DC1
Tatsuki Fujioka
Host institutions and supervisors
Tampere University (TAU), Prof. Reza Ghabcheloo (TAU), Prof. Esa Rahtu (TAU)
Bosch Rexroth, Dr. Christine Brach
Scientific background
I received B.Sc. in Electrical Engineering at University of Twente in the Netherlands, and M.Sc. in Systems and Control at Delft University of Technology, where I developed a strong foundation in control systems, system identification, along with a particular passion for Bayesian inference. I have a broad research interest, ranging from control, machine learning, and sensor fusion to compressed sensing and neuroscience.
Research project
“Real-time Multimodal perception for sensor fusion on a mobile working machine”
How should excavators perceive their surroundings to perform their tasks autonomously? My research focuses on developing multimodal perception frameworks that fuse vision, proprioception, vibration, and pressure signals to understand unstructured dynamic environments, including rocks, soils, pile materials, and other working machines. Better understanding of the surrounding environment is essential for improving the robustness of subsequent motion prediction and control strategies.
DC2
Faezeh Mosayyebi
Host institutions and supervisors
Tampere University (TAU), Prof. Esa Rahtu (TAU), Prof. Reza Ghabcheloo (TAU)
Volvo Autonomous Solutions, Dr. Mohammad Loni
Scientific background
I hold a master’s degree in electrical engineering from the University of Tabriz, Iran. My academic research primarily focused on computer vision, image processing, and deep learning. My research experience includes visual recognition, image analysis, and synthetic data generation. Professionally, I have experience as a Perception Engineer in the autonomous driving industry, focusing on 3D perception, multimodal sensor fusion, object detection, and semantic segmentation.
Research project
My research focuses on developing persistent multimodal 4D maps for dynamic mining environments. I aim to integrate continuous observations from LiDAR, cameras, radar, and GNSS/IMU into spatial-temporal representations that capture geometry, semantics, and environmental changes over time. The research investigates multimodal sensor fusion, 3D/4D scene representation, change detection, and incremental map updating to support autonomous navigation, planning, and monitoring.
DC3
Emanuele Rambaldi
Host institutions and supervisors
Technical University of Munich (TUM), Prof. Achim J. Lilienthal (Chair of Perception for Intelligent Systems)
Novatron, Dr. Antti Kolu
Scientific background
My academic journey began at the University of Bologna, where I earned a bachelor’s degree in Automation Engineering. I then moved to Genoa to pursue a master’s degree in Robotics Engineering at the University of Genoa, where I built a strong academic foundation in robot programming, autonomous navigation, SLAM, and perception. Towards the end of my studies, I participated in the Erasmus+ Traineeship Programme, conducting my master’s thesis research at the ARIES Laboratory of Heidelberg University.
Research project
As-built model updating using sensor data for an automatic documenting of a worksite
My research focuses on developing a pipeline for automatically updating as-built models for infrastructure worksites. The proposed approach integrates data collected from a sensor box mounted on a heavy-duty machine to generate a reliable 3D semantic map of the environment, which is then used to automatically produce an up-to-date as-built model with minimal human intervention. The main challenges include robust perception in unstructured environments, multimodal sensor data fusion, and reliable semantic scene understanding.
DC4
Abhimanyu Bhowmik
Host institutions and supervisors
Tampere University (TAU), Prof. Joni-Kristian Kämäräinen (TAU), Prof. Esa Rahtu (TAU)
Ponsse Oyj, Kalle Einola
Scientific background
I completed my undergraduate in Artificial Intelligence at Amity University Kolkata, India. I then pursued an Erasmus Mundus Joint Master’s degree in Marine and Maritime Intelligent Robotics at the Norwegian University of Science and Technology and Université de Toulon. I worked on uncertainty-aware mapping and exploration for vision-based underwater robots for my master’s thesis. After graduating, I joined the Czech Institute of Informatics, Robotics and Cybernetics at CTU in Prague as a Junior Researcher, focusing on situational awareness for robotic manipulation.
Research project
“Sensor-based positioning of mobile work machines in a dynamic forest environment”
Recently, after the emergence of Vision Language Models, there is a surge of methods for offroad navigation, specifically self-supervised traversibility estimation. My research focuses on understanding these learned traversibility specifically for mobile working machines in dynamic and unstructured forest environments which can enable autonomous harvesting. I am also interested in studying how traversibility knowledge can be transferred between different machines and load conditions. Another focus is understanding how environmental factors affect a vehicle’s ability to move safely.
