“Modelling and Control of a Robotic System – Hybrid Legged-Wheeled Robot Example” – Vítor Pinto (FEUP)
“Planning and Control for Hybrid Locomotion of Wheeled-Legged Robots” – Marko Bjelonic (ETH Zurich)
Monday, 22 of November, from 14:00 to 16:00
Escola Superior de Tecnologia e Gestão, Campus de Sta. Apolónia, room 114, 5300-253 Bragança – Portugal
The event is free and takes place in hybrid mode (in-person and remote) but requires pre-registration at:
“Modelling and Control of a Robotic System – Hybrid Legged-Wheeled Robot Example”
This talk will focus on modelling robotic systems realistically, considering some of their non-linearities. These models are very important for their realistic simulation, decreasing the time required for each controller development and testing iterations, for example. It can also be an important support for the development of trajectory planning algorithms in mobile robots. In order to arrive at these models with the greatest possible precision, real tests are carried out on the hardware in question, and the interesting parameters for the model are estimated. This entire process will be covered throughout the talk. After this, a control technique that makes it possible to deal with multi-input multi-output (MIMO) systems will be presented, namely the state space approach.
The application example will be a novel hybrid legged-wheeled robot prototype.
Vítor Pinto is currently a researcher and teaching assistant at the Department of Electrical and Computer Engineering at the University of Porto. In the past, he was awarded an Innovation for Services prize at FEUP. He has received his Master’s Degree in Electrical and Computer Engineering from the University of Porto in 2015, and a PhD in Robotic Systems at the University of Porto in 2021. So far, he has participated in 20 scientific articles in peer-reviewed international conferences, 8 articles in indexed journals and 1 book chapter. His main research interests are Autonomous Robotic Systems and Mobile Robotics, also undergoing Robotic Manipulation.
“Planning and Control for Hybrid Locomotion of Wheeled-Legged Robots”
This talk presents an optimization-based framework to perform complex and dynamic locomotion strategies for robots with legs and wheels. In particular, we perform novel manoeuvres, which exploit the roller-walking robot’s full capabilities over challenging obstacles. By combining innovative techniques in motion control and planning, this work reveals the full potential of wheeled-legged robots and their superiority compared to their legged counterparts. Its locomotion planning algorithms that rely on trajectory optimization (TO) and model predictive control (MPC) algorithms optimizing the robot’s whole-body trajectory over a receding horizon. We propose a novel whole-body MPC as a single task formulation that simultaneously optimizes wheel and torso motions. This approach accurately predicts the robot’s motion and automatically discovers complex and dynamic movements cumbersome to hand-craft through heuristics. Finding more complex motions over challenging obstacles and at the robot’s limits can be achieved through TO methods, optimizing computational-expensive variables like gait timings and considering high-dimensional Centroidal models. By combining offline TO for complex motions and online MPC for continuous optimization along the offline trajectory, we can execute novel manoeuvres, including fast motions over challenging obstacles, unique motions through confined spaces, dynamic motions at the robot’s limits, and artistic dance moves.
Marko Bjelonic (born 1990 in Banja Luka, Yugoslavia) is an enthusiastic roboticist and future entrepreneur. During his doctoral degree at ETH Zurich under the supervision of Prof. Dr. Marco Hutter (ETH Zurich, Prof. Dr. Stelian Coros (ETH Zurich), and Prof. Dr. Sangbae Kim (MIT in Boston), Marko developed optimization-based methods focusing on motion control and planning for wheeled-legged robots over challenging terrain (2017 – 2021), which has more than a million views through prestigious news articles like Reuters, CNN, BBC News, IEEE Spectrum and more. Marko holds a Bachelor’s and Master’s degree in Mechanical Engineering from TU Darmstadt, Germany (2010 – 2013, 2013 – 2016). He spent the first year of his Master’s degree at the Cybernetics Department of NTNU in Trondheim, Norway (2013 – 2014). During his studies, he worked at the Mercedes-AMG GmbH in Affalterbach (Germany), Fraunhofer LBF in Darmstadt (Germany), and most notably at CSIRO in Brisbane (Australia) with Dr. Alberto Elfes and Dr. Navinda Kottege on multi-legged robots. Previous work during his Master’s thesis consists of the development of the hexapod robot, Weaver, including its state estimation, perception, autonomy, and control algorithms. In 2020, he managed his family-owned business with an annual sales of 350K USD and negotiated a company sale in 2021