In robotics, Learning from Demonstration (LfD) aims to transfer skills to robots by using multiple demonstrations of the same task. These demonstrations are recorded and processed to extract a consistent skill representation. This process typically requires temporal alignment through techniques such as Dynamic Time Warping (DTW). In this pa per, we consider a novel algorithm, named Spatial Sampling (SS), specifically designed for robot trajectories, that enables time-independent alignment of the trajectories by providing an arc-length parametrization of the signals. This approach eliminates the need for temporal alignment, enhancing the accuracy and robustness of skill representation, especially when recorded movements are subject to intermittent motions or extremely variable speeds, a common characteristic of operations based on kinesthetic teaching, where the operator may encounter difficulties in guiding the end-effector smoothly. To prove this, we built a custom publicly available dataset of robot recordings to test real-world movements, where the user tracks the same geometric path multiple times, with motion laws that vary greatly and are subject to starting and stopping. The SS demonstrates better performances against state-of-the-art algorithms in terms of (i) trajectory synchronization and (ii) quality of the extracted skill.

The interest in Physical Human-Robot Interaction (pHRI) has significantly increased over the last two decades, thanks to the availability of collaborative robots that ensure user safety during force exchanges. For this reason, stability concerns remain a key issue in the literature whenever new control schemes for pHRI applications are proposed. Due to the nonlinear nature of robots, stability analyses generally rely on passivity concepts and consider ideal models of robot manipulators. Therefore, the first objective of this paper is to conduct a detailed analysis of the sources of instability in proxy-based constrained admittance controllers for pHRI applications, taking into account parasitic effects such as transmission elasticity, motor velocity saturation, and actuation delay. The second objective of this paper is to perform a sensitivity analysis, supported by experimental results, to identify how the control parameters affect the stability of the overall system. Finally, the third objective is to propose an adaptation technique for the proxy parameters, with the goal of maximizing transparency in pHRI. The proposed adaptation method is validated through simulations and experimental tests.

Trajectory learning is one of the key components of robot Programming by Demonstration approaches, which in many cases, especially in industrial practice, aim at defining complex manipulation patterns. In order to enhance these methods, which are generally based on a physical interaction between the user and the robot, guided along the desired path, an additional input channel is considered in this article. The hand stiffness, that the operator continuously modulates during the demonstration, is estimated from the forearm surface electromyography and translated into a request for a higher or lower accuracy level. Then, a constrained optimization problem is built (and solved) in the framework of smoothing B-splines to obtain a minimum curvature trajectory approximating, in this manner, the taught path within the precision imposed by the user. Experimental tests in different applicative scenarios, involving both position and orientation, prove the benefits of the proposed approach in terms of the intuitiveness of the programming procedure for the human operator and characteristics of the final motion.

In this paper, the problem of suppressing sloshing dynamics in liquid handling robotic systems has been faced by designing a dynamic filter that starting from the desired motion of the liquid container calculates the complete position/orientation trajectory for the robot end-effector. Specifically, a design philosophy mixing a filtering technique that suppresses the frequency contributions of the reference motion that may cause liquid oscillations and an active compensation of lateral accelerations by a proper container re-orientation has been adopted. In principle, the latter contribution requires the knowledge of acceleration of the reference trajectory, but because of the use of an harmonic smoother that performs a shaping of the original motion, it is possible to obtain the value of the acceleration in runtime. In this way, the proposed methods can be applied also to reference motions that are not known in advance, e.g. commands directly provided by a human operator. This possibility has been demonstrated by means of a number of experimental tests in which the user teleoperates the robot carrying the container with the liquid by simply moving in the free space its hand, whose 3D position is detected by a motion capture system.

The gap between human waiters and state-of-the-art robot systems that try to serve something to drink is often embarrassing, with the former able to manipulate glasses and trays or glasses on trays with incredible dexterity and the latter that move at incredible slowness. In this video, we want to show that robots can do it better by moving a bottle or a tankard full of beer that are simply placed on a flat steel plate connected the flange of a robot manipulator. The robot tracks the trajectory defined by a human operator that moves its hand in the 3D space, with a motion capture system that acquires in real time the position. A feed-forward controller, placed between the user and the robot and based on the combination of a smoother and proper orientation compensation, counteracts the lateral accelerations and suppress sloshing phenomena of the liquids. Eventually a camera mounted on the robot arm provides a visual feedback to the operator with monitoring purposes. The challenge for the operator was to drop the carried object. will the feed-forward control be robust enough to avoid this event, even at high speed? Watch the video and find out!

