Search

The paper addresses the problem of measuring whole-body dynamics for a multiple-branch kinematic chain in presence of unknown external wrenches. The main result of the paper is to give a methodology for computing whole body dynamics by aligning a model of the system dynamics with the measurements coming from the available sensors. Three primary sources of information are exploited: (1) em- bedded force/torque sensors, (2) embedded inertial sensors, (3) distributed tactile sensors (i.e. artificial skin). In order to cope with external wrenches applied at continuously chang- ing locations, we model the kinematic chain with a graph which dynamically adapts to the contact locations. Classi- cal pre-order and post-order traversals of this dynamically evolving graph allow computing whole-body dynamics and estimate external wrenches. Theoretical results have been implemented in an open-source software library (iDyn) re- leased under the iCub project. Experimental results on the iCub humanoid robot show the effectiveness of the proposed approach.

In this paper we describe how a humanoid robot can learn a representation of its own reachable space from motor experience: a Reachable Space Map. The map provides information about the reachability of a visually detected object (i.e. a 3D point in space). We propose a bio-inspired solution in which the map is built in a gaze-centered reference frame: the position of a point in space is encoded with the motor configuration of the robot head and eyes which allows the fixation of that point. We provide experimental results in which a simulated humanoid robot learns this map autonomously and we discuss how the map can be used for planning whole-body and bimanual reaching.

Humanoid robots use tactile fingertips to perform object manipulation. In this paper, an improved prototype of a tactile capacitive fingertip for the hands of the humanoid robot iCub is presented. In particular, a conductive silicone based material has been used to improve the characteristics of the capacitive sensor and an additional silicone foam layer has been also introduced in the sensor architecture to prevent behavior variation due to fingertip deterioration (e.g., caused by object manipulation). A characterization of the fingertip sensing behavior is presented. This work is funded by the European Commission as part of the project ICT-FP7-231500 RoboSKIN.

Visual motion is a simple yet powerful cue widely used by biological systems to improve their perception and adaptation to the environment. Examples of tasks that greatly benefit from the ability to detect movement are object segmentation, 3D scene reconstruction and control of attention. In computer vision several algorithms for computing visual motion and optic flow exist. However their application in robotics is not straightforward as in these platforms visual motion is often dominated by (self) motion produced by the movement of the robot (egomotion) making it difficult to disambiguate between motion induced by the scene dynamics or by the own actions of the robot. Independent motion detection is an active field in computer vision and robotics, however approaches in this area typically require that some models of both the environment and the robot visual system are available and are hardly suitable for real-time control. In this paper we describe the motionCUT, a derivation of the Lucas-Kanade optical flow algorithm that allows detecting moving objects, irrespectively of the egomotion produced by the robot. Our method is purely visual and does not require information other than the images coming from the cameras. As such it can be easily adapted to any robotic platform. The system was tested on a stereo tracking task on the iCub humanoid robot, demonstrating that the algorithm performs well and can easily execute in real-time.

The paper describes constructive approach of active perception for anthropomorphic robots. The key idea is that a robot tries to identify a human's action as an own action based on the observation of action effects for objects. In the proposed framework, the active perception are decomposed into the three phases; (1) a robot voluntarily generates actions to discover the own body and other objects. Here the object is defined as what the robot can interact with. (2) the robot characterizes its own action with its effect for objects. (3) the robot identifies the human action as the own action. The mirrored perception of the own action and the human's action allows the robot to share motor intelligence with humans. The approach was evaluated in experiments with a humanoid robot equipped with multi sensory modalities such as vision, tactile, and proprioceptive sensing.

The paper describes a constructive approach on active perception for anthropomorphic robots. The key idea is that a robot tries to identify a human's action as an own action based on the observation of action effects for objects. In the proposed framework, the active perception is decomposed into the three phases; First, a robot voluntarily generates actions to discover the own body and objects. Second, the robot characterizes its own action with the effect for the objects. Third, the robot identifies the human action with the own action. The mirrored perception of the own action and the human's action allows the robot to share the goal-directed behavior with humans. The proposed framework of active perception was experimentally validated with the integrated sensory modalities of vision, proprioception and touch.

