Soft robots' ability to safely navigate complex environments motivates the development of algorithms for accurate environmental interaction assessment, enabling greater autonomy. Specifically, strain-based shape and force estimation of continuum robots with embedded soft sensors poses an open challenge mainly owing to continuous softness, anisotropic deformation, and non-linear properties. Mathematical description of deformable soft bodies and accurate estimation of external forces are crucial for achieving controllable and intelligent behaviors of these robots. In this paper, a kinetostatic strain-based modeling for rod-driven soft robots (RDSR) with embedded stretch sensors is proposed, which incorporates local strains, actuation variables, and external interactions. The strain model enables full shape estimation of the robot and prediction of strain variations in soft bodies. Building on this, we develop a force estimator based on predicted and measured sensor and actuator lengths to evaluate 3D external forces, accounting for both orthogonal and tangential components relative to the backbone. Moreover, we introduce a methodology using a novel ellipsoid representation to handle tangential forces that may become insensitive in certain singular configurations. This estimator allows us to either disregard such forces when they do not influence deformation or estimate them when they become observable. Our simulations and experiments demonstrate how this approach can be used to analyze the robot's configuration and successfully estimate external forces. Finally, it is demonstrated that when the continuum arm follows trajectories with higher strain sensitivity, tangential force estimation is significantly improved.

Soft robotic grippers gently and safely manipulate delicate objects due to their inherent adaptability and softness. Limited by insufficient stiffness and imprecise force control, conventional soft grippers are not suitable for applications that require stable grasping force. In this work, we propose a soft gripper that utilizes an origami-inspired structure to achieve tunable constant force output over a wide strain range. The geometry of each taper panel is established to provide necessary parameters such as protrusion distance, taper angle, and crease thickness required for 3D modeling and FEA analysis. Simulations and experiments show that by optimizing these parameters, our design can achieve a tunable constant force output. Moreover, the origami-inspired soft gripper dynamically adapts to different shapes while preventing excessive forces, with potential applications in logistics, manufacturing, and other industrial settings that require stable and adaptive operations.

Recently, there has been a notable shift towards electronics-free designs, which offer promising integration possibilities with soft robots, reducing reliance on traditional electronics. Despite numerous demonstrations showcasing logical control, contact sensors, and gait control, the conventional quake valve-based design is gradually struggling to meet the demands of electronics-free soft robots with increasingly complex functionalities. Integrating multiple tubes, channels, and valves has led to larger and bulkier overall systems. In this study, we introduce a simple yet powerful electronics-free pneumatic valve that excels in various aspects: it allows for flexible function configurations (operating individually, in pairs, or in larger groups), offers high-frequency synchronous reverse outputs, stores valve status, and ensures efficient maintenance. We believe that this work lays the groundwork for developing straightforward yet highly effective fully autonomous soft robots.

Rod-driven soft robots (RDSR) with a well-balanced performance in terms of perception, precision, and intelligence have a great potential for application. Mathematical description and predicted sensing of deformable soft bodies are crucial to achieve controllable and intelligent behaviors of these robots. In this work, we propose a kinetostatic model for RDSR embedded with co-located sensors based on the Geometric Variable Strain (GVS) approach where local deformations, actuation lengths and external interactions are included. This approach allows us to estimate the shape of RDSR and predict the strain variation of soft bodies under internal and external interactions. Simulations and experimental results show that tip position errors are not greater than 1.8% with respect to the whole body length under different loads (0, 100, 200, 300 gf). The maximum error of predicted sensor length change is up to 2 mm and its percentage relative to the actual length does not exceed 4%. The results demonstrate the accuracy and effectiveness of the proposed model.

High precision control of soft robots is challenging due to their stochastic behavior and material-dependence nature. While RL has been applied in soft robotics, achieving precision in task execution is still a long way off. Traditionally, RL requires substantial data for convergence, often obtained from a training environment. Yet, despite exhibiting high accuracy in the training environment, RL-policies often fall short in reality due to the training-to-reality gap, and the performance is exacerbated by the stochastic nature of soft robots. This study paves the way for the implementation of RL for soft robot control to achieve high precision in task execution. Two sample-efficient adaptive control strategies are proposed, that leverage the RL-policy. The schemes can overcome stochasticity, bridge the training-to-reality gap, and attain desired accuracy even in challenging tasks such as obstacle avoidance. Additionally, deliberate and reversible damage is induced to the pneumatic actuation chamber, altering the soft robot's behavior to test the adaptability of our solutions. Despite the damage, desired accuracy was achieved in most scenarios without needing to retrain the RL-policy.

Controlling a soft robot poses a challenge due to its mechanical characteristics.Although the manufacturing process is well-established, there are still shortcomings in their control, which often limits them to static tasks.In this study, we aim to address some of these limitations by introducing a neural network-based controller specifically designed for the throwing task using a soft robotic arm.Drawing inspiration from previous research, we have devised a method for controlling the movement of the soft robotic arm during the ballistic task.By employing a feed-forward neural network, we approximate the relationship between the actuation pattern and the resulting landing position.This enables us to predict the input sequence that needs to be transmitted to the robot's actuators based on the desired landing coordinates.To validate our approach, we conducted experiments using a 2-module soft robotic arm, which was utilized to throw four different objects towards ten target boxes positioned beneath the robot.We considered two actuation modalities, depending on whether the distal module was activated.The results indicate a success rate, defined as the proportion of successful trials out of the total number of throws, of up to 68% when a single module was actuated.These findings demonstrate the potential of our proposed controller in achieving successful performance of the throwing task using a soft robotic arm.

Safe and efficient robot-environment interaction is a critical but challenging problem as robots are being increasingly employed to operate in unstructured and unpredictable environments. Soft robots are inherently compliant to safely interact with environments but their high nonlinearity exacerbates control difficulties. Meta-learning provides a powerful tool for fast online model adaptation because it can learn an efficient model from data across different environments. Thus, this work applies the idea of meta-learning for the control of soft robotics. In particular, a target-oriented proactive search strategy is firstly performed to collect environment-specific data efficiently when a new interaction environment occurs. Then meta-learning exploits past experience to train a data-driven probabilistic model prior, and the model prior is online updated to be fast adapted to the new environment. Lastly, a model-based optimal control policy is utilized to drive the robot to desired performance. Our approach controls a soft robotic manipulator to achieve the desired position and contact force simultaneously when interacting with unknown changing environments. Overall, this work provides a viable control approach for soft robots to interact with unknown environments.

