The validation of systems based on deep learning for use in safety-critical applications proves to be inherently difficult, since their sub symbolic mode of operation does not provide adequate levels of abstraction for representation and proof of correctness. The VeryHuman project aims to synthesize such levels of abstraction by observing and analysing the behaviour of upright walking of a two-legged humanoid robot. The theory to be developed is the starting point for the definition of an appropriate reward function to optimally control the movements of the humanoid by means of enhanced learning, as well as for verifiable abstraction of the corresponding kinematic models, which can be used to validate the behaviour of the robot more easily.
Duration:
01.06.2020
till
31.05.2024
Donee:
German Research Center for Artificial Intelligence GmbH
Overall workflow in the project VeryHuman. Source: DFKI, Foto: Daniel Harnack
Biologically inspired control algorithms for robots have been proven very successful. Often, these algorithms use techniques such as reinforcement learning or optimal control to perform sophisticated movement patterns with a robot (for e.g. humanoid walking). However, two main challenges exist for these learning-based approaches:
A robust hardware of the robot along with an accurate simulation of the system is required. For example, the robot can be subjected to a large amount of holonomic constraints including internal closed loops and external contacts which pose challenges to the accuracy of the simulation.
Second, control algorithms of this kind can be hard to implement due to lack of knowledge of reward and constraints. As an example, consider the upright walking movement for a two-legged humanoid robot. It is not immediately clear how one can specify the task of “upright walking”. We might try to relate different body parts (head above shoulders, shoulders above waist, waist above legs), use physical stability criteria (centre of pressure, zero moment point etc), but do these really specify walking and what are non-trivial properties? This leads to the non-trivial task of defining a suitable reward function for (deep) reinforcement learning approaches or cost function for optimal control approaches along with constraints.
This project aims at three basic research questions:
How can we formulate and prove properties of a complex humanoid robot, and
how can we efficiently combine reinforcement learning and optimal control-based approaches, and
how can we make use of symbolic properties to derive a reward function in a deep reinforcement learning approach or optimal control approach for complex use cases such as humanoid walking.
These three research questions are closely interwoven and are dealt with in three work areas that lead to the overarching goal of the project: a methodology to develop a hybrid of deep (reinforcement) learning and optimization based control of a robot together with a corresponding rational reconstruction of its observed and future behavior. This reconstruction is based on observations of the robot’s movements and general knowledge of physics (rigid body dynamics). The overall approach is shown in Fig 2. The demonstration scenario includes showing walking for a complex series-parallel humanoid robot RH5 which has been recently developed at DFKI-RIC (see Fig. 1).
Videos
Introducing RH5 Manus: A Powerful Humanoid Upper Body Design for Dynamic Movements
Design, Analysis and Control of the Series-Parallel Hybrid RH5 Humanoid Robot
Torque-limited simple pendulum: A toolkit for getting started with underactuated robotics
Publications
2023
AcroMonk: A Minimalist Underactuated Brachiating Robot
Mahdi Javadi, Daniel Harnack, Paula Stocco, Shivesh Kumar, Shubham Vyas, Daniel Pizzutilo, Frank Kirchner
In IEEE Robotics and Automation Letters, IEEE, volume 8, pages 1-8, Jun/2023.
A Recursive Lie-Group Formulation for the Second-Order Time Derivatives of the Inverse Dynamics of Parallel Kinematic Manipulators
Andreas Müller, Shivesh Kumar, Thomas Kordik
In IEEE Robotics and Automation Letters, IEEE, volume 8, pages 1-8, Apr/2023.
2022
Experimental Investigations into Using Motion Capture State Feedback for Real-Time Control of a Humanoid Robot
Mihaela Popescu, Dennis Mronga, Ivan Bergonzani, Shivesh Kumar, Frank Kirchner
In Sensors - Open Access Journal, Multidisciplinary Digital Publishing Institute (MDPI), volume 22, number 24, pages 1-12, Dec/2022.
Modular and Hybrid Numerical-Analytical Approach - A Case Study on Improving Computational Efficiency for Series-Parallel Hybrid Robots
Rohit Kumar, Shivesh Kumar, Andreas Mueller, Frank Kirchner
In 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), (IROS-2022), 23.10.-27.10.2022, Kyoto, IEEE, Nov/2022.
Torque-limited simple pendulum: A toolkit for getting familiar with control algorithms in underactuated robotics
Felix Wiebe, Jonathan Babel, Shivesh Kumar, Shubham Vyas, Daniel Harnack, Melya Boukheddimi, Mihaela Popescu, Frank Kirchner
In Journal of Open Source Software, The Open Journal, volume 7, number 74, pages 1-7, Jun/2022.
Kinematic Analysis of a Novel Humanoid Wrist Parallel Mechanism
Christoph Stoeffler, Adriano del Rio Fernandez, Heiner Peters, Moritz Schilling, Shivesh Kumar
Editors: Bruno Siciliano, Oussama Khatib
In ARK 2022: Advances in Robot Kinematics 2022, (ARK-2022), 27.6.-30.6.2022, Bilbao, Springer, series Springer Proceedings in Advanced Robotics, volume 24, pages 348-355, Jun/2022. ISBN: 978-3-031-08140-8.
A Survey on Design and Control of Lower Extremity Exoskeletons for Bipedal Walking
Ibrahim Tijjani, Shivesh Kumar, Melya Boukheddimi
In Applied Sciences, MDPI, volume 12, number 5, pages 1-31, Feb/2022.
Analytic Estimation of Region of Attraction of an LQR Controller for Torque Limited Simple Pendulum
Lukas Groß, Lasse Jenning Shala, Shivesh Kumar, Frank Kirchner, Christoph Lüth
In 61st IEEE Conference on Decision and Control, (CDC-2022), 6.12.-9.12.2022, Cancun, Institute of Electrical and Electronics Engineers (IEEE), Jan/2022.
Post-Capture Detumble Trajectory Stabilization for Robotic Active Debris Removal
Shubham Vyas, Lasse Maywald, Shivesh Kumar, Marko Jankovic, Andreas Mueller, Frank Kirchner
In Advances in Space Research, Elsevier Ltd., volume 1, pages 1-18, 2022.
2021
Closed-form time derivatives of the equations of motion of rigid body systems
Andreas Müller, Shivesh Kumar
In Multibody System Dynamics, Springer, volume o.A., pages o.A., Jul/2021.
Design, Analysis and Control of the Series-Parallel Hybrid RH5 Humanoid Robot
Julian Eßer, Shivesh Kumar, Heiner Peters, Vinzenz Bargsten, José de Gea Fernández, Carlos Mastalli, Olivier Stasse, Frank Kirchner
In 2020 IEEE-RAS 20th International Conference on Humanoid Robots (Humanoids), (Humanoids-2020), 19.7.-21.7.2021, Munich/Virtual, IEEE, pages 400-407, Jul/2021.
Nth Order Analytical Time Derivatives of Inverse Dynamics in Recursive and Closed Forms
Shivesh Kumar, Andreas Mueller
In 2021 IEEE International Conference on Robotics and Automation (ICRA), (ICRA-2021), 30.5.-05.6.2021, Xi'an, IEEE, Jun/2021. IEEE.