Even though the sense of touch is crucial for humans, most humanoid robots lack tactile sensing. While a large number of sensing technologies exist, it is not trivial to incorporate them into a robot. We have developed a compliant “skin” for humanoids that integrates a distributed pressure sensor based on capacitive technology. The skin is modular and can be deployed on nonflat surfaces. Each module scans locally a limited number of tactile-sensing elements and sends the data through a serial bus. This is a critical advantage as it reduces the number of wires. The resulting system is compact and has been successfully integrated into three different humanoid robots. We have performed tests that show that the sensor has favorable characteristics and implemented algorithms to compensate the hysteresis and drift of the sensor. Experiments with the humanoid robot iCub prove that the sensors can be used to grasp unmodeled, fragile objects.

This paper deals with the problem of estimating the position of tactile elements (i.e. taxels) that are mounted on a robot body part. This problem arises with the adoption of tactile systems with a large number of sensors, and it is particularly critical in those cases in which the system is made of flexible material that is deployed on a curved surface. In this scenario the location of each taxel is partially unknown and difficult to determine manually. Placing the device is in fact an inaccurate procedure that is affected by displacements in both position and orientation. Our approach is based on the idea that it is possible to automatically infer the position of the taxels by measuring the interaction forces exchanged between the sensorized part and the environment. The location of the contact is estimated through force/torque (F/T) measures gathered by a sensor mounted on the kinematic chain of the robot. Our method requires few hypotheses and can be effectively implemented on a real platform, as demonstrated by the experiments with the iCub humanoid robot.

We present an autonomous goal-directed strategy to learn how to control a redundant robot in the task space. We discuss the advantages of exploring the state space through goal-directed actions defined in the task space (i.e. learning by trying to do) instead of performing motor babbling in the joints space, and we stress the importance of learning to be performed online, without any separation between training and execution. Our solution relies on learning the forward model and then inverting it for the control; different approaches to learn the forward model are described and compared. Experimental results on a simulated humanoid robot are provided to support our claims. The robot learns autonomously how to perform reaching actions directed toward 3D targets in task space by using arm and waist motion, not relying on any prior knowledge or initial motor babbling. To test the ability of the system to adapt to sudden changes both in the robot structure and in the perceived environment we artificially introduce two different kinds of kinematic perturbations: a modification of the length of one link and a rotation of the task space reference frame. Results demonstrate that the online update of the model allows the robot to cope with such situations.

In this paper we describe the hand of the humanoid iCub, an open source robotic platform funded by the European Commission. The principal design rationale was the necessity to supply the robot with sufficiently dexterous and sensorized hands in order to study complex skills such as manipulation. The final design has 9 actuators for each hand, 12 tactile sensors at each fingertip, 48 pressure sensors in the palm and 17 position sensors. The electronics for the sensors are embedded in the hand. As a preliminary application we tested the hand in a simplified grasping scenario in order to quantitatively evaluate the reliability and repeatability of the overall system.

In this paper we describe the design of a Cartesian Controller for a generic robot manipulator. We address some of the challenges that are typically encountered in the field of humanoid robotics. The solution we propose deals with a large number of degrees of freedom, produce smooth, human-like motion and is able to compute the trajectory on-line. In this paper we support the idea that to produce significant advancements in the field of robotics it is important to compare different approaches not only at the theoretical level but also at the implementation level. For this reason we test our software on the iCub platform and compare its performance against other available solutions.