Multimodal information can significantly increase the perceptual capabilities of robotic agents, at the cost of a more complex sensory processing. This complexity can be reduced by employing machine learning techniques, provided that there is enough meaningful data to train on. This paper reports on creating novel datasets constructed by employing the iCub robot equipped with an additional depth sensor and color camera. We used the robot to acquire color and depth information for 210 objects in different acquisition scenarios. At the end, the results were large scale datasets that can be used for robot and computer vision applications: multisensory object representation, action recognition, rotation and distance invariant object recognition.

Soft robotic grippers are shown to be high effective for grasping unstructured objects with simple sensing and control strategies. However, they are still limited by their speed, sensing capabilities and actuation mechanism. Hence, their usage have been restricted in highly dynamic grasping tasks. This paper presents a soft robotic gripper with tunable bistable properties for sensor-less dynamic grasping. The bistable mechanism allows us to store arbitrarily large strain energy in the soft system which is then released upon contact. The mechanism also provides flexibility on the type of actuation mechanism as the grasping and sensing phase is completely passive. Theoretical background behind the mechanism is presented with finite element analysis to provide insights into design parameters. Finally, we experimentally demonstrate sensor-less dynamic grasping of an unknown object within 0.02 seconds, including the time to sense and actuate.

The cerebellum, which is responsible for motor control and learning, has been suggested to act as a Smith predictor for compensation of time-delays by means of internal forward models. However, insights about how forward model predictions are integrated in the Smith predictor have not yet been unveiled. To fill this gap, a novel bio-inspired modular control architecture that merges a recurrent cerebellar-like loop for adaptive control and a Smith predictor controller is proposed. The goal is to provide accurate anticipatory corrections to the generation of the motor commands in spite of sensory delays and to validate the robustness of the proposed control method to input and physical dynamic changes. The outcome of the proposed architecture with other two control schemes that do not include the Smith control strategy or the cerebellar-like corrections are compared. The results obtained on four sets of experiments confirm that the cerebellum-like circuit provides more effective corrections when only the Smith strategy is adopted and that minor tuning in the parameters, fast adaptation and reproducible configuration are enabled.

Dynamic control of soft robotic manipulators is an open problem yet to be well explored and analyzed. Most of the current applications of soft robotic manipulators utilize static or quasi-dynamic controllers based on kinematic models or linearity in the joint space. However, such approaches are not truly exploiting the rich dynamics of a soft-bodied system. In this paper, we present a model-based policy learning algorithm for closed-loop predictive control of a soft robotic manipulator. The forward dynamic model is represented using a recurrent neural network. The closed-loop policy is derived using trajectory optimization and supervised learning. The approach is verified first on a simulated piecewise constant strain model of a cable driven under-actuated soft manipulator. Furthermore, we experimentally demonstrate on a soft pneumatically actuated manipulator how closed-loop control policies can be derived that can accommodate variable frequency control and unmodeled external loads.

Designing and manufacturing actuators using soft materials are among the most important subjects for future robotics. In nature, animals made by soft tissues such as the octopus have attracted the attention of the robotics community in the last years. Suckers (or suction cups) are one of the most important and peculiar organs of the octopus body, giving it the ability to apply high forces on the external environment. The integration of suction cups in soft robots can enhance their ability to manipulate objects and interact with the environment similarly to what the octopus does. However, artificial suction cups are currently actuated using fluid pressure so most of them require external compressors, which will greatly increase the size of the soft robot. In this work, we proposed the use of the ElectroHydroDynamics (EHD) principle to actuate a suction cup. EHD is a fluidic phenomenon coupled with electrochemical reaction that can induce pressure through the application of a high-intensity electric field. We succeeded in developing a suction cup driven by EHD keeping the whole structure extremely simple, fabricated by using a 3D printer and a cutting plotter. We can control the adhesion of the suction cup by controlling the direction of the fluidic flow in our EHD pump. Thanks to a symmetrical arrangement of the electrodes, composed by plates parallel to the direction of the channel, we can change the direction of the flow by changing the sign of the applied voltage. We obtained the pressure of 643 Pa in one unit of EHD pump and pressure of 1428 Pa in five units of EHD pump applying 6 kV. The suction cup actuator was able to hold and release a 2.86 g piece of paper. We propose the soft actuator driven by the EHD pump, and expand the possibility to miniaturize the size of soft robots.

The concept of morphological computation holds that the body of an agent can, under certain circumstances, exploit the interaction with the environment to achieve useful behavior, potentially reducing the computational burden of the brain/controller. The conditions under which such phenomenon arises are, however, unclear. We hypothesize that morphological computation will be facilitated by body plans with appropriate geometric, material, and growth properties, while it will be hindered by other body plans in which one or more of these three properties is not well suited to the task. We test this by evolving the geometries and growth processes of soft robots, with either manually-set softer or stiffer material properties. Results support our hypothesis: we find that for the task investigated, evolved softer robots achieve better performances with simpler growth processes than evolved stiffer ones. We hold that the softer robots succeed because they are better able to exploit morphological computation. This four-way interaction among geometry, growth, material properties and morphological computation is but one example phenomenon that can be investigated using the system here introduced, that could enable future studies on the evolution and development of generic soft-bodied creatures.

International audience

Recent developments in robotics demonstrated that bioinspiration and embodiement are powerful tools to achieve robust behavior in presence of little control. In this context morphological design is usually performed by humans, following a set of heuristic principles: in general this can be limiting, both from an engineering and an artificial life perspectives. In this work we thus suggest a different approach, leveraging evolutionary techniques. The case study is the one of improving the locomotion capabilities of an existing bioinspired robot. First, we explore the behavior space of the robot to discover a number of qualitatively different morphology-enabled behaviors, from whose analysis design indications are gained. The suitability of novelty search -- a recent open-ended evolutionary algorithm -- for this intended purpose is demonstrated. Second, we show how it is possible to condense such behaviors into a reconfigurable robot capable of online morphological adaptation (morphosis, morphing). Examples of successful morphing are demonstrated, in which changing just one morphological parameter entails a dramatic change in the behavior: this is promising for a future robot design. The approach here adopted represents a novel computed-aided, bioinspired, design paradigm, merging human and artificial creativity. This may result in interesting implications also for artificial life, having the potential to contribute in exploring underwater locomotion "as-it-could-be".