This paper presents a new feedforward controller based on a continuous-time finite impulse response filter, designed to minimize the vibrations that usually affect robot manipulators with elastic joints. In particular, Variable Stiffness Joints (VSJ) robots are considered, since they are usually characterized by a very low level of damping which makes the problem of the oscillations quite important. The proposed approach allows to simplify the overall control structure of VSJ robots, which is based on a decentralized control of each servomotor, imposing the desired position and the desired stiffness at each joint, and on a novel feedforward control, filtering the reference signals. After analyzing some of the filter properties and the method for the parameters choice, experimental results on a VSJ robot demonstrate the importance of the proposed filtering action for minimizing vibrations and oscillations.

This paper presents a feed-forward approach to reduce sloshing dynamics in liquid handling robotic systems. According to our solution, the dynamics of a liquid into an open vessel manipulated by a robot can be described by means of a spherical pendulum mechanical model. By doing this, the sloshing problem can be addressed as a vibration suppression problem for a second order system. More in details, the pendulum model is utilized to tune an exponential filter which shapes the reference trajectory for the robot, thus achieving a sloshing-free motion of the liquid inside the vessel.

In this paper, a novel repetitive control scheme is presented and discussed. The general framework is the control of repetitive tasks of robotic systems or, more in general, of automatic machines. The key idea of the proposed scheme consists in modifying the reference trajectory provided to the plant in order to compensate for external loads or unmodelled dynamics that cyclically affect it. By exploiting the dynamic filters for the B-spline trajectory planning, it has been possible to integrate the trajectory generation within a repetitive control scheme able to modify in real-time the reference signal with the aims of nullify interpolation errors. Experimental results obtained controlling two joints of a standard industrial manipulator are reported, showing the effectiveness of the proposed method.

In this paper, the advantages of CACSD (Computer Aided Control System Design) tools for integrating a robotic system in a packaging machine are illustrated. Beside the mechanical integration of the robot into the machine architecture, it is necessary a functional integration, that requires a precise synchronization with the other parts of the system. In the proposed application, a robot with a parallel kinematics is used for pick-and-place tasks between two conveyor belts. It is therefore necessary a proper motion planning which allows to synchronize the grasp and release phases with the conveyor belts, avoiding obstacles and guaranteeing the compliance with bounds on velocity, acceleration and limits in the workspace. A trajectory composed by quintic polynomials has been considered and a specific tool has been designed in the Matlab environment, which allows to modify the parameters of the trajectory and to analyze the obtained motion profiles from both the kinematic and dynamic point of view.

In this paper, a novel technique for online generating trajectories in the 3-D space is presented. The trajectory planner is based on cubic B-splines. However, while the definition of B-splines requires the solution of a global problem that involves the entire set of via-points to be interpolated/approximated, and therefore it is not suitable for online implementation, the proposed generator is able to approximate spline functions with the prescribed precision on the basis of local computations, which only need the knowledge of a very limited number of via-points. FIR filters are the foundation of this result. As a matter of fact the planner is composed by a first FIR filter for the computation of the control points from the sequence of desired via-points, followed by a chain of moving average filters. Therefore, the generator combines the characteristics of B-spline trajectories (smoothness and minimum curvature) and those of FIR filters (simple structure and computational efficiency). Moreover, besides standard cubic curves, the so-called smoothing B-splines have been considered for online trajectory generation. This allows to find a tradeoff between the possibility of exactly crossing the given via-points and the smoothness of the resulting trajectory. A simple teleoperation task with a Puma 560 industrial manipulator has been arranged for experimentally validating the proposed method.

Joint stiffness plays an important role in both safety and control performance, particularly in human-friendly robots using artificial pneumatic muscles. Due to the limited control bandwidth of pneumatic muscles, stiffness characteristics and their effects on safety in the frequency domain should be taken into account. This paper introduces the concept of instantaneous stiffness and validates its model with the Stanford Safety Robot (S2ρ. The potential effects of instantaneous stiffness on safety is explored through experimental comparison of peak impact accelerations under various impact conditions. Instantaneous stiffness demonstrates different effects on the impact acceleration depending on impact velocity and controller gain. Finally, the paper discusses the stiffness characteristics as a guideline for design and control to improve the robot safety while maintaining the control performance.