One of the long term objectives of robotics and artificial cognitive systems is that robots will increasingly be capable of interacting in a cooperative and adaptive manner with their human counterparts in open-ended tasks that can change in real-time. In such situations, an important aspect of the robot behavior will be the ability to acquire new knowledge of the cooperative tasks by observing humans. At least two significant challenges can be identified in this context. The first challenge concerns development of methods to allow the characterization of human actions such that robotic systems can observe and learn new actions, and more complex behaviors made up of those actions. The second challenge is associated with the immense heterogeneity and diversity of robots and their perceptual and motor systems. The associated question is whether the identified methods for action perception can be generalized across the different perceptual systems inherent to distinct robot platforms. The current research addresses these two challenges. We present results from a cooperative human-robot interaction system that has been specifically developed for portability between different humanoid platforms. Within this architecture, the physical details of the perceptual system (e.g. video camera vs IR video with reflecting markers) are encapsulated at the lowest level. Actions are then automatically characterized in terms of perceptual primitives related to motion, contact and visibility. The resulting system is demonstrated to perform robust object and action learning and recognition on two distinct robotic platforms. Perhaps most interestingly, we demonstrate that knowledge acquired about action recognition with one robot can be directly imported and successfully used on a second distinct robot platform for action recognition. This will have interesting implications for the accumulation of shared knowledge between distinct heterogeneous robotic systems.

The sense of touch is of major importance for object handling. Nevertheless, adequate cutaneous sensors for humanoid robot hands are still missing. Designing such sensors is challenging, because they should not only give reliable measurements and integrate many sensing points into little space, but they should also be compliant and should not obstruct the other functions of the robot. This paper presents a capacitive pressure sensor system with 108 sensitive zones for the hands of the humanoid robot iCub. In particular, the palm has 48 taxels and each of the five fingertips has 12 taxels. The size and the shape of the hand are similar to that of a human child. When designing the sensors, we paid special attention to the integration on the robot. Also the ease and speed of production was an important design factor. Furthermore, the sensor incorporates silicone foam and is therefore compliant. We show the working principle of the sensor, how it has been integrated into the hands, and describe experiments that have been performed to show the characteristics of the sensor.

In order to successfully perform object manipulation, humanoid robots must be equipped with tactile sensors. However, the limited space that is available in robotic fingers imposes severe design constraints. In [1] we presented a small prototype fingertip which incorporates a capacitive pressure system. This paper shows an improved version, which has been integrated on the hand of the humanoid robot iCub. The fingertip is 14.5 mm long and 13 mm wide. The capacitive pressure sensor system has 12 sensitive zones and includes the electronics to send the 12 measurements over a serial bus with only 4 wires. Each synthetic fingertip is shaped approximately like a human fingertip. Furthermore, an integral part of the capacitive sensor is soft silicone foam, and therefore the fingertip is compliant. We describe the structure of the fingertip, their integration on the humanoid robot iCub and present test results to show the characteristics of the sensor.

We present an evaluation of different techniques for the estimation of forces and torques measured by a single six-axis force/torque sensor placed along the kinematic chain of a humanoid robot arm. In order to retrieve the external forces and detect possible contact situations, the internal forces must be estimated. The prediction performance of an analytically derived dynamic model as well as two supervised machine learning techniques, namely Least Squares Support Vector Machines and Neural Networks, are investigated on this problem. The performance are evaluated on the normalized mean square error (NMSE) and the comparison is made with respect to the dimension of the training set, the information contained in the input space and, finally, using a Euclidean subsampling strategy.

In this paper we consider the problem of ensuring that a multi-agent robot control system is both safe and effective in the presence of learning components. Safety, i.e., proving that a potentially dangerous configuration is never reached in the control system, usually competes with effectiveness, i.e., ensuring that tasks are performed at an acceptable level of quality. In particular, we focus on a robot playing the air hockey game against a human opponent, where the robot has to learn how to minimize opponent's goals (defense play). This setup is paradigmatic since the robot must see, decide and move fastly, but, at the same time, it must learn and guarantee that the control system is safe throughout the process. We attack this problem using automata-theoretic formalisms and associated verification tools, showing experimentally that our approach can yield safety without heavily compromising effectiveness.