This paper reports on the development and validation of the elastodynamics model of an innovative underwater soft-bodied robot inspired by cephalopods. The vehicle, for which the model is devised, is propelled by a discontinuous activation routine which entails the collapse of an elastic shell via cable transmission and its following passive re-inflation under the action of the elastic energy stored in the shell walls. Activation routine and thrust characterization have been determined to depend massively on the capability of the shell to elastically return to its unstrained state, hence an accurate description of the dynamics of the shell during all stages of actuation and at various degrees of deformation is essential. The model, based on a geometrically exact Cosserat theory, is validated against measurement achieved from an ad-hoc experimental apparatus, bringing evidence of its aptness at capturing the key parameters of the system. Eventually the model is employed for simulating a proper propulsion routine in water demonstrating that, upon suitable parametrization of the internal and external hydrodynamics, it can reliably be employed for the realistic quantitative characterization of the cephalopod-inspired robot.

The solution of the inverse kinematics problem of soft manipulators is essential to generate paths in the task space. The inverse kinematics problem of constant curvature or piecewise constant curvature manipulators has already been solved by using different methods, which include closed-form analytical approaches and iterative methods based on the Jacobian method. On the other hand, the inverse kinematics problem of nonconstant curvature manipulators remains unsolved. This study represents one of the first attempts in this direction. It presents both a model-based method and a supervised learning method to solve the inverse statics of nonconstant curvature soft manipulators. In particular, a Jacobian-based method and a feedforward neural network are chosen and tested experimentally. A comparative analysis has been conducted in terms of accuracy and computational time.

Soft Robotics is basically intended as building robots with highly compliant materials, but it is indeed more. Soft robots can safely interact with humans and with the environment and be able to adapt to different situations. These characteristics, combined with cheap materials and simple fabrication, candidate them to lead the next robotics revolution, when robots will massively move from the highly controlled industrial environments to the unpredictable real ones. As for the hardware, compliant materials such as silicones substitute stiff metals and rigid joints, following in under-actuated robots with virtually infinite number of degrees of freedom (DOF). Controlling this extremely high dexterity using the same approach of hard robotics (i.e. a hierarchical top-down control) simply does not apply. On the contrary, a new vision of the coupling between body and intelligence has to be adopted. Instead of considering brain and morphology separated entities (with the first commanding and the second obeying) morphological computation (Pfeifer and Bongard, 2007) proposes a radically different standpoint. The nonlinearity and complexity of the body dynamics rather than problems are considered as part of the solution, as the produced richness of behaviors allow the morphology to carry part of the computational load, thus simplifying the required (central) control instead of complicating it. This phenomenon is well observed in self-organization and embodiment of biological organisms, where characteristics like adaptability, robustness and agility are widely present without being explicitly controlled, and often without any (centralized) control. In the recent years many examples of embodiment in complex biological systems have been source of inspiration for robotics, ranging from cellular reproduction to the locomotion of more evolved animals like the salamander. A remarkable case where a central brain is present but still most of the sensory-motor control emerges from the body dynamics and the interaction with the environment is the octopus. The octopus (octopus vulgaris) is an invertebrate sea animal showing high dexterity, variable stiffness and much more complex behaviors than expected from its position in the evolutionary scale. Its body has virtually infinite DOF, resulting in a very high computational cost if pretended to be fully controlled by its central nervous system. It has been proved instead that its relatively simple brain mainly sends actuation patterns and most of the computation is performed by a combination of the peripheral nervous system, the dynamics of the compliant body and the interaction with the environment (Yekutieli et al., 2005). The result is an underactuated embodied system. The element at the base of this system is the muscular hydrostat. It is an isovolumetric structure widely present in nature as a component of compliant animal bodies. Besides generating the forces for movements it also provides skeletal support, featuring extremely high dexterity and variable stiffness ability. It is composed by a combination of muscular fibers arranged in longitudinal, transverse and oblique directions (Fig. 1). The different activation patterns of these

The new and promising field of soft robotics has many open areas of research such as the development of an exhaustive theoretical and methodological approach to dynamic modeling. To help contribute to this area of research, this paper develops a dynamic model of a continuum soft robot arm driven by cables and based upon a rigorous geometrically exact approach. The model fully investigates both dynamic interaction with a dense medium and the coupled tendon condition. The model was experimentally validated with satisfactory results, using a soft robot arm working prototype inspired by the octopus arm and capable of multibending. Experimental validation was performed for the octopus most characteristic movements: bending, reaching, and fetching. The present model can be used in the design phase as a dynamic simulation platform and to design the control strategy of a continuum robot arm moving in a dense medium.

This paper reports on the development of a new kind of unmanned underwater vehicle which draws inspiration from cephalopods both in terms of morphology and swimming routine. The robot developed here is the first in its kind, being a soft aquatic robot which travels in water by pulsed-jet propulsion. The general design principles of this innovative kind of underwater robot are illustrated and a first prototype is built and tested. The experiments demonstrate an inverse correlation between the frequency of pulsation and the speed of the robot. A mathematical model which associates the kinematics of the pulsating routine to the dynamics of the swimming is devised and compared with the experiments in order to better investigate the interplay of the various design parameters.

This paper presents the concept design, the fabrication and the experimental characterization of a unit of a modular manipulator for minimal access surgery. Traditional surgical manipulators are usually based on metallic steerable needles, tendon driven mechanisms or articulated motorized links. In this work the main idea is to combine flexible fluidic actuators enabling omnidirectional bending and elongation capability and the granular jamming phenomenon to implement a selective stiffness changing. The proposed manipulator is based on a series of identical modules, each one consisting of a silicone tube with pneumatic chambers for allowing 3D motion and one central channel for the implementation of the granular jamming phenomenon for stiffening. The silicone is covered by a novel bellows-shaped braided structure maximizing the bending still limiting lateral expansion. In this paper one single module is tested in terms of bending range, elongation capability, generated forces and stiffness changing.

