The paper describes a human-robot cooperative installation methodology of heavy and bulky components based on marker-based visual servoing, force control, and human-robot cooperation. The main advance in the human-robot cooperation is achieved by a shared-control of the interaction during the installation task, relieving the human operator by the manipulated load and giving to the robot a partially autonomous behaviour in the force-tracking direction. Experimental results are shown in the context of the H2020 CleanSky 2 EURECA project in which a side-wall panel is installed in a 1:1 scale mock-up scenario of an A320 plane fuselage environment.

Thanks to the increasing interest on robotics prosthetic devices controlled by means of physiological signals, a continuously increasing number of solutions are proposed. Usually the proposed solutions are very expensive and created ad-hoc for the final user. For this reason, a large part of the possible users can not afford this kind of technology. Furthermore, the software adaptation to the user is time consuming and physically stressing for the subject. The paper presents a low cost prosthesis framework, which covers the three fundamental aspects of a rehabilitation system, i.e. the prosthesis, the sensors used to record the physiological signals, and the software connecting the two previous points. To reduce the costs we chose a 3D printed prosthetic hand from an open-source project. We recorded Electromyography (EMG) signals from the subjects' muscles by using a low cost armband, a all-in-one solution easy to wear and remove. The EMG signals are preprocessed in order to be used online, and they are used to train a probabilistic model for classification purposes. Furthermore, the model is built on data from different subjects, in order to develop a subject-independent framework, which can be used by any subject, with no need of draining training phases. We test the goodness of our solution with a leave-one-out approach by classifying three different hand grasps. Finally, the 3D printed hand reproduces the movement performed by the subject. Data were recorded from four different subjects, each of them repeating the selected movements five times, and we obtained an overall accuracy of 76.8%.

The paper describes how the graduate course "Autonomous Robotics" innovatively introduces robotics to Master of Science students of the Faculty of Computer Engineering of the University of Padova (Italy). The main contributions are: 1) The adoption of a Project-Based Learning constructivist approach. This teaching methodology makes students able to autonomously build their robotic knowledge base; 2) The assignment of laboratory experiences according to an increasing difficulty, from mobile robots (the simple Lego Mindstorms NXT) to humanoids (the Vstone Robovie-X and the Aldebaran NAO). Humanoids are not a widespread teaching tool because of their complexity: the course simplifies the resolution of the robots stability problem by adopting teleoperation; 3) The adoption of the open-source Robot Operating System framework. The framework encourages students to implement reusable code. The effectiveness of the adopted approach has been proven building a team of students that had successfully concluded the course. The team is participating in the European Robotics Challenges and has successfully accomplished the challenge's first stage.

In this work, we propose a robust and efficient method to build dense 3D maps, using only the images grabbed by an omnidirectional camera. The map contains exhaustive information about both the structure and the appearance of the environment and it is well suited also for large scale environments. We start from the assumption that the surrounding environment (the scene) forms a piecewise smooth surface represented by a triangle mesh. Our system is able to infer, without any odometry information, the structure of the environment along with the ego-motion of the camera by performing a robust tracking of the projection of this surface in the omnidirectional image. The key idea is to use a guess of the triangle mesh subdivision based on a constrained Delaunay triangulation built according to a set of point features and edgelet features extracted from the image. In such a way, we take into account both the corners and the edges of the scene imaged by the camera, constrained by the topology of the triangulation in order to improve the stability of the tracking process. Both motion and structure parameters are estimated using a direct method inside an optimization framework, taking into account the topology of the subdivision in a robust and efficient way. We successfully tested our system in a challenging urban scenario along a large loop using an omnidirectional camera mounted on the roof of a car.

The inclusion of robotic systems in physiotherapy allows developing new solutions for the rehabilitation and support of disabled people. Our research addresses the core problem for the advancement of such applications: the availability of a human machine interface offering intuitive control of robotic devices. In this paper we present an EMG-driven model of the human lower limb based on that previously developed by Lloyd et al. We then introduce a set of enhancements that allow reducing time and memory requirements and provide real-time performances for the control of a lower limb powered orthosis.