During the last decades, interaction (with humans and with the environment) has become an increasingly interesting topic of research within the field of robotics. At the basis of interaction, a fundamental role is played by the ability to actively regulate the interaction forces. In this paper we propose a technique for controlling the interaction forces exploiting a proximal six axes force/torque sensor. The major assumption is the knowledge of the point where external forces are applied. The proposed approach is tested and validated on the four limbs of the iCub, a humanoid robot designed for research in embodied cognition. Remarkably, the proposed approach can be used to implement active compliance in other non passively back-drivable manipulators by simply inserting one or more force/torque sensor anywhere along the kinematic chain.

We describe a humanoid robot platform - the iCub - which was designed to support collaborative research in cognitive development through autonomous exploration and social interaction. The motivation for this effort is the conviction that significantly greater impact can be leveraged by adopting an open systems policy for software and hardware development. This creates the need for a robust humanoid robot that offers rich perceptuo-motor capabilities with many degrees of freedom, a cognitive capacity for learning and development, a software architecture that encourages reuse & easy integration, and a support infrastructure that fosters collaboration and sharing of resources. The iCub satisfies all of these needs in the guise of an open-system platform which is freely available and which has attracted a growing community of users and developers. To date, twenty iCubs each comprising approximately 5000 mechanical and electrical parts have been delivered to several research labs in Europe and to one in the USA.

This paper describes the control of a human-like robotic neck actuated with tendons. The controller regulates the length of the tendons to achieve a desired orientation of the neck and at the same time it maintains the tension of the tendons within certain limits. The solution we propose does not use any model of the system, but it relies on online learning of the different Jacobian mappings required by the controller. Learning, data acquisition and control are simultaneous; thus learning is completely autonomous, and purely online. We show that after enough iterations the controller produces straight trajectories in the task space and is able to maintain the tension of the tendons within safe limits.

This paper describes the design and integration of three joint torque sensors on the arm of the iCub platform [1], [2]. The objective is to enhance the robot arm with joint torque control capability. This activity is part of a general upgrade of the humanoid robot to provide the its 53 degrees of freedom low-level joint torque control capabilities. In particular, the shoulder joint is challenging because of its complex and compact mechanical structure. We first modeled the behaviour of the sensors with analytical equations and the sensor geometry were subsequently optimized using finite element structural simulations. The sensors were then constructed, and integrated in the arm assembly. Finally we present preliminary experiments to validate the design.

This paper describes the design and realization of a force sensor and its use in the control of a human-like robotic neck actuated with tendons. The sensor is designed to have high sensitivity and to be robust to large loads. We use the sensor to control the force exerted by the tendons and realize a controller for the two degrees of freedom of the neck. With respect to previous work [1] we show that the use of force feedback improves the robustness of the controller. Although quite specific to the robotic setup that we considered the work we describe shows some insights that could be useful for the control and realization of similar structures actuated with tendons.

This chapter reports to the development of the tools and methodologies that are in development within the EU, with an emphasis on the Open Source approaches with a view to performance analysis and comparison, and to provide an overview of cooperative research and especially on the use of Open platforms.

Tactile feedback is of crucial importance for object manipulation in unknown environments. In this paper we describe the design and realization of a fingertip which includes a capacitive pressure sensor with 12 sensitive zones. It is naturally shaped and its size is small enough so that it can be mounted on the fingers of the humanoid robot iCub. It also embeds the electronic device which performs A/D conversion: this is beneficial for the signal to noise ratio and reduces the number of wires required to connect the fingertip to the robot. The fingertip is made of silicone, which makes its surface and inner structure compliant and flexible. We present preliminary experiments performed with the first prototype.

