-
Sim-to-Real of Soft Robots with Learned Residual Physics
Authors:
Junpeng Gao,
Mike Yan Michelis,
Andrew Spielberg,
Robert K. Katzschmann
Abstract:
Accurately modeling soft robots in simulation is computationally expensive and commonly falls short of representing the real world. This well-known discrepancy, known as the sim-to-real gap, can have several causes, such as coarsely approximated geometry and material models, manufacturing defects, viscoelasticity and plasticity, and hysteresis effects. Residual physics networks learn from real-wor…
▽ More
Accurately modeling soft robots in simulation is computationally expensive and commonly falls short of representing the real world. This well-known discrepancy, known as the sim-to-real gap, can have several causes, such as coarsely approximated geometry and material models, manufacturing defects, viscoelasticity and plasticity, and hysteresis effects. Residual physics networks learn from real-world data to augment a discrepant model and bring it closer to reality. Here, we present a residual physics method for modeling soft robots with large degrees of freedom. We train neural networks to learn a residual term -- the modeling error between simulated and physical systems. Concretely, the residual term is a force applied on the whole simulated mesh, while real position data is collected with only sparse motion markers. The physical prior of the analytical simulation provides a starting point for the residual network, and the combined model is more informed than if physics were learned tabula rasa. We demonstrate our method on 1) a silicone elastomeric beam and 2) a soft pneumatic arm with hard-to-model, anisotropic fiber reinforcements. Our method outperforms traditional system identification up to 60%. We show that residual physics need not be limited to low degrees of freedom but can effectively bridge the sim-to-real gap for high dimensional systems.
△ Less
Submitted 1 February, 2024;
originally announced February 2024.
-
Real Robot Challenge 2022: Learning Dexterous Manipulation from Offline Data in the Real World
Authors:
Nico Gürtler,
Felix Widmaier,
Cansu Sancaktar,
Sebastian Blaes,
Pavel Kolev,
Stefan Bauer,
Manuel Wüthrich,
Markus Wulfmeier,
Martin Riedmiller,
Arthur Allshire,
Qiang Wang,
Robert McCarthy,
Hangyeol Kim,
Jongchan Baek,
Wookyong Kwon,
Shanliang Qian,
Yasunori Toshimitsu,
Mike Yan Michelis,
Amirhossein Kazemipour,
Arman Raayatsanati,
Hehui Zheng,
Barnabas Gavin Cangan,
Bernhard Schölkopf,
Georg Martius
Abstract:
Experimentation on real robots is demanding in terms of time and costs. For this reason, a large part of the reinforcement learning (RL) community uses simulators to develop and benchmark algorithms. However, insights gained in simulation do not necessarily translate to real robots, in particular for tasks involving complex interactions with the environment. The Real Robot Challenge 2022 therefore…
▽ More
Experimentation on real robots is demanding in terms of time and costs. For this reason, a large part of the reinforcement learning (RL) community uses simulators to develop and benchmark algorithms. However, insights gained in simulation do not necessarily translate to real robots, in particular for tasks involving complex interactions with the environment. The Real Robot Challenge 2022 therefore served as a bridge between the RL and robotics communities by allowing participants to experiment remotely with a real robot - as easily as in simulation.
In the last years, offline reinforcement learning has matured into a promising paradigm for learning from pre-collected datasets, alleviating the reliance on expensive online interactions. We therefore asked the participants to learn two dexterous manipulation tasks involving pushing, gras**, and in-hand orientation from provided real-robot datasets. An extensive software documentation and an initial stage based on a simulation of the real set-up made the competition particularly accessible. By giving each team plenty of access budget to evaluate their offline-learned policies on a cluster of seven identical real TriFinger platforms, we organized an exciting competition for machine learners and roboticists alike.
In this work we state the rules of the competition, present the methods used by the winning teams and compare their results with a benchmark of state-of-the-art offline RL algorithms on the challenge datasets.
△ Less
Submitted 24 November, 2023; v1 submitted 15 August, 2023;
originally announced August 2023.