The next generation of tools for rehabilitation robotics requires advanced human-robot interfaces able to activate the device as soon as patient's motion intention is raised. This paper investigated the suitability of Support Vector Machine (SVM) classifiers for identification of locomotion intentions from surface electromyography (sEMG) data. A phase-dependent approach, based on foot contact and foot push off events, was employed in order to contextualize muscle activation signals. Good accuracy is demonstrated on experimental data from three healthy subjects. Classification has also been tested for different subsets of EMG features and muscles, aiming to identify a minimal setup required for the control of an EMG-based exoskeleton for rehabilitation purposes.

Brain computer interface is a system that offers also a support to the patients with neuromuscular diseases as amyotrophic lateral sclerosis.In this paper are presented some works with the aim to integrate brain computer interfaces and mobile robots.The two aim of this project are: (i) to test an improved BCI experience through the help of a physical robot, so that brain signals are stronger stimulate. (ii) to use a remote robot controlled by a highly paralyzed patient via a BCI through a friendly graphic user. Some preliminary experiments are presented in this paper about one of the possible application: a robotic museum guide(PeopleBot and Pioneer3 robot), that can transmit remote visual perceptions to the patient.

This paper presents the localization of a mobile robot while simultaneously mapping the position of the nodes of a Wireless Sensor Network using only range measurements. The robot can estimate the distance to nearby nodes of the Wireless Sensor Network by measuring the Received Signal Strength Indicator (RSSI) of the received radio messages. The RSSI measure is very noisy, especially in an indoor environment due to interference and reflections of the radio signals. We adopted an Extended Kalman Filter SLAM algorithm to integrate RSSI measurements from the different nodes over time, while the robot moves in the environment. A simple pre-processing filter helps in reducing the RSSI variations due to interference and reflections. Successful experiments are reported in which an average localization error less than 1 m is obtained when the SLAM algorithm has no a priori knowledge on the wireless node positions, while a localization error less than 0.5 m can be achieved when the position of the node is initialized close to the their actual position. These results are obtained using a generic path loss model for the transmission channel. Moreover, no internode communication is necessary in the WSN. This can save energy and enables to apply the proposed system also to fully disconnected networks.

Motion blur is a severe problem in images grabbed by legged robots and, in particular, by small humanoid robots. Standard feature extraction and tracking approaches typically fail when applied to sequences of images strongly affected by motion blur. In this paper, we propose a new feature detection and tracking scheme that is robust even to non-uniform motion blur. Furthermore, we developed a framework for visual odometry based on features extracted out of and matched in monocular image sequences. To reliably extract and track the features, we estimate the point spread function (PSF) of the motion blur individually for image patches obtained via a clustering technique and only consider highly distinctive features during matching. We present experiments performed on standard datasets corrupted with motion blur and on images taken by a camera mounted on walking small humanoid robots to show the effectiveness of our approach. The experiments demonstrate that our technique is able to reliably extract and match features and that it is furthermore able to generate a correct visual odometry, even in presence of strong motion blur effects and without the aid of any inertial measurement sensor.

In this article, we propose a new approach to the map building task: the implementation of the Spatial Semantic Hierarchy (SSH), proposed by B. Kuipers, on a real robot fitted with an omnidirectional camera. The original Kuiper's formulation of the SSH was slightly modified, in order to manage in a more efficient way the knowledge the real robot collects while moving in the environment. The sensory data experienced by the robot are transformed by the different levels of the SSH in order to obtain a compact representation of the environment. This knowledge is stored in the form of a topological map and, eventually, of a metrical map. The aim of this article is to show that a catadioptric omnidirectional camera is a good sensor for the SSH and nicely couples with several elements of the SSH. The panoramic view and rotational invariance of our omnidirectional camera makes the identification and labelling of places a simple matter. A deeper insight is that the tracking and identification of events on an omnidirectional image such as occlusions and alignments can be used for the segmentation of continuous sensory image data into the discrete topological and metric elements of a map. The proposed combination of the SSH and omnidirectional vision provides a powerful general framework for robot maping and offers new insights into the concept of “place.” Some preliminary experiments performed with a real robot in an unmodified office environment are presented.

Different coherence dimensions can be considered to improve the performances of an algorithm for computing collision translations of pairs of convex polyhedra. The algorithm's peculiar approach, based on convex minimization, is well suited to work without initialization and also endowed with an inherently embedded mechanism to exploit spatial coherence in a broader sense than other related approaches usually do. After a brief outline of the algorithm, we summarize the outcomes of several numerical experiments meant to explore extensively the incremental behavior of the algorithm while controlling the coherence parameters. In order to assess the efficacy and the potential of the approach, the performances are also discussed in the light of the results on H-Walk, an algorithm specifically designed to adapt to variable coherence.