One of the long-term goals for humanoid robotics is to have these robots working side-by side with humans, helping the humans in a variety of open ended tasks, which can change in real-time. In such contexts a crucial component of the robot behavior will be to adapt as rapidly as possible to regularities that can be learned from the human. This will allow the robot to anticipate predictable events, in order to render the interaction more fluid. This will be particularly pertinent in the context of tasks that will be repeated several times, or that contain sub-tasks that will be repeated within the global task. Through exposure to repetition the robot should automatically extract and exploit the underlying regularities. Here we present results from human-robot cooperation experiments in the context of a cooperative assembly task. The architecture is characterized by the maintenance and use of an ldquointeraction historyrdquo - a literal record of all past interactions that have taken place. During on-line interaction, the system continuously searches the interaction history for sequences whose onset matches the actions that are currently being invoked. Recognition of such matches allows the robot to take different levels of anticipatory activity. As predicted sequences are successively validated by the user, the level of anticipation and learning increases. Level 1 anticipation allows the system to predict what the user will say, and thus eliminate the need for verification when the prediction holds. At Level 2 allows the system to take initiative to propose the predicted next event. At Level 3, the robot is highly confident and takes initiative to perform the predicted action. We demonstrate how these progressive levels render the cooperative interaction more fluid and more rapid. Implications for further refinement in the quality of human-robot cooperation are discussed.

We report about the iCub, a humanoid robot for research in embodied cognition. At 104 cm tall, the iCub has the size of a three and half year old child. It will be able to crawl on all fours and sit up to manipulate objects. Its hands have been designed to support sophisticate manipulation skills. The iCub is distributed as Open Source following the GPL/FDL licenses. The entire design is available for download from the project homepage and repository (http://www.robotcub.org). In the following, we will concentrate on the description of the hardware and software systems. The scientific objectives of the project and its philosophical underpinning are described extensively elsewhere [1].

This paper describes a developmental approach to the design of a humanoid robot. The robot, equipped with initial perceptual and motor competencies, explores the "shape" of its own body before devoting its attention to the external environment. The initial form of sensorimotor coordination consists of a set of explorative motor behaviors coupled to visual routines providing a bottom-up sensory-driven attention system. Subsequently, development leads the robot from the construction of a "body schema" to the exploration of the world of objects. The "body schema" allows controlling the arm and hand to reach and touch objects within the robot's workspace. Eventually, the interaction between the environment and the robot's body is exploited to acquire a visual model of the objects the robot encounters which can then be used to guide a top-down attention system.

In this paper we discuss the implementation of a precise reaching controller on an upper-torso humanoid robot. The proposed solution is based on a learning strategy which does not rely on a priori models of the kinematics of the arm nor of that of the head. After learning, the robot can reach for visually identified objects in 3-D space by integrating an open loop and a closed loop component; the open loop controller allows ballistic movements, while the closed loop one performs precise positioning of the hand in visual space. Differently from other approaches we handle the critical case of redundancy in the head and the arm and propose a solution that although preliminary possesses some biological relevance.

In this paper, we describe the implementation of a precise reaching controller on an upper-torso humanoid robot. The solution we propose does not rely on prior models of the kinematic structure of either the arm or the head. A learning strategy enables the robot to acquire the required sensory-motor transformations. After learning the robot is able to precisely reach for a visually identified object in the 3-dimensional space. In this technique we use the fixation point (represented in the head joints motor space) as a reference frame to code the position of the object and to represent the eye-to-hand Jacobian matrix. This strategy successfully deals with the kinematic redundancy of the structure and constraints the dimensionality of the problem.

This paper describes a developmental approach to the design of a humanoid robot. The robot, equipped with initial perceptual and motor competencies, explores the "shape" of its own body before devoting its attention to the external environment. The initial form of sensorimotor coordination consists of a set of explorative motor behaviors coupled to visual routines providing a bottom-up sensory-driven attention system. Subsequently, development leads the robot from the construction of a "body schema" to the exploration of the world of objects. The "body schema" allows controlling the arm and hand to reach and touch objects within the robot's workspace. Eventually, the interaction between the environment and the robot's body is exploited to acquire a visual model of the objects the robot encounters which can then be used to guide a top-down attention system.

