implemented and evaluated in the prototype of a highly wearable fingertip device. Final experimental results show how the implementation of the proposed method is capable of an effective haptic rendering, that better matches the desired frequency response.

Deep transfer learning (DTL) allows for the efficient building of intelligent fault diagnosis systems (IFDS). On the other hand, DTL methods still heavily rely on large amounts of labelled data. Obtaining such an amount of data can be challenging when dealing with machines or structures faults. This document proposes a novel approach to the design of vibration-based IFDS using DTL in condition of strong data scarcity. A periodic multi-excitation level procedure leveraging intrinsic non-linearities of real-world systems is used to produce images that can be conveniently analysed by pre-trained Convolutional Neural Networks to diagnose faults. A new data visualization method and its augmentation technique are proposed in this paper to tackle the typical lack of data encountered during the design of IFDS. Experimental validation on a railway pantograph structure provides effective support for the proposed method.

Functional verification of railway pantographs is performed within periodic maintenance programs by means of specifically designed automation and robotic devices that can check their structural integrity and correct functionality. In this paper, we present the design and validation of a new portable inspection robotic device that through structural dynamic excitation and passive movement can assess the health status of railway pantographs. The device is endowed with a new actuation system that combines the large range of force attained through a macro-actuator with the high-frequency capabilities of a micro-actuator while preserving lightweight structure. The reported design and experiments confirm that excitation transmission by accurate force control can be achieved in the entire frequency range, despite the interaction between actuator and structure, and that simulated defects can be revealed by low and high-frequency alterations.

Periodic assessment of railway pantograph is essential for the safety of railways. Nevertheless, actual devices can only perform a rough analysis of suspensions and no clue on the structural integrity is provided. In this paper, we introduce a portable robotic device that achieves the goal through two kinds of tests. Laboratory experiments show how the preliminary quasi-static test can assess the suspension in terms of uplift force, joints friction and damping. The diagnosis capability of the dynamic test is ensured by simulating both a suspension and a structural fault. Damages evidence appear even in the low-frequency band, below 5 Hz. Results encourage further studies to introduce a new class of maintenance routines based on the present device.