Among the many challenges of parallel kinematic manipulators, achieving high-speed and accurate control remains crucial. Estimating their dynamic properties is essential for designing precise and efficient control schemes. Conventional methods for dynamic model identification have been effective, though deep learning approaches have historically faced limitations due to data inefficiencies. However, recent advancements in physics-informed neural networks (PINNs) offer a way to improve both control and the extraction of interpretable physical properties from these robots. In this work, we propose and validate a PINN-based dynamic model for a Delta parallel robot, specifically the ABB IRB 360-6/1600. Our approach incorporates known physical properties, such as mass matrix sparsity, to improve accuracy and computational efficiency in dynamic model identification. To the best of our knowledge, this is the first study applying PINNs to model parallel robots. The method is validated experimentally, and its performance is compared to a validated identification technique for physically consistent identification, demonstrating the effectiveness of this approach for real-world applications in parallel robots.

The determination of the dynamic properties of a robot is especially important for designing highly accurate and efficient control systems. Conventional methods for dynamic model identification have proven to be effective, where deep learning (DL) approaches have shown limits due to data inefficiencies. However, thanks to novel physics-informed DL architectures, such as Deep Lagrangian Networks (DeLaN) [1], it is possible to control and extract interpretable physical information of a robot. This paper introduces an augmented DeLaN architecture for linear viscous and non-symmetrical Coulomb friction identification, which also learns motor parameters such as rotor inertia. An approach is proposed for comparing this method with the conventional dynamic identification and previous DeLaN implementations. Moreover, our friction and rotor inertia identification is validated, and the performance of our model is analyzed with a real robot (UR5e).

High precision industrial applications call for equally precise functioning of industrial manipulators, which in turn requires accurate modeling of the manipulators. This paper carries out a detailed study on the modeling of industrial manipulators with elastic joints to improve their accuracy. In particular, the effect of adopting a simple harmonic drive (HD) model and ignoring a dynamic effect called low inertia coupling between the actuators and links on the model accuracy has been analyzed from a parameter estimation perspective. Since the aforementioned model characteristics have been generally ignored for high gear reduction ratios, this study is carried out with five different reduction ratios ranging from low to high, where three different models of a three-joints elastic manipulator are considered. The accuracy of the models is compared using the torque performance metrics of a predefined joint motion of the robot. Furthermore, the impact of the models with different accuracy is assessed by carrying out a state-of-the-art dynamic parameter estimation, and the resulting errors are compared to ascertain the merits of adopting a detailed elastic dynamic model of a manipulator.

One major problem in industrial automation is the workload balancing problem. It consists of making the robots or, more generally, the machines, involved in the assembly process to work exactly the same, either by picking and placing the same number of pieces or by having the same number of operational cycles. This paper presents a novel strategy for solving such a problem by means of an evolutionary algorithm. The specific application of this strategy is to balance the workload of a pick-and-place process developed in the facilities of the industrial company Fameccanica Spa Data within the framework of an industrial project between the company and our research group. The novelties concerning the state-of-the-art contributions are: (1) instead of using an explicit fitness function, the candidate solutions at each iteration are evaluated by using a simulation of the entire process; (2) the parameters optimized are the velocity and acceleration of the robots involved in the line and (3) the strategy includes an algorithm for distributing the workload between the robots during the process.

International audience

Robotics in the last decades is moving towards bioinspired solutions in order to develop systems increasingly integrated with the human environment. Among them, legged robots fascinates more and more researchers thanks to their ability of moving in unstructured environment such as the ones typical of earthquakes, where in the near future robots are planned to be send to help humans while performing dangerous tasks. On the base of this findings, the authors propose a novel concept for a jumping humanoid leg, based on the key role played from the structural flexibility. The geometric and dynamic features of this leg have been selected thanks to a targeted set of numerical simulations. An extensive campaign of experimental tests useful for the validation of the numerical model here presented will be a matter of future works.