This paper describes a developmental sequence that allows a humanoid robot to learn about the shape of its body and successively about certain parts of the environment. We equipped the humanoid robot with an initial set of motor and perceptual competencies ranging from simple stereotyped actions to more sophisticated visual routines providing a bottom-up attention system. This initial form of sensorimotor coordination is sufficient to initiate the interaction with the environment and allows the robot to improve its motor and perceptual skills by first constructing a "body-schema" and later by learning about objects. The body-schema allows controlling movements to fixate, reach and touch objects in the environment. The interaction is further used to form a visual model of the objects grasped by the robot which eventually modulate the attention system in a top-down way. In another experiment we show an initial effort to study the acquisition of object affordances. We discuss the importance of sensorimotor coordination as a required step not only for the control of action but also, and more importantly, for perceptual development

This paper reports about our investigation on action understanding in the brain. We review recent results of the neurophysiology of the mirror system in the monkey. Based on these observations we propose a model of this brain system which is responsible for action recognition. The link between object affordances and action understanding is considered. To support our hypothesis we describe two experiments where some aspects of the model have been implemented. In the first experiment an action recognition system is trained by using data recorded from human movements. In the second experiment, the model is partially implemented on a humanoid robot which learns to mimic simple actions performed by a human subject on different objects. These experiments show that motor information can have a significant role in action interpretation and that a mirror-like representation can be developed autonomously as a result of the interaction between an individual and the environment.

This paper is about the implementation of grasping skills in a humanoid robot. Following a developmental approach, the robot is initially equipped with little perceptual and motor competencies whose role is to bootstrap learning through the exploration of the external environment. This crude form of sensorimotor coordination consists of a set of control systems and explorative behaviors as well as simple visual routines. The developmental path leads the robot from the exploration of the physics and geometry of its body to the probing of the external environment. The robot experience builds and modifies continuously its internal representations of the environment, being this, its body or the objects it encounters. We discuss the implications of our approach to the study of cognition and our effort to build a cognitive artificial system.

The role of technology for the brain functions are fundamental tools for the analysis of biological data and neuroscience research. In this paper the human cognition, implementing artificial systems as explicit physical models are implementing humanoid systems. Our reference framework is human sensorimotor and cognitive developments and we approach the problems by trying to implement motor and cognitive abilities of system in an artificial systems.

In biological systems vision is always in the context of a particular body and tightly coupled to action. Therefore it is natural to consider visuo-motor methods (rather than vision alone) for learning about objects in the world. Indeed, initially it may be necessary to act on something to learn that it is an object! Learning to act involves not only learning the visual consequences of performing a motor action, but also the other direction, i.e. using the learned association to determine which motor action will bring about a desired visual condition.In this paper we show how a humanoid robot uses its arm to try some simple pushing actions on an object, while using vision and proprioception to learn the effects of its actions. We show how the robot learns a mapping between the initial position of its arm and the direction the object moves in when pushed, and then how this learned mapping is used to successfully position the arm to push/pull the target object in a desired direction.

The goal of this paper is to propose a biologically plausible, functional model of the acquisition of visual, acoustic and multi-modal motor responses. Within this context visual and acoustic spatial cues are considered, fused in a coherent percept and eventually employed to control the orienting behavior of a humanoid robot. The rationale of the approach lies in the possibility to test and empirically prove the correctness of the model through the embodiment and the real interaction of the system with the environment. The model takes into account the fact that (i) acoustic and visual cues are represented with respect to different coordinate frames (head-centric versus retino-centric) and consequently they need to be “aligned” in order to be properly fused, (ii) a teaching signal has to be generated in order to inform the system that the motor performance is not adequate to perform the task (i.e. orient toward the stimulus) and thus adaptation is required, and (iii) vision plays a major role in driving the acquisition of the appropriate map of space but other sources of feedback might be employed as well.