The subject of this paper is an asymptotically fast and incremental algorithm for computing collision translations of convex polyhedra, where the problem at hand is reduced to determining collision translations of pairs of planar sections and minimizing a bivariate convex function. There are two main reasons, in our view, why the algorithm is worth consideration. On the one hand, the addressed proximity measure, namely collision translation, is not as widely studied as distance. On the other, its peculiar computation strategy may be interesting in itself, being well suited to work without initialization and also endowed with an inherently embedded mechanism to exploit spatial coherence. After outlining the main ideas of this novel approach and providing an estimation of the computational costs, we summarize a broad set of numerical experiments meant to explore extensively the behavior of the algorithm, both without and with initialization. Finally, in order to assess the efficacy and the potential of the approach under analysis, the attained performances are contrasted with those of other popular algorithms designed to compute distances between polyhedra. A thorough comparison of the reported query times and, more significantly, of the corresponding trends shows that the behavior of the collision translation algorithm is quite interesting, especially when used without initialization or under variable coherence, which should encourage further work on this approach.

This paper presents a comparative analysis relative to the experimental performances of an asymptotically fast and incremental algorithm, recently developed to compute collision translations for pairs of convex polyhedra. The algorithm may be worth considering because it solves a proximity problem which is less widely addressed than distance, as well as because of its peculiar computation strategy, well suited to work without initialization, but also endowed with an inherently embedded mechanism to exploit spatial coherence. Numerical data characterizing the behavior of the algorithm with respect to the complexity of the polyhedra have already been discussed elsewhere, thus here the main focus is on contrasting its performances with those of two popular algorithms designed to compute distances between polyhedra. Although the considered "yardsticks" answer different proximity queries, and although one of the techniques is meant to deal with general polyhedra, the results presented in this paper should help to assess the efficacy and potential of the approach under analysis. All the three algorithms, indeed, share the same kind of application context; moreover, on the basis of the asymptotic bounds discussed in the literature, distances and collision translations require similar computational efforts. A thorough comparison of the reported query times and, more significantly, of the corresponding trends seems to show that the behavior of the novel algorithm is quite interesting, especially when used without initialization, what should encourage further work on its peculiar approach

This paper considers the properties a multirobot system should exhibit to perform an assigned task cooperatively. Our experiments regard specifically the domain of RoboCup middle-size league (MSL) competitions. But the illustrated techniques can be usefully applied also to other service robotics fields like, for example, videosurveillance. Two issues are addressed in the paper. The former refers to the problem of dynamic role assignment in a team of robots. The latter concerns the problem of sharing the sensory information to cooperatively track moving objects. Both these problems have been extensively investigated over the past years by the MSL robot teams. In our paper, each individual robot has been designed to become reactively aware of the environment configuration. In addition, a dynamic role assignment policy among teammates is activated, based on the knowledge about the best behavior that the team is able to acquire through the shared sensorial information. We present the successful performance of the Artisti Veneti robot team at the MSL Challenge competitions of RoboCup-2003 to show the effectiveness of our proposed hybrid architecture, as well as some tests run in laboratory to validate the omnidirectional distributed vision system which allows us to share the information gathered by the omnidirectional cameras of our robots

The localization problem for an autonomous robot moving in a known environment is a well-studied problem which has seen many elegant solutions. Robot localization in a dynamic environment populated by several moving obstacles, however, is still a challenge for research. In this paper, we use an omnidirectional camera mounted on a mobile robot to perform a sort of scan matching. The omnidirectional vision system finds the distances of the closest color transitions in the environment, mimicking the way laser rangefinders detect the closest obstacles. The similarity of our sensor with classical rangefinders allows the use of practically unmodified Monte Carlo algorithms, with the additional advantage of being able to easily detect occlusions caused by moving obstacles. The proposed system was initially implemented in the RoboCup Middle-Size domain, but the experiments we present in this paper prove it to be valid in a general indoor environment with natural color transitions. We present localization experiments both in the RoboCup environment and in an unmodified office environment. In addition, we assessed the robustness of the system to sensor occlusions caused by other moving robots. The localization system runs in real-time on low-cost hardware.