In packaging industry, the duration of the carton folding plays a fundamental role in the production process; in particular in the erection process when each panel rotates around the die-pressed lines called creases. Their bending response can be very complex, depending on forming and environment conditions. The crease mechanical properties, such as geometrical parameters, temperature, moisture and folding speed, influence the overall production. It is therefore necessary to control all of these parameters, from both a theoretical and experimental point of view. About this, an experimental setup is expressly designed and built w.r.t. the rotation angle for a paper around the respective crease. The results of this research allow demonstrating the reliability of the experimental setup, the substantial negligibility of the geometrical errors and determining the number of sample so that the dispersion is compatible with the experimental errors. Then the test-ring is suitable for carton folding investigation.

An increasing interest towards functionally independent and self-reconfigurable robots featured the research direction in the recent past. Such attention is well explained by the characteristics of flexibility and cost-effectiveness which distinguish these modular machines. This scenario framed the birth of many different devices characterized by simplicity of use and versatility. The present manuscript follows these trends by presenting a newly conceived device suitable for being used as a limb, as well as a planar two degrees of freedom manipulator. The features of both parallel and serial kinematic mechanisms have been exploited to the aim of obtaining a device as much as possible independent from the frame to whom it is connected. Formal aspects about position and velocity kinematics are addressed and a specifically built prototype is then used for experimental tests on the new limb.

In electronic manufacturing system, the design of the robotic hand with sufficient dexterity and configuration is important for the successful accomplishment of the assembly task. It is significant that the robot can grasp assembly parts and do some simple in-hand manipulation so as to fit them with the package slots. In this research, we study the process of precise in-hand posture transition problem using a novel jaw like gripper with human-sized anthropomorphic features. We transform the in-hand manipulation problem into a series of static grasping problems. Then we study the successful twisting condition on each grasp frame by analyzing its dynamic performance and requirements. Based on this data-driven idea, simulation and experimental data is obtained from both successful and failed trials. Finally, we create the distribution of parameters grasp map for successful twisting.

International audience

Intonational variation is widely regarded as a source of information about the topic structure of spoken discourse. However, many factors other than topic can influence this variation. We compared two models of intonation in terms of their ability to account for these other sources of variation. In dealing with this variation, the models paint different pictures of the intonational correlates of topic.

Contrairement a certaines langues comme l'anglais et l'italien standard, l'italien parle a Naples partage avec d'autres varietes du Sud de l'Italie un patron melodique ascendant/descendant (LHL) tres similaire pour les questions oui/non et pour les affirmations a focalisation etroite (D'Imperio, 200I ; Grice et al., a paraitre). Neanmoins, il a ete propose que les pics d'accent melodique des questions oui/non (L*+H) sont alignes temporellement plus tard que ceux des affirmations (L+H*) et que cette difference d'alignement est employee par les auditeurs de langue maternelle napolitaine pour identifier perceptuellement les deux modalites (D'Imperio and House, I997). Cette etude teste au niveau acoustique l'hypothese que les trois cibles tonales de la montee/descente sont alignees temporellement differemment et sont produites sur des echelles differentes pour les questions et les affirmations. De plus, les differences de pente des montees et des descentes ont aussi ete analysees grâce a un modele de regression logistique. Deux locuteurs napolitains ont produit des phrases dont les mots cibles, la structure syllabique et l'environnement segmental differaient pour la modalite question et la modalite affirmation. Les resultats montrent que les trois cibles dans les montees/descentes sont alignees plus tard pour les questions que pour les affirmations. Par contre, aucune difference systematique n'a ete mise en evidence pour les pentes des montees ou pour les pentes des descentes. La contribution exacte de la hauteur de la F o au contraste question/affirmation n'a pu etre determinee. En effet, alors qu'un locuteur marquait le contraste en produisant des pics plus eleves pour les affirmations, l'autre locuteur n'a produit aucune difference significative entre les deux modalites.

This paper addresses the problem of the perception of two different pitch accents in Italian which signal two utterance types (interrogative and declarative). The questions asked concern whether the major perceptual cue to this category distinction involves only the temporal alignment of the high level target with the syllable or if the category percept also depends on the presence of a rising or falling melodic movement within the syllable nucleus. The results show that the primary perceptual cue for questions is a rise through the vowel, while the primary cue for statements is a fall through the vowel. The results bear upon a general theory of intonation and our understanding of intonation in Italian as well as on current models of tonal perception in speech.