-
Low Voltage Electrohydraulic Actuators for Untethered Robotics
Authors:
Stephan-Daniel Gravert,
Elia Varini,
Amirhossein Kazemipour,
Mike Y. Michelis,
Thomas Buchner,
Ronan Hinchet,
Robert K. Katzschmann
Abstract:
Rigid robots can be precise in repetitive tasks, but struggle in unstructured environments. Nature's versatility in such environments inspires researchers to develop biomimetic robots that incorporate compliant and contracting artificial muscles. Among the recently proposed artificial muscle technologies, electrohydraulic actuators are promising since they offer performance comparable to that of m…
▽ More
Rigid robots can be precise in repetitive tasks, but struggle in unstructured environments. Nature's versatility in such environments inspires researchers to develop biomimetic robots that incorporate compliant and contracting artificial muscles. Among the recently proposed artificial muscle technologies, electrohydraulic actuators are promising since they offer performance comparable to that of mammalian muscles in terms of speed and power density. However, they require high driving voltages and have safety concerns due to exposed electrodes. These high voltages lead to either bulky or inefficient driving electronics that make untethered, high-degree-of-freedom bio-inspired robots difficult to realize. Here, we present hydraulically amplified low voltage electrostatic (HALVE) actuators that match mammalian skeletal muscles in average power density (50.5 W kg-1) and peak strain rate (971 % s-1) at a driving voltage of just 1100 V. This driving voltage is approx. 5-7 times lower compared to other electrohydraulic actuators using paraelectric dielectrics. Furthermore, HALVE actuators are safe to touch, waterproof, and self-clearing, which makes them easy to implement in wearables and robotics. We characterize, model, and physically validate key performance metrics of the actuator and compare its performance to state-of-the-art electrohydraulic designs. Finally, we demonstrate the utility of our actuators on two muscle-based electrohydraulic robots: an untethered soft robotic swimmer and a robotic gripper. We foresee that HALVE actuators can become a key building block for future highly-biomimetic untethered robots and wearables with many independent artificial muscles such as biomimetic hands, faces, or exoskeletons.
△ Less
Submitted 30 August, 2023; v1 submitted 1 June, 2023;
originally announced June 2023.
-
Beyond Regular Grids: Fourier-Based Neural Operators on Arbitrary Domains
Authors:
Levi Lingsch,
Mike Y. Michelis,
Emmanuel de Bezenac,
Sirani M. Perera,
Robert K. Katzschmann,
Siddhartha Mishra
Abstract:
The computational efficiency of many neural operators, widely used for learning solutions of PDEs, relies on the fast Fourier transform (FFT) for performing spectral computations. As the FFT is limited to equispaced (rectangular) grids, this limits the efficiency of such neural operators when applied to problems where the input and output functions need to be processed on general non-equispaced po…
▽ More
The computational efficiency of many neural operators, widely used for learning solutions of PDEs, relies on the fast Fourier transform (FFT) for performing spectral computations. As the FFT is limited to equispaced (rectangular) grids, this limits the efficiency of such neural operators when applied to problems where the input and output functions need to be processed on general non-equispaced point distributions. Leveraging the observation that a limited set of Fourier (Spectral) modes suffice to provide the required expressivity of a neural operator, we propose a simple method, based on the efficient direct evaluation of the underlying spectral transformation, to extend neural operators to arbitrary domains. An efficient implementation* of such direct spectral evaluations is coupled with existing neural operator models to allow the processing of data on arbitrary non-equispaced distributions of points. With extensive empirical evaluation, we demonstrate that the proposed method allows us to extend neural operators to arbitrary point distributions with significant gains in training speed over baselines while retaining or improving the accuracy of Fourier neural operators (FNOs) and related neural operators.
△ Less
Submitted 20 May, 2024; v1 submitted 31 May, 2023;
originally announced May 2023.