The aim of this paper is to investigate how stigmergic information allow each individual of a group of autonomous robots to take advantages from other individual behaviors. The proposed analysis is based on the roboticle model where sensor data and effector commands are treated as energy exchange between the robot and its environment, eventually populated by other robots. Without explicit communication, the collective behavior of a group of teammates can be forced only if the robot designer makes each robot to become aware of distinguishing configuration patterns in the environment. Usually, the job is accomplished both by evaluating descriptive conditions as macroparameters and an appropriate dynamic role assignment among teammates. Since observed individual behaviors can affect the normal course of operations for each robot propagating to other teammates, we want to address some issues on how a collective behavior is fired and maintained.

One of the most challenging issues in mobile robot navigation is the localization problem in densely populated environments. In this paper, we present a new approach for vision-based localization able to solve this problem. The omnidirectional camera is used as a range finder sensitive to the distance of color transitions, whereas classical range finder;, like lasers or sonars, are sensitive to the distance of the nearest obstacles. The well-known Monte-Carlo localization technique was adapted for this new type of range sensor. The system runs in real time on a low-cost pc. In this paper we present experiments, performed in a crowded RoboCup middle-size field, proving the robustness of the approach to the occlusions of the vision sensor by moving obstacles (e.g other robots); occlusions that are very likely to occur in a real environment. Although, the system was implemented for the RoboCup environment, the system can be used in more general environments.

The robotic model considered hereafter is an attempt to deal with the problem of mobility for simplified holonomic mobile robots, such as Braitenberg vehicles. The aim is to find out which behavioural constraints are needed to make emerging a collective behaviour in the form of "mobility task". In our model "situatedness" and "embodiment" are defined as in the traditional behaviour-based approaches. The situated body of such a simplified robot is reach enough to be more than a merely symbol manipulating process. The model provides sensor data and effector commands with the same meaning of energy exchange between the robot and its environment, eventually populated by other robots. When a group of robots want to cooperate for a common goal it uses "stigmergy" to allow each individual to take advantage from other individual behaviours. Stigmergy detection is evaluated by the so called "macroparameters" and triggered by a dynamical assignment of "roles", which can force a collective behaviour. The concept of "perceptual perturbation" introduced in our robotic model is a useful abstract tool to figure out collective mobility tasks. This property stems from the "autopoietic loop" shown by each individual and which, under specified conditions, can interact with other ones to form one or more autopoietic loops involving two or more individuals. Collective behaviours eventually emerge as a consequence of a non-linear interaction among behaviours.

Reports the steps undertaken in our work aimed to demonstrate the effectiveness of an omnidirectional vision sensor when conjugated with the spatial semantic hierarchy. The spatial semantic hierarchy was proposed by Kuipers (2000) as a method for map building with robots. In our work, a robot builds a topological map of an unknown environment, using the spatial semantic hierarchy and an omnidirectional vision system as the only sensor. We present a new omnidirectional mirror and a new robot. The new mirror was expressly designed for this application, the robot's chassis was designed to create a synergy with the omnidirectional vision sensor. A complete description of our project is reported, underlying the strict link it is possible to create between omnidirectional vision and the spatial semantic hierarchy. Experiments in simulated environments and in real environments produced positive results.

The paper presents a fast algorithm to compute collision translations for pairs of convex polyhedra with some interesting features. From a theoretical viewpoint, besides the novelty of the approach, the polylog asymptotic trend in the average case is as good as that of the best algorithms proposed to solve the similar problems. On the other hand, the measured performances to detect possible collisions from scratch are satisfactory, and this is especially true in cases where the bodies do not collide. However, the most peculiar feature is a simple and flexible mechanism to exploit spatial coherence in a continuous range, which distinguishes this algorithm from all the other proposals we know. Furthermore, the nature of the approach is such that the self-tuning capability is attained at negligible additional costs even for unrelated collision tests. After a brief outline of the main ideas characterizing the approach, a set of numerical results are summarized. The proposed algorithm may be appropriate to plan collision-free paths, both online and off-line, on the basis of fine-grain descriptions of the objects in the workspace.