In this work we address the inverse kinetics problem of a non-constant curvature manipulator driven by three cables. An exact geometrical model of this manipulator has been employed. The differential equations of the mechanical model are non-linear, therefore the analytical solutions are difficult to calculate. Since the exact solutions of the mechanical model are not available, the elements of the Jacobian matrix can not be calculated. To overcome intrinsic problems of the methods based on the Jacobian matrix, we propose for the first time a neural network learning the inverse kinetics of the soft manipulator moving in three-dimensional space. After the training, a feed-forward neural network (FNN) is able to represent the relation between the manipulator tip position and the forces applied to the cables. The results show that a desired tip position can be achieved with a degree of accuracy of 1.36% relative average error with respect to the total arm length.

This paper describes the design and development of a robot with six soft limbs, with the dual capability of pushing-based locomotion and grasping by wrapping around objects. Specifically, a central platform lodges six silicone limbs, radially distributed, with cables embedded. A new mechanism-specific gait, invariant regarding the number of limbs, has been implemented. Functionally, some limbs provide stability while others push and pull the robot to locomote in the desired direction. Once the robot is close to a target, one limb is elected to wrap around the object and, thanks to the particular limb structure and the soft material, a friction-based grasping is achieved. The robot is inspired by the octopus and implements the key principles of locomotion in this animal, without coping the full body structure. For this reason it works in water, but it is not restricted to this environment. The experiments show the effectiveness of the original solution in locomotion and grasping.

The octopus is a marine animal whose body has no rigid structures. It has eight arms mainly composed of muscles organized in a peculiar structure, named muscular hydrostat, that can change stiffness and that is used as a sort of a modifiable skeleton. Furthermore, the morphology of the arms and the mechanical characteristics of their tissues are such that the interaction with the environment, namely water, is exploited to simplify the control of movements. From these considerations, the octopus emerges as a paradigmatic example of embodied intelligence and a good model for soft robotics. In this paper the design and the development of an artificial muscular hydrostat are reported, underling the efforts in the design and development of new technologies for soft robotics, like materials, mechanisms, soft actuators. The first prototype of soft robot arm is presented, with experimental results that show its capability to perform the basic movements of the octopus arm (like elongation, shortening, and bending) and demonstrate how embodiment can be effective in the design of robots.

Control of soft robots remains nowadays a big challenge, as it does in the larger category of continuum robots. In this paper a direct and inverse kinetics models are described for a non-constant curvature structure. A major effort has been put recently in modelling and controlling constant curvature structures, such as cylindrical shaped manipulators. Manipulators with non-constant curvature, on the other hand, have been treated with a piecewise constant curvature approximation. In this work a non-constant curvature manipulator with a conical shape is built, taking inspiration from the anatomy of the octopus arm. The choice of a conical shape manipulator made of soft material is justified by its enhanced capability in grasping objects of different sizes. A different approach from the piecewise constant curvature approximation is employed for direct and inverse kinematics model. A continuum geometrically exact approach for direct kinetics model and a Jacobian method for inverse case are proposed. They are validated experimentally with a prototype soft robot arm moving in water. Results show a desired tip position in the task-space can be achieved automatically with a satisfactory degree of accuracy.

Soft robots have significant advantages over traditional robots made of rigid materials. However, controlling this type of robot by conventional approaches is difficult. Reservoir computing has been demonstrated to be an effective approach for achieving rapid learning in benchmark tasks and conventional robots. In this study, we investigated the feasibility and capacity of the reservoir computing approach to embedding and switching between multiple behaviors in a on-line manner in a soft robotic arm. The result shows that this approach can successfully achieve this task.

Recently, continuum manipulators have drawn a lot of interest and effort from the robotic community, nevertheless control and modeling of such manipulators are still a challenging task especially because they require a continuum approach. In this paper, a general mechanical model with a geometrically exact approach for tendon-driven continuum manipulators is presented. This model can be applied to a wide range of manipulators thanks to the generality of the parameters which can be set. The approach proposed could as well be a powerful tool for developing the control strategy. The model is also capable of properly simulating the coupled tendon drive, because it takes into account the torsion of the robot arm rather than neglecting it, as it is common practice in other existing models.

This paper describes a walking stabilization control on a soft ground based on gait analysis for a humanoid robot. There are many researches on gait analysis on a hard ground, but few scientists analyze the walking ability of human beings on a soft ground. Therefore, we conducted anthropometric measurement using a motion capture system on a soft ground. By analyzing experimental data, we obtained two findings. The first finding is that although there are no significant differences in step width and step length, step height tends to increase to avoid the collision between the feet and a soft ground. The second finding is that there are no significant differences in the lateral CoM trajectories but the vertical CoM amplitude increases when walking on a soft ground. Based on these findings, we developed a walking stabilization control to stabilize the CoM motion in the lateral direction on a soft ground. Verification of the proposed control is conducted through experiments with a human-sized humanoid robot WABIAN-2R. The experimental videos are supplemented.

The results of the neuroscientific research show that humans tend to stabilize the head orientation during locomotion. In this paper we describe the implementation of inverse kinematics based head stabilization controller on the humanoid platform. The controller uses the IMU feedback and controls neck joints in order to align the head orientation with the global orientation reference. Thanks to the method, we can decouple the orientational motion of the head from the rest of the body. This way stabilized head becomes better platform for proprioceptive sensory apparatus, such as cameras or IMU. In the paper we present three experiments which prove that the method has good performance in damping both, high and low frequency motion of the head. We also prove that the proposed controller improves the stability of the tracked goal point on the image of in-built camera.

In this work we propose an adaptive model for the head stabilization based on a feedback error learning (FEL). This model is capable to overcome the delays caused by the head motor system and adapts itself to the dynamics of the head motion. It has been designed to track an arbitrary reference orientation for the head in space and reject the disturbance caused by trunk motion. For efficient error learning we use the recursive least square algorithm (RLS), a Newton-like method which guarantees very fast convergence. Moreover, we implement a neural network to compute the rotational part of the head inverse kinematics. Verification of the proposed control is conducted through experiments with Matlab SIMULINK and a humanoid robot SABIAN.

