Automatic Behavior Tree Expansion with LLMs for Robotic Manipulation
Jonathan Styrud, Matteo Iovino, Mikael Norrlöf, Mårten Björkman, Christian Smith.
Matteo Iovino, PhD.
Since starting university I have been attracted by the domains of Robotics and Automation.
I am adaptable, resourceful and always dedicated to the objectives I am assigned to.
I am very comfortable with teamwork and always open to dialogue. I think that my best assets are rigor and work planning.
Work Experience
Scientist
June 2023 - Present | ABB Corporate Research, Västerås, Sweden
Project Manager of WASP Research Arena for Robotics.
Enhancing collaborative robots with AI-ready skills.
Competences: Project Management, Python, ROS2, Git.
Industrial PhD, Computer Science
February 2019 - June 2023 | ABB Corporate Research, Västerås, Sweden
Learning Behavior Trees for ease of use of robots.
Competences: Python, ROS2, Git.
Summer School Tutorial
August 2021, August 2022 | WASP, Sweden
Set up of a tutorial about Behavior Trees in Robotics in the WASP Summer School.
The task for the students is to design a Behavior Tree to control a robot solving a mobile manipulation task.
The robot has to pick items from two different conveyor belts and place them in a delivery station. The robot should recharge batteries when needed.
The code repository for the tutorial is available here.
Engineer Internship
April 2018 - September 2018 | ABB Corporate Research, Västerås, Sweden
Set up of a simulation environment with ROS and Gazebo for a mobile manipulation task.
Competences: C++, C#, RAPID (RobotStudio), ROS, Mobile Manipulation.
Engineer Internship
March 2017 - August 2017 | Safran Nacelles, Le Havre, France
Set up of a simulation environment in AMESim for a thrust reverser test bench.
Competences: AMESim modelling and contorl of hydraulic components.
Education
PhD, Computer Science
2019 - 2023 | KTH - Royal Institute of Technology, Stockholm, Sweden
School of Electrical Engineering and Computer Science (EECS).
Division of Robotics, Perception and Learning (RPL).
Research on learning Behavior Trees for ease of use of robots operating in dynamic industrial environments.
Third cycle courses on Machine Learning and Robotics.
Student Project: Vision-based Manipulation and Mobility
Student Project: The History of Heavy Metal
Student Project: RottenBanana - Movie recommender
Student Project: Dual Arm Manipulation: cooperative object carrying
Research Visit
April 2022 - September 2022 | ETH Zurich, Zurich, Switzerland
Autonomous System Lab (ASL), led by Prof. Roland Siegwart.
M.Sc. in Control System Engineering
2015 - 2018 | Università degli Studi di Padova, Padova, Italy | 110/110 Cum Laude
Double Degree in the T.I.M.E. (Top International Managers in Engineering) Association.
Digital Control, Control of Mechanical Systems (PID control), Systems Theory (state space control, LQ control), Estimation and Filtering (maximum likelihood estimator, Wiener filter, Kalman filter, Particle filter), Digital Signal Processing, Advanced Topics in Control (non linear control theory).
M.Sc. in Engineering
2016 - 2018 | École Centrale de Nantes, Nantes, France | Major: Robotics | GPA 3.82
Double Degree in the T.I.M.E. (Top International Managers in Engineering) Association.
Manipulator Robot Modelling, Advanced Programming (C++), Vision for Robotics, Non Linear Control and Observation, Intelligent Vehicle and Transport, Modelling and Control of Unmanned Systems (aerial/submarine), Middleware (ROS).
Student Project: Influence de la musique sur les micro-organismes
B.Sc. in Information Engineering
2015 - 2018 | Università degli Studi di Padova, Padova, Italy | 103/110
Mathematical Analysis 1 and 2, General Physics 1 and 2, Linear Algebra and Geometry, Computer Architecture, Data Structure and Algorithms 1, Signals and Systems, Systems and Models, Statistical Data Analysis, Control Theory, Electronics, Digital Electronics, Microelectronics Laboratory, Telecommunications.
Publications
Behavior Trees in Industrial Applications: A Case Study in Underground Explosive Charging
Mattias Hallen, Matteo Iovino, Shiva Sander-Tavallaey, Christian Smith.
Comparison between Behavior Trees and Finite State Machines
Matteo Iovino, Julian Förster, Pietro Falco, Jen Jen Chung, Roland Siegwart, Christian Smith.
A Framework for Learning Behavior Trees in Collaborative Robotic Applications
Matteo Iovino, Jonathan Styrud, Pietro Falco, Christian Smith.
On the programming effort required to generate Behavior Trees and Finite State Machines for robotic applications
Matteo Iovino, Julian Förster, Pietro Falco, Jen Jen Chung, Roland Siegwart, Christian Smith.
Interactive Disambiguation for Behavior Tree Execution
Matteo Iovino, Fethiye Irmak Doğan, Iolanda Leite, Christian Smith.
Combining Context Awareness and Planning to Learn Behavior Trees from Demonstration
Oscar Gustavsson, Matteo Iovino, Jonathan Styrud, Christian Smith.
Combining Planning and Learning of Behavior Trees for Robotic Assembly
Jonathan Styrud, Matteo Iovino, Mikael Norrlöf, Mårten Björkman, Christian Smith.
Learning behavior trees with genetic programming in unpredictable environments
Matteo Iovino, Jonathan Styrud, Pietro Falco, Christian Smith.
A survey of Behavior Trees in robotics and AI
Matteo Iovino, Edvards Scukins, Jonathan Styrud, Petter Ögren, Christian Smith.