In this paper, the relation between B-splines and FIR (Finite Impulse Response) filters is demonstrated and exploited to design a digital filter for trajectory planning, combining the very simple structure and computational efficiency of FIR filters with the flexibility of splines. In particular, the trajectory generator consists of two main elements. The former is devoted to the solution of an optimization problem that, given a set of points to be interpolated (or approximated), provides the control points defining the spline. The latter, a cascade of moving average filters, gives the trajectory profile at each sampling time on the basis of such points. The proposed method has been applied to several robotic and industrial applications, and in this paper two case studies are reported as examples: an industrial robot performing a welding operation and a mobile robot moving in an environment with obstacles. With respect to these tasks, the main features of the trajectory generator are shown: the possibility of planning trajectories with high degree of smoothness (continuity of the derivatives), the possibility of easily changing the duration of the trajectory (and therefore the velocity, acceleration, jerk, etc. of the trajectory) maintaining the same geometric path, the possibility of locally modifying the pre-planned path.

The first part of this paper describes the development of a humanoid robot hand based on an endoskeleton made of rigid links connected with elastic hinges, actuated by sheath routed tendons and covered by continuous compliant pulps. The project is called UB Hand 3 (University of Bologna Hand, 3rd version) and aims to reduce the mechanical complexity of robotic end effectors yet maintaining full anthropomorphic aspect and a good level of dexterity. In the second part this paper focuses on the early experiences of the UB Hand 3 in performing manipulation tasks.

In this work the static and dynamic characterization of viscoelastic pads for robotic hands is performed. A quasi-linear model, developed to describe the behavior of human hand pads and, more generally, of biological tissues, is adopted in order to overcome the problems tied to classical (linear) models, often used in the robotic field. Through experimental tests, the values of the parameters of this model have been found for two different materials (a polyurethane gel and a silicon rubber) which show a behavior similar to that of human pads and seem very suitable for robotic applications. Finally, the model has been extended with the use of digital filters, making the identification process very straightforward.

The paper describes work in progress at the University of Bologna concerning the design of a new anthropomorphic robot hand. The hand is based on the modular assembly of articulated fingers that adopt an original configuration of their structure, made with rigid links connected by elastic hinges that are coaxially crossed by flexible tendons. This innovative design is suitable to host distributed sensory equipment and continuous compliant cover, allowing a high level of anthropomorphism together with great structural simplification, reliability enhancement and cost reduction. Furthermore, the proposed solution is very flexible, as it can be adapted to many different hand configurations and is not dependent on a particular type of actuation, being compatible with future availability of any kind of artificial muscles.

In this paper, a novel design approach for the development of robot hands is presented. This approach, that can be considered alternative to the "classical" one, takes into consideration compliant structures instead of rigid ones. Compliance effects, which were considered in the past as a "defect" to be mechanically eliminated, can be viceversa regarded as desired features and can be properly controlled in order to achieve desired properties from the robotic device. In particular, this is true for robot hands, where the mechanical complexity of "classical" design solutions has always originated complicated structures, often with low reliability and high costs. In this paper, an alternative solution to the design of dexterous robot hand is illustrated, considering a "mechatronic approach" for the integration of the mechanical structure, the sensory and electronic system, the control and the actuation part. Moreover, the preliminary experimental activity on a first prototype is reported and discussed. The results obtained so far, considering also reliability, costs and development time, are very encouraging, and allows to foresee a wider diffusion of dexterous hands for robotic applications.

In this work, a cartesian impedance controller purposely designed for dexterous manipulation is described. Based on the main features of the DLR Hand II, concerning kinematic structure and sensory equipment of fingers, this control strategy allows to overcome the main problems encountered in fine manipulation, namely: effects of the friction (and unmodeled dynamics) on robot performances and occurrence of singularity conditions. The achieved control scheme bas been experimentally validated by testing it on a finger of the DLR Hand.

In advanced robotic applications where a physical interaction of the robot with the environment takes place, the measurement of interaction forces is of basic importance for a safe execution of the desired tasks. In the paper, a 'stress' sensor is presented and its basic characteristics illustrated and discussed. Among its positive features, one may find the very compact size, good electrical properties, and the fact that it can be easily 'constructed' directly on the mechanical part by deposition of the sensing elements directly on the mechanical structure. These are very interesting properties for its use in advanced robotic systems.

In space applications, it is conceivable that part of the robotic activities could involve the grasp and/or manipulation of free-floating objects in the absence of gravity. In this case, synchronous application of contacts seems to represent a basic feature in order to efficiently grasp the floating items. In this sense, an additional difficulty is that objects may have irregular shape and/or be non well positioned in the gripper workspace. These difficulties cannot be handled in a simple way with standard 2-jaw grippers, with one (or two) degrees of freedom. In the paper, an activity for designing and experimenting with a gripper for this type of operations is reported, and the first laboratory results are presented and discussed. The main features of the gripper are its kinematic configuration (3 fingers with 3 dof) and its sensorial equipment, features that improve the dexterity of this device if compared to more classical devices.