-We describe the design, realization and first sea trials of a small-sized Autonomous Surface Vehicle (ASV) for environmental monitoring. The robot is being developed in the framework of the HydroNet European project [1] aiming at realizing a new multi-robot system for monitoring the quality of coastal waters, rivers and lagoons. One main innovation of the robot is the capability to measure heavy metals concentrations directly onboard using sensors ad hoc developed by the consortium. This enables the system to provide real-time measurements of heavy metals concentrations potentially changing the current water monitoring methodology in which the samples are collected by a dedicated boat and analyzed in laboratory. The robot is designed for long range missions and for lodging the onboard water analysis system. Some severe constraints imposed by the addressed scenarios are considered in the design: reduced length and limited weight for ease of transportability and deployment; low draft and protected propellers to enable the ASV to move safely in shallow waters with likely floating obstacles such as plastic bags.

We report on the design and the experimental results of DustCart, a wheeled autonomous robot for door-to-door garbage collection. DustCart is able to navigate in urban environments avoiding static and dynamic obstacles and to interact with human users. The robot is managed by an Ambient Intelligence system (AmI) through a wireless connection: it navigates to collect garbage bags to the houses of users and then moves to discharge the collected waste to a dedicated area. The architecture, navigation and localization systems are described along with the results achieved in different urban sites. In particular, a localization approach based on optical beacons was used and guaranteed position errors sufficient for a safe robot navigation. We report also the first results of a long-term experimentation of the DustCart robot in Peccioli, a small town of Tuscany (Italy). This can be considered as a first step in using robotics in the everyday life of a real town for providing a real service.

The maintenance of a stable and coherent representation of the surrounding environment is an essential capability in cognitive robotic systems. Most systems employ some form of 3D perception to create internal representations of space (maps) to support tasks such as navigation, manipulation and interaction. The creation and update of such representations may represent a significant effort in the overall computation performed by the robot. In this paper we propose an architecture based on the concept of Expected Perception that allows lightweight map updates whenever the course of action happens according to the robot's expectations. It is only when the robot's predictions and the real world outcomes differ, that corrections must be done at its full extent. We performed experiments and show results in a real robotic platform with stereo (3D) perception where map corrections are proposed by simple image level (2D) comparisons.

Water, recognized as one of the most important and endangered resources to mankind, is very difficult to monitor in real time using conventional methods. Thanks to recent advancements in technology, the use of robots, able to fulfill missions such as sampling from aquatic environments, becomes feasible and can increase dramatically the quality of water monitoring. Due to the complications for the robot's movement in some aquatic scenarios, for instance, shallow waters or dry banks of rivers and lakes, amphibian locomotion appears to be necessary to guarantee a satisfying coverage of the monitoring activities in such areas. In this paper, we focus on developing a practical amphibian solution by introducing a pair of screw rotors, which rotate in opposite directions to generate locomotion on the ground. The primitive idea of this novel design is to enhance the versatility of a marine robot with a terrestrial locomotion without undermining its performance during the movements in the water. In particular, we elaborate the principle of this novel design and present some test results of a prototype on selected terrain types, which are similar to the target aquatic environment. We conclude the paper with some preliminary conclusions based on the analysis of the test results about effectiveness and efficiency issues.

This work deals with a new technique for the estimation of the parameters and number of components in a finite mixture model. The learning procedure is performed by means of a expectation maximization (EM) methodology. The key feature of our approach is related to a top-down hierarchical search for the number of components, together with the integration of the model selection criterion within a modified EM procedure, used for the learning the mixture parameters. We start with a single component covering the whole data set. Then new components are added and optimized to best cover the data. The process is recursive and builds a binary tree like structure that effectively explores the search space. We show that our approach is faster that state-of-the- art alternatives, is insensitive to initialization, and has better data fits in average. We elucidate this through a series of experiments, both with synthetic and real data.

In this paper we propose a new algorithm for the least square fitting of ellipses from scattered data. Originally based on the one proposed by Fitzgibbon et Al in 1999, our procedure is able to overcome the numerical instability of that algorithm. We test our approach versus the latter and another approach with different ellipses. Then, we present and discuss our results.

In the domain of psychology and medical science, many experiments have been conducted referring to research on animal behaviors, to study the mechanism of mental disorders and to develop psychotropic drugs to treat them. Rodents such as rats are often chosen as experimental subjects in these experiments. However, according to some researchers, the experiments on social interactions using animals are poorly- reproducible. Therefore, we consider that the reproducibility of these experiments can be improved by using a robotic agent that interacts with an animal subject. We have developed a novel quadruped rat-inspired robot, the WR-2 (Waseda Rat No.2), based on the dimension and body structure of a mature rat. It is capable of reproducing the behaviors such as walking, mounting, rearing and grooming of the rat.

The usage of Gaussian mixture models for video segmentation has been widely adopted. However, the main difficulty arises in choosing the best model complexity. High complex models can describe the scene accurately, but they come with a high computational requirements, too. Low complex models promote segmentation speed, with the drawback of a less exhaustive description. In this paper we propose an algorithm that first learns a description mixture for the first video frames, and then it uses these results as a starting point for the analysis of the further frames. Then, we apply it to a video sequence and show its effectiveness for real-time tracking multiple moving objects. Moreover, we integrated this procedure into a foreground/background subtraction statistical framework. We compare our procedure against the state-of-the-art alternatives, and we show both its initialization efficacy and its improved segmentation performance.

The purpose of this work is to investigate the applicability of a visual tracking model on humanoid robots in order to achieve a human-like predictive behavior. In humans, in case of moving targets the oculomotor system uses a combination of the smooth pursuit eye movement and saccadic movements, namely “catch up” saccades to fixate the object of interest. This work aims to validate the "catch up" saccade model in order to obtain a human-like tracking system able to correctly switch from a zero-lag predictive smooth pursuit to a fast orienting saccade for the position error compensation. Experimental results on the iCub simulator show several correspondences with the human behavior.

