Purpose of Review: The review presents an overview of research approaches on human-robot interfaces in industrial and service robotics. Recent Findings: Research approaches address especially speech and gesture recognition in both fields but are more explored in service robotics. However, the importance of interfaces in industrial robots increases. Summary: The development of human-robot interfaces is leading towards intuitive interfaces, especially using speech and gesture recognition and combining them to multimodal interfaces.

In this paper, we introduce the problem of cross-modal visuo-tactile object recognition with robotic active exploration. With this term, we mean that the robot observes a set of objects with visual perception, and later on, it is able to recognize such objects only with tactile exploration, without having touched any object before. Using a machine learning terminology, in our application, we have a visual training set and a tactile test set, or vice versa. To tackle this problem, we propose an approach constituted by four steps: finding a visuo-tactile common representation, defining a suitable set of features, transferring the features across the domains, and classifying the objects. We show the results of our approach using a set of 15 objects, collecting 40 visual examples and five tactile examples for each object. The proposed approach achieves an accuracy of 94.7%, which is comparable with the accuracy of the monomodal case, i.e., when using visual data both as training set and test set. Moreover, it performs well compared to the human ability, which we have roughly estimated carrying out an experiment with ten participants.

In this work, we propose a framework to deal with cross-modal visuo-tactile object recognition. By cross-modal visuo-tactile object recognition, we mean that the object recognition algorithm is trained only with visual data and is able to recognize objects leveraging only tactile perception. The proposed cross-modal framework is constituted by three main elements. The first is a unified representation of visual and tactile data, which is suitable for cross-modal perception. The second is a set of features able to encode the chosen representation for classification applications. The third is a supervised learning algorithm, which takes advantage of the chosen descriptor. In order to show the results of our approach, we performed experiments with 15 objects common in domestic and industrial environments. Moreover, we compare the performance of the proposed framework with the performance of 10 humans in a simple cross-modal recognition task.

In the permanent mold casting process, the distribution of mold coating thickness is a significant variable with respect to the coating’s thermal resistance, as it strongly influences the mechanical properties of cast parts and the thermal erosion of expensive molds. However, efficient online coating thickness measurement is challenging due to the high working temperatures of the molds. To address this, we propose an indirect monitoring concept based on the analysis of the as-cast surface corresponding to the coated area. Our previous research proves linear correlations between the as-cast surface roughness parameter known as arithmetical mean height (Sa) and the coating thickness for various coating materials. Based on these correlations, we can derive the coating thickness from the analysis of the corresponding as-cast surface. In this work, we introduce a method to quickly evaluate the as-cast surface roughness by analyzing optical images with a deep-learning model. We tested six different models due to their high accuracies on ImageNet: Vision Transformer (ViT), Multi-Axis Vision Transformer (MaxViT), EfficientNetV2-S/M, MobileNetV3, Densely Connected Convolutional Networks (DenseNet), and Wide Residual Networks (Wide ResNet). The results show that the Wide ResNet50-2 model achieves the lowest mean absolute error (MAE) value of 1.060 µm and the highest R-squared (R²) value of 0.918, and EfficientNetV2-M reaches the highest prediction accuracy of 98.39% on the test set. The absolute error of the surface roughness prediction remains well within an acceptable tolerance of ca. 2 µm for the top three models. The findings presented in this paper hold significant importance for the development of an affordable and efficient online method to evaluate mold coating thickness. In future work, we plan to enrich the sample dataset to further enhance the stability of prediction accuracy.

Imitation learning provides an intuitive approach for robot programming by enabling robots to learn directly from human demonstrations. While recent visual imitation learning methods have shown promise, they often depend on large datasets, which limits their applicability in manufacturing scenarios where tasks and objects are highly specialized. This paper proposes a one-shot visual imitation learning framework that allows robots to acquire multi-step pick & place tasks from a single video demonstration. The framework integrates hand detection, object detection, trajectory segmentation, and skill learning through Dynamic Movement Primitives (DMPs). Hand trajectories are mapped to the robot’s end-effector, enabling the system to generalize to new object positions while significantly reducing data requirements. The approach is evaluated in simulation and achieves reliable reproduction of multi-step tasks. These results demonstrate the potential of one-shot visual imitation learning to reduce programming complexity and increase flexibility for industrial robot applications.

In this work, an approach for robot skill learning from voice command and hand movement sequences is proposed. The motion is recorded by a 3D camera. The proposed framework consists of three elements. Firstly, a hand detector is applied on each frame to extract key points, which are represented by 21 landmarks. The trajectories of index finger tip are then taken as hand motion for further processing. Secondly, the trajectories are divided into five segments by voice command and finger moving velocities. These five segments are: reach, grasp, move, position and release, which are considered as skills in this work. The required voice commands are grasp and release, as they have short duration and can be viewed as discrete events. In the end, dynamic movement primitives are learned to represent reach, move and position. In order to show the result of the approach, a human demonstration of a pick-and-place task is recorded and evaluated.