The subject of the paper is a fast algorithm for detecting collisions of two convex polyhedra translating in space. A major feature is the novelty of the approach: collision detection for two convex bodies is reduced to collision detection for pairs of planar sections and minimization of a bivariate convex function; furthermore, most of the subproblems are solved using two-dimensional geometry. As proved by previous theoretical work, on this basis it is possible to design an algorithm, which runs in O(log/sup 2/n) time in the average and O(log/sup 3/n) in the worst case, where n is the total number of vertices. Here the focus is on a more practical version of the algorithm, which is particularly suited to plan collision-free paths on the basis of fine-grain descriptions of the objects in the workspace, as it is the case for the systems supported by sophisticated geometric modelers. After explaining the main ideas underlying the approach, a set of experimental results are presented and discussed in some depth.

We discuss how to induce a set of collective emergent behaviors into a team of real robots used for soccer robotics. The activation of robot behaviors is organized according to a multi-level control architecture. The emergent cooperative abilities, like exchanging a ball, are achieved through the use of efficient collision avoidance algorithms implemented by a small set of robots able to frequently swap their roles. Our algorithms have been tested on a couple of real robots, Bart and Homer, which played the final game with the middle size league, at RoboCup'99, in Stockholm. This approach can be generalized to allow multi-robot systems to perform various kind of collective tasks in the entertainment field.

We discuss a practical motion planning strategy based on a two-step approach. First, an approximation of the C-space is built by a plane-sweep algorithm. Then, the search for a solution path drives the necessary refinement steps. Our claim is that an approach based on the incremental characterization of the C-space can be competitive with the best proposed motion planning techniques. We substantiate this claim in the simple case of planning translations of a convex body in the plane. Since the shape of the free space is incrementally recognized by probing the space via collision detection, every item of geometric information is obtained from the analysis of contact configurations involving convex bodies. At any stage the probes provide a partial characterization, represented by a simple cell subdivision and a suitable set of chains approximating the boundaries of the grown obstacles. The cells and their adjacencies do not change during the refinement step, so that the search strategy is straightforward. Although the performances are not optimal in theory, the planning algorithm shows a good behaviour, as demonstrated by a few experiments where it is compared with a quadtree-based strategy.

This paper is focused on solving the multiple robots motion planning problem in a robust way, whereas we say that a motion plan is robust if it can be used in spite of small variations in the motion context. After a critical review of their previous research on generating multiple robots motion plans (1995), the authors classify several "sources of instability" useful for establishing criteria for robustness. These sources include variations of the environment model, variations in the number of robots and paths, and variations in the motion parameters of the moving items. We define various "impact factors", that evaluate how much a variation affects a given plan. For example, a new path (for a robot) can be associated with a collision impact factor, that estimate how many collisions among moving objects may occur along that path. Hence, to evaluate the motion plan robustness, we define how quality values for motion plans are modified by the impact factors.

We address the problem of planning the motion of multiple autonomous robots, by analysing their behaviour in space and time. Proper representations have been studied for both domains, with a particular emphasis on the definition of some performance indexes to weight the goodness of a path concerning motion and time performances. Sublinear algorithms have been used for planning in space while reasoning in the temporal domain is based on a proper subdivision of the time axis that leads to polynomial algorithms.

A general scenario of disassembly is figured out, starting from the authors' past study of the internal structures of machines and devices, based on a mechanical engineer's point of view. The authors argue that among all data relevant to assembly, the key data for generating assembly sequences are the liaison type and the assembly direction. From these data many important geometrical, topological and technological constraints can be inferred that can be used as criteria for reducing the user's input data and the information to be extracted from the CAD database. These criteria minimize the possible decompositions at each stage of disassembly, and may even guide the search towards the best plan. A fast planner has been implemented in PROLOG, giving a user friendly system capable of generating optimal assembly plans very quickly, with a computational cost roughly proportional to the number of parts in the final assembly.>

Starting from their (1989) past work on geometric modelling, the authors discuss various techniques for the robot motion planning problem and suggest further work in this field. The efficiency of any approach to motion planning is strongly related to the ability of getting relevant information from a small subset of the geometric items. Considering a restricted portion of the configuration space around the current configuration of the introduces some local issues. So, a spatial reasoning system is defined by the interactions between local focusing and a global view of the planning process.>