This paper describes and discusses the preliminary results of a behavioural study on robot social acceptability, which was carried out during a public demonstration in South Korea. Data was collected by means of direct observation of people behaviour during interaction with robots. The most interesting result to emerge is that of young people: they tended to react to the robots presence with extreme curiosity and, quite often, to treat them aggressively. In this paper, the word bullying is used to describe any kind of improper and violent behaviour, intended to cause damages or impede the robot operation. It is the authors' opinion that if not tackled appropriately, abuses towards robots may become a serious hindrance to their future deployment, and safety. Hence, the necessity to tackle this issue with dedicated solutions during the early phases of design.

Neuro-developmental engineering is a new interdisciplinary research area at the intersection of developmental neuroscience and bioengineering. Applications can be found in early detection of neuro-developmental disorders via a new generation of mechatronic toys for assessing the regular development of perceptual and motor skills in infants, in particular coordination of mobile and multiple frames of reference during manipulation. This paper focuses on the design of a novel mechatronic toy, shaped as a 5 cm (diameter) ball, i.e. small enough to be grasped with a single hand by a 1 year old child. The sensorized ball is designed to embed a kinematics sensing unit, able to sense both the orientation in 3D space and linear accelerations, as well as a force sensing unit, to detect grasping patterns during manipulation. Dimensioning of batteries able to operate for 1 hour during experimental sessions as well as a wireless communication unit are also included in the design.

The sixteen papers in this special section are devoted to human-robot interaction models and tools.

The biomedical application this paper refers to is the neuroendoscopy of the sub-arachnoid spinal space. Such a kind of endoscopy is strongly challenging due to the tiny space to be explored and to the delicate anatomical structures which lie into it. In order to enhance the degree of safety of the endoscopic intervention, a robotic system for neuroendoscopy has been developed, so as to obtain a robot-assisted exploration which aims to lower the risks for the patient and to ease the catheter maneuvering task. This paper explains how robot sensory feedbacks can be exploited to provide the surgeon with useful information for the navigation and to implement automatic control strategies. It is described how sensory feedbacks processing can improve the safety of the operation by implementing a human-robot cooperation for the understanding of the anatomical environment and the monitoring of physiological parameters. The focus of the paper is on vision and pressure sensory feedbacks. Experimental results prove the reliability and the effectiveness of the proposed algorithms and their suitability to be employed in real time operation

This paper presents a self-organizing neural network model for visuo-motor coordination of a redundant humanoid robot arm in reaching tasks. The proposed approach is based on a biologically-inspired model which replicates some characteristics of human control: learning occurs through an action-perception cycle and does not requires explicit knowledge of the geometry of the manipulator. The transformation learned is a mapping from spatial movement direction to joint rotation. During learning, the system creates relations between the motor data associated to endogenous movements performed by the robotic arm and the sensory consequences of such motor actions, i.e. the final position and orientation of the end effector. The learnt relations are stored in the neural map structure and are then used, after learning, for generating motor commands aimed at reaching a given point in 3D space. The work is an extension of (E. Guglielmelli, et al.) including the end-effector orientation control. Experimental trials confirmed the system capability to control the end effector position and orientation and also to manage the redundancy of the robotic manipulator in reaching the 3D target point even with additional constraints, such as one or more clamped joints without additional learning phases

This paper introduces the main objectives of the neurobotics project aimed at designing and developing innovative hybrid bionic systems (HBSs) by fusing neuroscience and robotics. Eight different HBSs have been jointly designed and are being developed. This paper presents in detail the telepresence platform. The neurobotics artificial lamprey model has been designed to validate a number of neuroscience models and to investigate new telepresence strategies

This paper presents the application of a neural approach in the control of a 7-DOF robotic head. The inverse kinematics problem is addressed, for the control of the gaze fixation point of two cameras mounted on the robotic head. The proposed approach is based on a biologically-inspired model, which replicates the human brain capability of creating associations between motor and sensory data, by learning. The model is implemented here by self organizing neural maps. During learning, the system creates relations between the motor data associated to endogenous movements performed by the robotic head and the sensory consequences of such motor actions, i.e. the final position of the gaze fixation point. The learnt relations are stored in the neural map structure and are then used, after learning, for generating motor commands aimed at reaching a given fixation point. The approach proposed here allows to solve the inverse kinematics and joint redundancy problems for the ARTS robotic head, with good accuracy and robustness. Experimental trials confirmed the system capability to control the gaze direction and fixation point and also to manage the redundancy of the robotic head in reaching the target fixation point even with additional constraints, such as a clamped joint or two symmetric joint angles (e.g. eye joints).

In recent years, advances and improvements in engineering and robotics have been strengthening interactions between biological science and robotics in the goal of mimicking the complexity of biological systems. In this paper, motor control paradigms inspired by human mechanisms of sensory-motor coordination are applied to a biologically-inspired, purpose-designed robotic platform. The goal was to define and implement a multi-network architecture and to demonstrate that progressive learning of object grasping and manipulation can greatly increase performance of a robotic system in terms of adaptability, flexibility, growing competences and generalization, while preserving the robustness of traditional control. The paper presents the neural approach to sensory-motor coordination and shows preliminary results of the integration with the robotic system by means of simulation tests and experimental trials.

This paper presents an experimental investigation on a novel interface for high level control of robotic hands, based on selected foot movements. A prototype has been developed that integrates 4 sensitive areas, battery, and electronics for data acquisition and wireless transmission into a wearable insole. The prototype foot interface has been experimentally validated in the control of a robotic hand prosthesis. Comparative experimental trials were conducted with 10 able-bodied subjects, with both the foot interface and an EMG-based control, which represents the most advanced interface currently available in clinical implants for amputees. The results confirmed the effectiveness of the foot interface in the control of the hand prosthesis and showed a significant decrease in required adaptation and learning from the user's side.