-
Aquarium: A Fully Differentiable Fluid-Structure Interaction Solver for Robotics Applications
Authors:
Jeong Hun Lee,
Mike Y. Michelis,
Robert Katzschmann,
Zachary Manchester
Abstract:
We present Aquarium, a differentiable fluid-structure interaction solver for robotics that offers stable simulation, accurately coupled fluid-robot physics in two dimensions, and full differentiability with respect to fluid and robot states and parameters. Aquarium achieves stable simulation with accurate flow physics by directly integrating over the incompressible Navier-Stokes equations using a…
▽ More
We present Aquarium, a differentiable fluid-structure interaction solver for robotics that offers stable simulation, accurately coupled fluid-robot physics in two dimensions, and full differentiability with respect to fluid and robot states and parameters. Aquarium achieves stable simulation with accurate flow physics by directly integrating over the incompressible Navier-Stokes equations using a fully implicit Crank-Nicolson scheme with a second-order finite-volume spatial discretization. The fluid and robot physics are coupled using the immersed-boundary method by formulating the no-slip condition as an equality constraint applied directly to the Navier-Stokes system. This choice of coupling allows the fluid-structure interaction to be posed and solved as a nonlinear optimization problem. This optimization-based formulation is then exploited using the implicit-function theorem to compute derivatives. Derivatives can then be passed to downstream gradient-based optimization or learning algorithms. We demonstrate Aquarium's ability to accurately simulate coupled fluid-robot physics with numerous 2D examples, including a cylinder in free stream and a soft robotic fish tail with hardware validation. We also demonstrate Aquarium's ability to provide analytical gradients by performing gradient-based shape-and-gait optimization of an oscillating diamond foil to maximize its generated thrust.
△ Less
Submitted 7 March, 2023; v1 submitted 17 January, 2023;
originally announced January 2023.
-
Planar Modeling and Sim-to-Real of a Tethered Multimaterial Soft Swimmer Driven by Peano-HASELs
Authors:
Stephan-Daniel Gravert,
Mike Y. Michelis,
Simon Rogler,
Dario Tscholl,
Thomas Buchner,
Robert K. Katzschmann
Abstract:
Soft robotics has the potential to revolutionize robotic locomotion, in particular, soft robotic swimmers offer a minimally invasive and adaptive solution to explore and preserve our oceans. Unfortunately, current soft robotic swimmers are vastly inferior to evolved biological swimmers, especially in terms of controllability, efficiency, maneuverability, and longevity. Additionally, the tedious it…
▽ More
Soft robotics has the potential to revolutionize robotic locomotion, in particular, soft robotic swimmers offer a minimally invasive and adaptive solution to explore and preserve our oceans. Unfortunately, current soft robotic swimmers are vastly inferior to evolved biological swimmers, especially in terms of controllability, efficiency, maneuverability, and longevity. Additionally, the tedious iterative fabrication and empirical testing required to design soft robots has hindered their optimization. In this work, we tackle this challenge by providing an efficient and straightforward pipeline for designing and fabricating soft robotic swimmers equipped with electrostatic actuation. We streamline the process to allow for rapid additive manufacturing, and show how a differentiable simulation can be used to match a simplified model to the real deformation of a robotic swimmer. We perform several experiments with the fabricated swimmer by varying the voltage and actuation frequency of the swimmer's antagonistic muscles. We show how the voltage and frequency vary the locomotion speed of the swimmer while moving in liquid oil and observe a clear optimum in forward swimming speed. The differentiable simulation model we propose has various downstream applications, such as control and shape optimization of the swimmer; optimization results can be directly mapped back to the real robot through our sim-to-real matching.
△ Less
Submitted 2 August, 2022; v1 submitted 1 August, 2022;
originally announced August 2022.