Within the field of Neuro Robotics we are driven primarily by the desire to understand how humans and animals live and grow and solve every day's problems. To this aim we adopted a "learn by doing" approach by building artificial systems, e.g. robots that not only look like human beings but also represent a model of some brain process. They should, ideally, behave and interact like human beings (being situated). The main emphasis in robotics has been on systems that act as a reaction to an external stimulus (e.g. tracking, reaching), rather than as a result of an internal drive to explore or "understand" the environment. We think it is now appropriate to try to move from acting, in the sense explained above, to "understanding". As a starting point we addressed the problem of learning about the effects and consequences of self-generated actions. How does the robot learn how to pull an object toward itself or to push it away? How does the robot learn that spherical objects roll while a cube only slides if pushed? Interacting with objects is important because it implicitly explores object representation, event understanding, and can provide definition of object-hood that could not be grasped with a mere passive observation of the world. Further, learning to understand what one's own body can do is an essential step toward learning by imitation. In this view two actions are similar not only if their kinematics and dynamics are similar but rather if the effects on the external world are the same. Along this line of research we discuss some recent experiments performed at the AI-Lab at MIT and at the LIRA-Lab at the University of Genova on COG and Babybot respectively. We show how the humanoid robots can learn how to poke and prod objects to obtain a consistently repeatable effect (e.g. sliding in a given direction), to help visual segmentation, and to interpret a poking action performed by a human manipulator.

In this paper we describe an autonomous strategy which enables a humanoid robot to learn how to reach for a visually identified object in the 3D space. The robot is a 22-DOF upper-body humanoid with moving eyes, neck, arm and hand. The robot is bootstrapped with limited a-priori knowledge, sufficient to start the interaction with the environment; this interaction allows the robot to learn different sensorimotor mappings, required for reaching. The arm-head forward kinematic model and a visuo-motor inverse model are learned from sensory experience. Learning is performed purely online (without any separation between training and execution) through a goal-directed exploration of the environment. During the learning the robot is also able to build an internal representation of its reachable space.

Object detection is a fundamental ability for robots interacting within an environment. While stunningly effective, state-of-the-art deep learning methods require huge amounts of labeled images and hours of training which does not favour such scenarios. This work presents a novel pipeline resulting from integrating (Maiettini et al. in 2017 IEEE-RAS 17th international conference on humanoid robotics (Humanoids), 2017) and (Maiettini et al. in 2018 IEEE/RSJ international conference on intelligent robots and systems (IROS), 2018), which naturally trains a robot to detect novel objects in few seconds. Moreover, we report on an extended empirical evaluation of the learning method, justifying that the proposed hybrid architecture is key in leveraging powerful deep representations while maintaining fast training time of large scale Kernel methods. We validate our approach on the Pascal VOC benchmark (Everingham et al. in Int J Comput Vis 88(2): 303–338, 2010), and on a challenging robotic scenario (iCubWorld Transformations (Pasquale et al. in Rob Auton Syst 112:260–281, 2019). We address real world use-cases and show how to tune the method for different speed/accuracy trades-off. Lastly, we discuss limitations and directions for future development.

Object grasping and manipulation in humanoid robots depend, for advanced and long lasting performance, on the development of reliable tactile sensors. Materials used for sensor implementation have to be chosen accurately and optimized in order to endow the device with sufficient resilience to endure long lasting grasping tasks. In particular care has to be taken when trying to enhance durability by the use of materials with high shore hardness as this choice can noticeably reduce pressure sensitivity. In this paper, a comparison between two implementations of the fingertip sensor of the humanoid robot iCub is presented by using a raw characterization of the response along a straight line. Prototypes differ in the materials used for the dielectric and protective layers, while maintaining the same capacitive architecture. The relevant result that emerges from the characterization is that the prototype with Soma Foama presents a higher sensitivity compared to the prototype in PDMS, while the second has a higher mechanical strength.

