- Manufacturing Process and Optimization
- Smart Agriculture and AI
- Robot Manipulation and Learning
- Human-Automation Interaction and Safety
- Robotics and Sensor-Based Localization
- Industrial Vision Systems and Defect Detection
- Digital Transformation in Industry
- Flexible and Reconfigurable Manufacturing Systems
- Ergonomics and Human Factors
- Robotics and Automated Systems
- Robotic Path Planning Algorithms
- IoT and Edge/Fog Computing
- Additive Manufacturing and 3D Printing Technologies
- Insect behavior and control techniques
- Virtual Reality Applications and Impacts
- Additive Manufacturing Materials and Processes
- Advanced Manufacturing and Logistics Optimization
- Remote Sensing and LiDAR Applications
- Advanced Statistical Process Monitoring
- Robotic Mechanisms and Dynamics
- Injection Molding Process and Properties
- Augmented Reality Applications
- Time Series Analysis and Forecasting
- Hand Gesture Recognition Systems
- Advanced Machining and Optimization Techniques
Siemens (Germany)
2020-2024
Technical University of Munich
2020-2024
Fondazione Sandro Pitigliani
2022
Politecnico di Milano
2021
Wire Arc Additive Manufacturing is a near-net-shape processing technology which allows cost-effective manufacturing of large and customized metal parts. Processing aluminium in quite challenging, especially terms porosity. In the present work, pore behaviour AW4043/AlSi5(wt%) was investigated post-process monitoring approach developed. It has been observed that as shielding gas flow rate increases, porosity parts also increases due to rapid solidification melt pool by forced convection. The...
The presence of Deformable Linear Objects (DLOs) such as wires, cables or ropes in our everyday life is massive. However, the applicability robotic solutions to DLOs still marginal due many challenges involved their perception. In this letter, a methodology generate datasets from mixture synthetic and real samples for training segmentation approaches thus presented. method composed two steps. First, key-points along real-world DLO are labeled by employing VR tracker operated user. Second,...
This paper discusses the potential benefits of using augmented reality (AR) technology to enhance human–robot collaborative industrial processes. The authors describe a real-world use case at Siemens premises in which an AR-based authoring tool is used reduce cognitive load, assist human workers training robots, and support calibration inspection tasks during assembly tasks. study highlights AR as solution for optimizing collaboration improving productivity. article describes methodology...
Wire Arc Additive Manufacturing (WAAM) offers the possibility to build up large-scale metal parts. Data which is obtained from a multivariate sensor system in-situ must be analyzed automatically ensure an early and reliable detection of defects reduce costs due production scrap. For that reason, modular anomaly detector for time series in WAAM was investigated this paper. The approach adressed major topics real-life data sets industrial applications such as miscellaneous signal sample rates,...
The factory of the future requires flexibility and adaptability to satisfy market demands. However, these properties are difficult ensure when collaborative manipulators considered. Thus, approaches for integrating devices, fostering adoption new engineering methodologies augmenting studied. In this work we address how integration human operators with robotic might accommodate requirements using a concept lean automation. To answer question, compare current research in field, propose design...
Human–robot collaboration (HRC) is one of the key aspects Industry 4.0 (I4.0) and requires intuitive modalities for humans to communicate seamlessly with robots, such as speech, touch, or bodily gestures. However, utilizing these usually not enough ensure a good user experience consideration human factors. Therefore, this paper presents software component, Multi-Modal Offline Online Programming (M2O2P), which considers characteristics establishes communication channel robot predefined yet...
Abstract In subtractive manufacturing, differences in machinability among batches of the same material can be observed. Ignoring these deviations potentially reduce product quality and increase manufacturing costs. To consider influence batch process optimization models, needs to efficiently identified. Thus, a smart service is proposed for in-situ identification. This driven by supervised machine learning model, which analyzes signals machine’s control, especially torque data,...
The research community is puzzled with words like skill, action, atomic unit and others when describing robots' capabilities. However, for giving the possibility to integrate capabilities in industrial scenarios, a standardization of these descriptions necessary. This work uses structured review approach identify commonalities differences skill frameworks. Through this method, 210 papers were analyzed three main results obtained. First, vast majority authors agree on taxonomy based task,...
Building on the idea of Industry 4.0, new models highly connected factory that leverage factory-generated data to introduce cost-effective automation and involve human worker for creating higher added value are possible. Within this context, collaborative robots becoming more common in industry. However, promises regarding flexibility cannot be satisfied due challenging process ensuring safety. This is because current regulations standards require updates risk assessment every change robotic...
This study proposed an approach for robot localization using data from multiple low-cost sensors with two goals in mind, to produce accurate and keep the computation as simple possible. The used wheel odometry, inertial-motion Inertial Motion Unit (IMU), a location fix Real-Time Kinematics Global Positioning System (RTK GPS). Each of is prone errors some situations, resulting inaccurate localization. odometry affected by caused slipping when turning or putting it on slippery ground. IMU...
Agriculture needs to optimization for satisfying the rising demands of food due world population growth. An approach this problem is digitization agriculture through IT tools by creating digital twins. However, in twin creation uniqueness plant lost, therefore, based agricultural cultivation cannot be performed. Hence, paper proposes a methodology assign an identification marker plants crop using image analysis pipeline. To show effectiveness algorithm proposed method evaluated on Rovitis...
Adoption of human–robot collaboration is hindered by barriers in collaborative task design. A new approach for solving these problems to empower operators the design their tasks. However, how this may affect user welfare or performance industrial scenarios has not yet been studied. Therefore, research, results an experiment designed identify influences operator’s self-designed on physical ergonomics and are presented. At first, a framework able accept operator definition via parts’ locations...
Autonomous agricultural machines are attracting significant attention from the industry. Automatization helps to reduce costs, and increase productivity. This paper designs a global localization algorithm for an drone designed autonomous operation in vineyards. Particularly, main objective is obtain accurate also when GPS fails. The scheme has components: Extended Kalman Filter (EKF) online magnetometer calibration module. first fuses information inertial sensors, track encoders azimuth...
With the trend of low batch manufacturing, more and small medium enterprises are leaning towards adopting collaborative robots to increase productivity improve operator well-being. However, often programmed by robot experts rather than operators effectively working with machine. Therefore, may perceive levels task autonomy due unpredictability motions. Empowering make their own choices regarding motions can such feelings. research in cognitive science shows that allowing decide on a could be...
The manufacturing industry is undergoing rapid evolution, necessitating flexible and adaptable robots. However, configuring such machines requires technical experts, which are hard to find, especially for small medium enterprises. Therefore, the process needs be simplified by allowing non-experts configure During configuration, one key aspect definition of objects' grasping poses. literature proposes deep learning techniques compute poses automatically facilitate process. Nevertheless,...
One of the biggest challenges for artificial intelligence in industry is lack labeled application data. Particularly time series data, labeling requires a large amount data preparation and expert knowledge both analysis domain. In this work, we propose methodology solving two barriers identified above an additive manufacturing use case. Our approach correlates spatial temporal features process defects by means sensor. By applying our method, were able to achieve shorter while obtaining...
In this preliminary work, we propose to design an activity recognition system that is suitable for Industrie 4.0 (I4.0) applications, especially focusing on Learning from Demonstration (LfD) in collaborative robot tasks. More precisely, focus the issue of data exchange between and a robotic system. We with linked using functional object-oriented network (FOON) facilitate industrial use cases. Initially, drafted FOON our case. Afterwards, action estimated by object hand systems coupled...