This work presents an adaptive segmentation algorithm for endoscopic images. It is part of a complete system for robot-assisted endoscopy of the human sub-arachnoid spinal space. The role of the vision system is to provide a feedback for assisting the navigation of the endoscope and helping avoid damages to delicate tissues. Due to the presence of small blood vessels, nerves, and possible fibrosis, a multi-step approach has been followed for segmentation of the lumen (corresponding to free space for navigation) and the other tissues. Histogram analysis, together with blob analysis and a modified implementation of the convex hull algorithm bring to the isolation of the lumen; by means of adaptive thresholding nerves are isolated; thresholding on the hue and saturation helps in recognizing the vessels. A special condition of dirty lumen helps in managing doubtful situations. Experimental trials have been conducted on video streams from endoscopic explorations of animal (pig) spinal cord in-vivo. Experimental results show that membranes, vessels, nerves, and lumen are recognized in a reliable way, so as to contribute to robot-assisted endoscopy. The speed of the processing resulted compatible with an envisaged use in real tune support of endoscopic navigation.

This paper describes an experimental work conducted in order to estimate the conditions of applicability of expected perception (EP) based on a scheme for robot sensorimotor coordination. The starting hypothesis is that predictions of incoming sensory data can improve sensorymotor coordination respect to pure feedback loops. This implies that the environment presents a level of predictability, as in realistic environments. An implementation of the EP-based scheme has been realized on a platform composed by the Dexter 8-d.o.f. robotic arm and a color camera, for executing a pushing task in a real-world environment. Its performance, where defined as a combination of the error in the trajectory following and the computational effort, has been compared with that of a feedback-based system executing the same task in the same environmental conditions. The results have been put in relation with the degree of environmental predictability, which was controlled in the experimental trials. The experimental results give support and useful insights for analyzing the applicability of the EP-based scheme.

This paper presents the design and development of a 3-axial artificial vestibular system to be integrated on robotic heads, in order to provide a sense of head position and motion. In accordance with the role of the vestibular system in humans, this artificial vestibular system is specifically devoted to regulate and stabilize gaze during head motion, by means of the vestibulo-ocular reflex (VOR) mechanism. Future applications explore the possibility of using the artificial vestibular system on human heads, as an indirect interface between humans and robots, for 'tele-control'.

The application of robots in contact with humans, in services or assistance, implies the realization of effective and acceptable human-robot interaction. In humans, vision plays a significant role in social interaction, like for example in the recognition of faces and expressions, as well as in gesture understanding. Visuo-motor coordination of eye and neck movements in humans presents high performances in terms of accuracy, speed, effectiveness. This is due to the mechanical, cinematic, and dynamic features of the head muscular-skeletal apparatus and to the peculiar processing of visual data, detected with a space variant resolution. The work presented in this paper aims at employing retina-like cameras in human-like robotic head, reproducing similar degrees of freedom, ranges of motion, speeds, and accelerations of human neck and eye movements, in order to improve visuo-motor coordination. The use of retina-like cameras allows a faster human-like processing of visual information, better suited for the control of head movements. On the other hand, retina-like cameras provide high resolution information only in a small area of the image and thus need to be dynamically focused on the points of interest. Experimental trials were focused on the capability of identifying and tracking human faces, as a first step of human-robot interaction. Preliminary experimental results show the feasibility of a smooth face tracking, by a closed control loop for the head movements, based on retina-like vision.

The paper proposes an anticipation mechanism to improve the perception-action loop of robots interacting with real-world environments. According to recent neuroscientific findings, sensory anticipation can increase the effectiveness of perception-action loops and reduce the delays in obtaining the sensory information, especially in case of complex sensory modalities like vision, that affect pure feed-back structures. In the proposed scheme, perception crucially involves comparison processes between incoming stimuli and expected perceptions (EPs), built from previous perceptions, current motor commands, and internal models of the robot and the environment. Background knowledge plays here a helpful role, as it reduces the computational burden of perception and motor coordination tasks in partially structured environments. In the work presented here, an EP mechanism has been applied in the visuo-motor coordination of an anthropomorphic 8 d.o.f. robotic manipulator equipped with a vision system, in order to evaluate the conditions of applicability of the proposed strategy, and to validate the viability and effectiveness of the initial hypothesis.

In the development of humanoid personal robots, accompanying the life of humans in their everyday activities, vision is no doubt an extremely important sensory capability to provide the robots with. We propose the application of robot vision to the identification of hand posture, as a first step in gesture detection, for two main classes of reasons: 1) the recognition of human hand posture can help human-robot interaction, especially in relation to 'teaching by demonstration'; 2) the recognition of the robot hand can allow visual servoing, especially helpful in environments and tasks with a high level of uncertainty. The paper presents the development of a vision system for hand posture recognition, by describing the basic algorithmic implementation, based on symbolic representations, and reporting the experimental results obtained in the recognition of a three-fingered robotic hand.

The development of humanoid robots and their application as personal robots introduce problems related to the interaction of robots with humans and with environments specifically designed for human beings. In tasks where human-robot cooperation is required, compliance of the robot arm can provide a helpful solution, not only to ensure safety, but especially to increase the robot functionality and usability. The concept of functional compliance is illustrated and an experimental validation is reported, where the functionality of an anthropomorphic robot arm is comparatively assessed with and without compliant control. The paper describes the implementation of a compliant control scheme on an anthropomorphic robotic manipulator, illustrates in detail the methodology adopted for the experimental comparative validation and reports the results obtained in the execution of a sample task by a set of potential users, showing the increased performance in the case of compliant control. Indications are also provided on the improvement of acceptability, which is also affected by the enhanced performance.

This paper presents a research work on compliant control of an anthropomorphic robot arm used as a personal robot. In personal applications of robotics, human-robot interaction represents a critical factor for a robot design and introduces strict requirements on its behavior and control, which has to ensure safety and effectiveness. In this work, the problem of controlling the Dexter anthropomorphic robot arm with variable compliance has been investigated, not only to ensure safety in the interaction with humans, but especially to increase the robot functionality in tasks of physical interaction, performed in co-operation with humans. Two different control schemes have been formulated and implemented, to compare their performance experimentally. Both schemes aim at realising a self-controlled compliant behavior without using information from force/torque sensors. The experimental comparison outlines how the performance of the two control systems are inverted with respect to the theoretical considerations, based on the classical control theory, on their accuracy and effectiveness.