-
Fast Aquatic Swimmer Optimization with Differentiable Projective Dynamics and Neural Network Hydrodynamic Models
Authors:
Elvis Nava,
John Z. Zhang,
Mike Y. Michelis,
Tao Du,
**chuan Ma,
Benjamin F. Grewe,
Wojciech Matusik,
Robert K. Katzschmann
Abstract:
Aquatic locomotion is a classic fluid-structure interaction (FSI) problem of interest to biologists and engineers. Solving the fully coupled FSI equations for incompressible Navier-Stokes and finite elasticity is computationally expensive. Optimizing robotic swimmer design within such a system generally involves cumbersome, gradient-free procedures on top of the already costly simulation. To addre…
▽ More
Aquatic locomotion is a classic fluid-structure interaction (FSI) problem of interest to biologists and engineers. Solving the fully coupled FSI equations for incompressible Navier-Stokes and finite elasticity is computationally expensive. Optimizing robotic swimmer design within such a system generally involves cumbersome, gradient-free procedures on top of the already costly simulation. To address this challenge we present a novel, fully differentiable hybrid approach to FSI that combines a 2D direct numerical simulation for the deformable solid structure of the swimmer and a physics-constrained neural network surrogate to capture hydrodynamic effects of the fluid. For the deformable solid simulation of the swimmer's body, we use state-of-the-art techniques from the field of computer graphics to speed up the finite-element method (FEM). For the fluid simulation, we use a U-Net architecture trained with a physics-based loss function to predict the flow field at each time step. The pressure and velocity field outputs from the neural network are sampled around the boundary of our swimmer using an immersed boundary method (IBM) to compute its swimming motion accurately and efficiently. We demonstrate the computational efficiency and differentiability of our hybrid simulator on a 2D carangiform swimmer. Due to differentiability, the simulator can be used for computational design of controls for soft bodies immersed in fluids via direct gradient-based optimization.
△ Less
Submitted 22 June, 2022; v1 submitted 30 March, 2022;
originally announced April 2022.
-
Physics-constrained Unsupervised Learning of Partial Differential Equations using Meshes
Authors:
Mike Y. Michelis,
Robert K. Katzschmann
Abstract:
Enhancing neural networks with knowledge of physical equations has become an efficient way of solving various physics problems, from fluid flow to electromagnetism. Graph neural networks show promise in accurately representing irregularly meshed objects and learning their dynamics, but have so far required supervision through large datasets. In this work, we represent meshes naturally as graphs, p…
▽ More
Enhancing neural networks with knowledge of physical equations has become an efficient way of solving various physics problems, from fluid flow to electromagnetism. Graph neural networks show promise in accurately representing irregularly meshed objects and learning their dynamics, but have so far required supervision through large datasets. In this work, we represent meshes naturally as graphs, process these using Graph Networks, and formulate our physics-based loss to provide an unsupervised learning framework for partial differential equations (PDE). We quantitatively compare our results to a classical numerical PDE solver, and show that our computationally efficient approach can be used as an interactive PDE solver that is adjusting boundary conditions in real-time and remains sufficiently close to the baseline solution. Our inherently differentiable framework will enable the application of PDE solvers in interactive settings, such as model-based control of soft-body deformations, or in gradient-based optimization methods that require a fully differentiable pipeline.
△ Less
Submitted 30 March, 2022;
originally announced March 2022.
-
On Linear Interpolation in the Latent Space of Deep Generative Models
Authors:
Mike Yan Michelis,
Quentin Becker
Abstract:
The underlying geometrical structure of the latent space in deep generative models is in most cases not Euclidean, which may lead to biases when comparing interpolation capabilities of two models. Smoothness and plausibility of linear interpolations in latent space are associated with the quality of the underlying generative model. In this paper, we show that not all such interpolations are compar…
▽ More
The underlying geometrical structure of the latent space in deep generative models is in most cases not Euclidean, which may lead to biases when comparing interpolation capabilities of two models. Smoothness and plausibility of linear interpolations in latent space are associated with the quality of the underlying generative model. In this paper, we show that not all such interpolations are comparable as they can deviate arbitrarily from the shortest interpolation curve given by the geodesic. This deviation is revealed by computing curve lengths with the pull-back metric of the generative model, finding shorter curves than the straight line between endpoints, and measuring a non-zero relative length improvement on this straight line. This leads to a strategy to compare linear interpolations across two generative models. We also show the effect and importance of choosing an appropriate output space for computing shorter curves. For this computation we derive an extension of the pull-back metric.
△ Less
Submitted 8 May, 2021;
originally announced May 2021.