DC5
Anh Hiep Hoang
Host institutions and supervisors
Technical University of Munich (TUM), Prof. Achim J. Lilienthal (Chair of Perception for Intelligent Systems)
Gim Robotics, Dr. Jose Luis Peralta
Scientific background
My academic journey began with a B.E. in Control Engineering at Hanoi University of Science and Technology (HUST), Vietnam. During my undergraduate studies, I gained research experience as a remote intern at the Dynamic Robotics and Control Lab (DRCL), University of Southern California (USC), USA. I later earned an M.Sc. in Information Communication Convergence Technology from Soongsil University, South Korea, where I specialized in 3D object detection for autonomous driving. Before starting my Ph.D. at the Technical University of Munich (TUM) under the MSCA ENGAGE Doctoral Network, I worked as an AI Researcher at DeltaX in Seoul. My research focuses on 3D perception, object detection, object tracking, scene understanding, and multimodal perception for robotics and autonomous systems.
Research project
This project aims to develop AI-driven 3D perception methods for robust real-time understanding of dynamic environments. The research focuses on multimodal sensor fusion using LiDAR, cameras, and radar, combined with foundation models for object detection, tracking, and semantic scene understanding. It also investigates self-learning, language-guided reasoning, and efficient deployment on resource-constrained edge devices for autonomous mobile machines.
DC6
Jannik Heinen
Host institutions and supervisors
Aalto University, Prof. Joni Pajarinen
Tampere University, Prof. Reza Ghabcheloo
Bosch Rexroth, Dr. Christine Brach
Scientific background
My academic journey began at the Bonn-Rhein-Sieg University of Applied Sciences, where I earned a bachelor’s degree in Mechanical Engineering with an emphasis on Mechatronics. I then moved to Berkeley to pursue a master’s degree in Mechanical Engineering at the University of California, Berkeley, where I built a strong academic foundation in the control of robotics and autonomous systems.
Research project
Offline learning for adaptable control of a multi-actuator boom using existing large but sub-optimal datasets
This research project focuses on advancing the autonomy and efficiency of heavy-duty mobile machinery, with a primary application on excavator arms. By integrating deep learning with advanced data-driven control strategies—including offline reinforcement learning and/or imitation learning—the project aims to map complex machine dynamics and automate highly repetitive or hazardous industrial tasks. The work is being conducted within the ENGAGE network across Aalto University, Tampere University, and Bosch Rexroth.
DC7
Mohammad Hosein Saeidi Mostaghim
Host institutions and supervisors:
Tampere University (TAU): Prof. Gökhan Alcan, Prof. Reza Ghabcheloo
Novatron Oy: Dr. Antti Kolu
Scientific background
I have my bachelor’s degree from KNTU university, Iran, in mechanical engineering, with research focus on control theory. I pursued my master’s degree in automation and control engineering at Politecnico di Milano, where I focused on multi agent systems and robot learning from human demonstration.
Research project
Modelling and Control for Excavation
This research addresses predictive force and loss models for control of hydraulic machines. The trajectory touches optimal experiment design, system identification, control and estimation theory. A reliable force model will enable generalizing contact-aware control, helping to adapt to variable soil conditions. And an accurate loss model will help increase efficiency of automation.
DC9
Marco Rositani
Host institutions and supervisors
FZI Forschungszentrum Informatik: Prof. Gerhard Neumann, Dr. Karam Daaboul
Bosch: Dr. Ozan Demir
Scientific background
My academic journey began at the University of Naples Federico II, where I earned a Bachelor’s degree in Automation Engineering. I then moved to Politecnico di Milano to pursue a Master’s degree in Automation and Control Engineering, where I built a strong academic foundation in advanced control theory, numerical optimization, and physics-informed machine learning.
Research project
This research aims to develop learning-based control methods for mobile working machines that generalize across machine types, tasks, and environments, addressing the sim-to-real gap through improved simulation environments and transfer techniques such as domain randomization. It investigates expressive, diffusion-based policies trained via reinforcement learning to capture the multimodal, contact-rich dynamics of soil interaction more effectively, enabling machines trained largely in simulation to adapt robustly to real-world tasks with minimal additional data.