The aim of the work is the integration of an anthropomorphic robotic platform, starting from a neurophysiological model of grasping, in order to provide tools for an experimental validation of the model and also to provide new anthropomorphic solutions for robot grasping. The resulting robotic system is composed of an anthropomorphic arm/hand system and visual and tactile sensors. Grasp planning, control and learning have been achieved by a neural approach, inspired by a model of the inter-connections among the brain areas involved in grasping, as formulated by neurophysiologists. After a description of objectives and specifications for the robotic platform, the system is illustrated and experimental results are reported. Finally, the results are discussed as a starting point for current activities, involving the development of novel human-like robotic components and the implementation of more sophisticated learning schemes.

A research work on the interaction control of a cable-actuated robot arm, the Dexter arm, is presented in this paper. Firstly, general considerations on the cable-actuated structures and their application potential are provided and then the Dexter structure peculiarities are accurately analyzed in order to develop proper control solutions. Starting from the analysis of the limitations of the compliance control schemes in Cartesian space and in joint space, previously implemented and experimentally validated on the Dexter arm, a novel control strategy, named impedance-compliance controller, is developed. The proposed control strategy tries to combine the benefits of a compliance control scheme in Cartesian space with the benefits of an impedance control scheme in the operational space by compensating the dynamics of the sole proximal joints. The impedance-compliance controller is capable to achieve accurate smooth motions while guaranteeing functional control of the whole structure, even though a greater computational complexity is required The last section of the paper, dedicated to the experimental results, points out the differences with the previously experimented control solutions anti provides some proofs of the increased Dexter functionality.

The research in the field of advanced robotics is turning its attention more and more to man and his assistance, by developing systems such as service robots, personal robots, and even humanoid robots. Interaction control of such robot manipulators is of paramount importance for an effective execution of manipulation and tracking and, over all, for a safe and effective interaction with the humans. The paper concerns the problem of the control of an 8 degree of freedom anthropomorphic arm named DEXTER, mounted on the mobile platform of the MOVAID System, a robotic system for household personal assistance. The goal is to realize a compliant control for this manipulator in tasks of assistance to disabled and elderly people. On the basis of the control theory applied to industrial robotics, a specific compliant control solution has been developed for the DEXTER peculiar mechanical structure and actuation system, which cause a coupled joint configuration. The solution provides the capability of regulating the robot compliance according to the level of stiffness of the interaction environment. The paper describes the theoretical model of the control system, the implementation on the MOVAID platform and the experimental results in the execution of a set of demonstration tasks.

The field of humanoids robotics is widely recognized as the current challenge for robotics research. Developing humanoids poses fascinating problems in the realization of manipulation capability, which is still one of most complex problem in robotics. The paper, starting from an overview of current activities in the development of humanoid robots, with special focus on manipulation, presents the authors' approach to the design and development of anthropomorphic sensorized hands and of anthropomorphic control and sensory-motor coordination schemes. Current achievements at the Scuola Superiore Sant'Anna and Centro INAIL RTR (Research Centre on Rehabilitation Bioengineering) in the development of a robotic human prosthesis are described, together with preliminary experimental results, as well as in the implementation of biologically-inspired schemes for control and sensory-motor co-ordination, derived from models of well-identified human brain areas.

The capability of autonomously discovering relations between perceptual data and motor actions is crucial for the development of robust adaptive robotic systems intended to operate in a changing and unknown environment. In the case of robotic tactile perception, proper interaction between contact sensing and motor control is the basic step towards the execution of complex motor procedures such as grasping and manipulation. In this paper we propose an approach to the development of tactile-motor coordination in robotics, based on a neural model of the human tactile-motor system. The definition of such model is based on the features of biological systems as investigated by neuroscience. The autonomous development of tactile-motor coordination achieved through the implementation of the neural model is evaluated by experimental trials using a sensorised prosthetic hand and a robotic manipulator. The proposed neural network architecture linking changes in the sensed tactile pattern with the motor actions performed is described and experimental results are analysed and discussed.

A general framework of artificial perception for personal robots is proposed, and a subset of the framework, devoted to the face problem of robotic grasping and manipulation, is implemented. A series of experiments has been carried out using an anthropomorphic approach, both in the sensory system and in the processing modules. In particular planning of the pre-grasping hand shaping, learning of motor co-ordination strategies, exploration and grasping of an object and object classification based on the visuo-tactile information perceived during the exploration phase are described. Experiments indicate that the proposed framework may lead to practical results towards the implementation of "humanoid" robots.

This paper presents an integrated approach to tactile perception, both in terms of data acquisition and data interpretation. In humans, touch sensing is implemented through a number of different sensing elements embedded in the skin. The interpretation of perceived data to the level of detection of basic features, such as material, shape of surface, shape of contact, is achieved by integrating the different sensorial inputs at a low level, with no involvement of high level cognitive processes. The approach we propose in this paper follows this anthropomorphic model of tactile perception, by including, on one hand, a miniature fingertip integrating different sensors and, on the other hand, a parallel data interpretation module implemented through a fuzzy neural-network which processes all the different inputs at the same level. The paper describes the characteristics of the integrated fingertip sensor and of the neuro-fuzzy system, and discusses experimental results achieved during exploratory tasks on a set of common objects are discussed in detail.

The traditional approach to automation and robotics has focused so far mainly on assembly problems, whereas the managing of manufactured products at the end of their life cycle has been almost entirely neglected. However, disassembly and recycling are becoming important factors as the ecological and economical implications of manufacturing raise increasing concerns. As an initial investigation of the very general problem of disassembly, in this paper the authors outline first the motivations and the potentially very important perspectives of this approach for robotics and automation research and for industrial application. Then the authors present a robotic system for the extraction, recognition and sorting of individual objects from an "agglomerate". Such operations are of crucial importance in several disassembly tasks, such as recycling and raw materials recovering. The system is based on the integration of different sensory modalities with motor actions. An example of application of the system is described and experimental results are discussed.>

Disassembly and recycling are becoming increasingly important in our society, especially for their ecological implications. In this paper the authors present an approach to disassembly problems, which is essentially based on the concepts of multisensory integration and fusion and on the use of purposive actions to simplify perceptual tasks. The authors present a robotic system for the recognition and sorting of individual objects from a group, operations that are extremely important in most disassembly tasks. An example application of the system is described and experimental results are discussed. The effectiveness of the system in operating in a partially structured environment, shows how problems which are difficult to manage by using a single sensory modality can be solved by integrating multisensory data.>