Musculoskeletal disorders, the single largest category of workrelated injuries in many industrial countries, are associated with very high costs in terms of lost productivity. In highvolume production facilities, large parts of the workstation should ideally be adapted to individual workers in real time to prevent such injuries. However, in smaller production lines, especially those found in small and medium enterprises (SMEs), regularly adapting the entire workstation to accommodate flexibility is a major challenge. A solution to the problem is the development of reconfigurable human-robot collaboration (HRC) workstations and frameworks. These are key to enabling agile manufacturing, by merging the dexterity, flexibility, and problem-solving ability of humans with the strength and precision of robotics.

For robots to share the environment and cooperate with humans without barriers, we need to guarantee safety to the operator and, simultaneously, to maximize the robot's usability. Safety is typically guaranteed by controlling the robot movements while, possibly, taking into account physical contacts with the operator, objects or tools. If possible, also the safety of the robot must be guaranteed. Not less importantly, as the complexity of the robots and their skills increase, usability becomes a concern. Social interaction technologies can save the day by enabling natural human-robot collaboration. In this paper we show a possible integration of physical and social Human-Robot Interaction methods (pHRI and sHRI respectively). Our reference task is object hand-over. We test both the case of the robot initiating the action and, vice versa, the robot receiving an object from the operator. Finally, we discuss possible extension with higher-level planning systems for added flexibility and reasoning skills.

Robots are now physically capable of locomotion, object manipulation, and an essentially unlimited set of sensory motor behaviors. This sets the scene for the corresponding technical challenge: how can non-specialist human users interact with these robots for human robot cooperation? Crangle and Suppes stated in [1] : “the user should not have to become a programmer, or rely on a programmer, to alter the robot’s behavior, and the user should not have to learn specialized technical vocabulary to request action from a robot.” To achieve this goal, one option is to consider the robot as a human apprentice and to have it learn through its interaction with a human. This chapter reviews our approach to this problem.

In robotic systems with moving cameras control of gaze allows for image stabilization, tracking and attention switching. Proper integration of these capabilities lets the robot exploit the kinematic redundancy of the oculomotor system to improve tracking performance and extend the field of view, while at the same time stabilize vision to reduce image blur induced by the robot’s own movements. Gaze may be driven not only by vision but also by other sensors (e.g. inertial sensors or motor encoders) that carry information about the robot’s own movement. Humanoid robots have sophisticated oculomotor systems, usually mounting inertial devices and are therefore an ideal platform to study this problem. We present a complete architecture for gaze control of a humanoid robot. Our system is able to control the neck and the eyes in order to track a 3D cartesian fixation point in space. The redundancy of the kinematic problem is exploited to implement additional behaviors, namely passive gaze stabilization, saccadic movements, and vestibuloocular reflex. We implement this framework on the iCub’s head, which is equipped with a 3-DoFs neck and a 3-DoFs eyes system and includes an inertial unit that provides feedback on the acceleration and angular speed of the head. The framework presented in this work can be applied to any robot equipped with an anthropomorphic head. In addition we provide an opensource, modular implementation, which has been already ported to other robotic platforms.

This paper presents a set of plugins for the Gazebo simulator that enables the interoperability between a robot, controlled using the YARP framework, and Gazebo itself. Gazebo is an open-source simulator that can handle different Dynamic Engines developed by the Open Source Robotics Foundation. Since our plugins conform with the YARP layer used on the real robot, applications written for our robots, COMAN and iCub, can be run on the simulator with no changes. Our plugins have two main components: a YARP interface with the same API as the real robot interface, and a Gazebo plugin which handles simulated joints, encoders, IMUs, force/torque sensors and synchronization. Different modules and tasks for COMAN and iCub have been developed using Gazebo and our plugins as a testbed before moving to the real robots.