DC10
Chuanlin Zuo
Host institutions and supervisors
Tampere University, Prof. Reza Ghabcheloo
Volvo Autonomous Solutions, Dr. Mohammad Loni
University of Tartu (not host institution), Prof. Arun Singh
Scientific background
Master of Science in Smart Aerospace and Autonomous Systems from Paris-Saclay University, France (2025) with title of thesis: “Multimodal Gait analysis and Exoskeleton control”. Bachelor of Engineering in Automation Science and Engineering from Nanjing University of Aeronautics and Astronautics, China (2022) with title of thesis: “Generalization and Application of Reinforcement Learning”.
Research project
Safe navigation on a rough and dynamic terrain
This research focuses on enabling autonomous vehicles to navigate safely in challenging off-road environments where terrain–vehicle interactions are difficult to predict. The project investigates how uncertainties in terrain conditions, vehicle dynamics, and terrain-vehicle contact can be propagated into stochastic dynamics and incorporated into motion planning and control. The goal is to enable heavy mobile working machines to navigate safely and efficiently over rough and uneven terrain.
DC11
Alireza Azimi
Host institutions and supervisors
Karlsruhe Institute of Technology (KIT), Prof. Dr.-Ing. Marcus Geimer
AGCO, Fendt, MSc. Fridolin Schuler
Scientific background
I received B.Sc. in Electrical Engineering-Electronics at University of Science and Culture in the Tehran, Iran and M.Sc. in Electrical Engineering-Control at Kharazmi University. I focused on developing RL-based control techniques for nonlinear dynamics and their application on mobile robots. My research interest includes application of control theory in agriculture autonomous systems.
Research project
Sufficient safety performance level (AgPL) for geofence-solution for autonomous agricultural vehicles
This project aims to develop and validate a virtual boundary (geofencing) system for autonomous agricultural vehicles. The proposed system ensures that the vehicle remains within a predefined operational boundary and safely prevents unintended exits from the designated area. The objectives are to maximize the usable field coverage, minimize the distance between the vehicle’s operating path and the field boundaries, and satisfy the safety requirements associated with different Agricultural Performance Levels (AgPLs).
DC12
Emil Keshishian
Host institutions and supervisors
Umeå University, Department of Physics, Dr. Martin Servin
Algoryx Simulation, Dr. Viktor Wiberg
Scientific background
I received my B.Sc. in Mechanical Engineering from K.N. Toosi University of Technology, Iran, where my bachelor’s thesis focused on the Real-Time Computation of Vehicle Body Plastic Deformation Due to Impact in a Driving Simulator. My research interests include physically based simulation, physical AI, computer graphics, and computational mechanics.
Research project:
Simulation Intelligence
My research focuses on developing novel methods for intelligent simulation of mobile working machines. This includes simulation of rigid-body and granular interactions, with a particular emphasis on strongly non-homogeneous soils, as well as scenario generation, reliability monitoring, and control using the developed simulation methods.
DC13
Selin Yavuz
Host institutions and supervisors
Karlsruher Institut für Technologie (KIT) – Karlsruhe, Prof. Dr.-Ing. Marcus Geimer
Liebherr-Werk Bischofshofen GmbH – Bischofshofen, Dr.-Ing. Manuel Bös
Scientific background
I received my B.Sc. in Electrical and Electronics Engineering at Koç University in Istanbul, Türkiye in 2023. Later I completed my M.Sc. in the Erasmus Mundus Joint Master’s Program in Intelligent Field Robotic Systems at the University of Girona (Girona, Spain), Eötvös Loránd University (Budapest, Hungary), and the University of Zagreb (Zagreb, Croatia). I worked on Urban Object Detection using Sensor Fusion for Autonomous Navigation in my master’s thesis.
Research project
“Off-highway Road Quality Maintenance for Autonomous Earth-moving Machines”
My research focuses on developing methods to evaluate terrain accessibility and drivability using digital environment maps derived from available sensor data, combining semantic-based and geometric information to characterize the environment. A central aim is to quantify surface deformation severity relative to a given machine’s off-road capability. Building on this, the project addresses optimal planning of corrective measures without disrupting the primary work task.