-
Solvability of the Inverse Optimal Control problem based on the minimum principle
Authors:
Afreen Islam,
Guido Herrmann,
Joaquin Carrasco
Abstract:
In this paper, the solvability of the Inverse Optimal Control (IOC) problem based on two existing minimum principal methods, is analysed. The aim of this work is to answer the question regarding what kinds of trajectories, that is depending on the initial conditions of the closed-loop system and system dynamics, of the original optimal control problem, will result in the recovery of the true weigh…
▽ More
In this paper, the solvability of the Inverse Optimal Control (IOC) problem based on two existing minimum principal methods, is analysed. The aim of this work is to answer the question regarding what kinds of trajectories, that is depending on the initial conditions of the closed-loop system and system dynamics, of the original optimal control problem, will result in the recovery of the true weights of the reward function for both the soft and the hard-constrained methods [1], [2]. Analytical conditions are provided which allow to verify if a trajectory is sufficiently conditioned, that is, holds sufficient information to recover the true weights of an optimal control problem. It was found that the open-loop system of the original optimal problem has a stronger influence on the solvability of the Inverse Optimal Control problem for the hard-constrained method as compared to the soft-constrained method. These analytical results were validated via simulation.
△ Less
Submitted 14 March, 2024;
originally announced March 2024.
-
Towards Semi-Autonomous Robotic Arm Manipulation Operator Intention Detection from Forces Feedback
Authors:
Abdullah Alharthi,
Ozan Tokatli,
Erwin Lopez,
Guido Herrmann
Abstract:
In harsh environments such as those found in nuclear facilities, the use of robotic systems is crucial for performing tasks that would otherwise require human intervention. This is done to minimize the risk of human exposure to dangerous levels of radiation, which can have severe consequences for health and even be fatal. However, the telemanipulation systems employed in these environments are bec…
▽ More
In harsh environments such as those found in nuclear facilities, the use of robotic systems is crucial for performing tasks that would otherwise require human intervention. This is done to minimize the risk of human exposure to dangerous levels of radiation, which can have severe consequences for health and even be fatal. However, the telemanipulation systems employed in these environments are becoming increasingly intricate, relying heavily on sophisticated control methods and local master devices. Consequently, the cognitive burden on operators during labor-intensive tasks is growing. To tackle this challenge, operator intention detection based on task learning can greatly enhance the performance of robotic tasks while reducing the reliance on human effort in teleoperation, particularly in a glovebox environment. By accurately predicting the operator's intentions, the robot can carry out tasks more efficiently and effectively, with minimal input from the operator. In this regard, we propose the utilization of Convolutional Neural Networks, a machine learning approach, to learn and forecast the operator's intentions using raw force feedback spatiotemporal data. Through our experimental study on glovebox tasks for nuclear applications, such as radiation survey and object gras**, we have achieved promising outcomes. Our approach holds the potential to enhance the safety and efficiency of robotic systems in harsh environments, thus diminishing the risk of human exposure to radiation while simultaneously improving the precision and speed of robotic operations.
△ Less
Submitted 19 January, 2024;
originally announced February 2024.
-
Sim-and-Real Reinforcement Learning for Manipulation: A Consensus-based Approach
Authors:
Wenxing Liu,
Hanlin Niu,
Wei Pan,
Guido Herrmann,
Joaquin Carrasco
Abstract:
Sim-and-real training is a promising alternative to sim-to-real training for robot manipulations. However, the current sim-and-real training is neither efficient, i.e., slow convergence to the optimal policy, nor effective, i.e., sizeable real-world robot data. Given limited time and hardware budgets, the performance of sim-and-real training is not satisfactory. In this paper, we propose a Consens…
▽ More
Sim-and-real training is a promising alternative to sim-to-real training for robot manipulations. However, the current sim-and-real training is neither efficient, i.e., slow convergence to the optimal policy, nor effective, i.e., sizeable real-world robot data. Given limited time and hardware budgets, the performance of sim-and-real training is not satisfactory. In this paper, we propose a Consensus-based Sim-And-Real deep reinforcement learning algorithm (CSAR) for manipulator pick-and-place tasks, which shows comparable performance in both sim-and-real worlds. In this algorithm, we train the agents in simulators and the real world to get the optimal policies for both sim-and-real worlds. We found two interesting phenomenons: (1) Best policy in simulation is not the best for sim-and-real training. (2) The more simulation agents, the better sim-and-real training. The experimental video is available at: https://youtu.be/mcHJtNIsTEQ.
△ Less
Submitted 17 September, 2023; v1 submitted 26 February, 2023;
originally announced February 2023.
-
Drivers' Manoeuvre Modelling and Prediction for Safe HRI
Authors:
Erwin Jose Lopez Pulgarin,
Guido Herrmann,
Ute Leonards
Abstract:
As autonomous machines such as robots and vehicles start performing tasks involving human users, ensuring a safe interaction between them becomes an important issue. Translating methods from human-robot interaction (HRI) studies to the interaction between humans and other highly complex machines (e.g. semi-autonomous vehicles) could help advance the use of those machines in scenarios requiring hum…
▽ More
As autonomous machines such as robots and vehicles start performing tasks involving human users, ensuring a safe interaction between them becomes an important issue. Translating methods from human-robot interaction (HRI) studies to the interaction between humans and other highly complex machines (e.g. semi-autonomous vehicles) could help advance the use of those machines in scenarios requiring human interaction. One method involves understanding human intentions and decision-making to estimate the human's present and near-future actions whilst interacting with a robot. This idea originates from the psychological concept of Theory of Mind, which has been broadly explored for robotics and recently for autonomous and semi-autonomous vehicles. In this work, we explored how to predict human intentions before an action is performed by combining data from human-motion, vehicle-state and human inputs (e.g. steering wheel, pedals). A data-driven approach based on Recurrent Neural Network models was used to classify the current driving manoeuvre and to predict the future manoeuvre to be performed. A state-transition model was used with a fixed set of manoeuvres to label data recorded during the trials for real-time applications. Models were trained and tested using drivers of different seat preferences, driving expertise and arm-length; precision and recall metrics over 95% for manoeuvre identification and 86% for manoeuvre prediction were achieved, with prediction time-windows of up to 1 second for both known and unknown test subjects. Compared to our previous results, performance improved and manoeuvre prediction was possible for unknown test subjects without knowing the current manoeuvre.
△ Less
Submitted 3 June, 2021;
originally announced June 2021.
-
The Complex of Hypersurfaces in a Homology Class
Authors:
Gerrit Herrmann,
José Pedro Quintanilha
Abstract:
For a compact oriented smooth $n$-manifold $M$ and a codimension-$1$ homology class $φ\in \operatorname{H}_{n-1}(M, \partial M)$, we investigate a simplicial complex $\mathcal{S}^\dagger(M, φ)$ relating the properly embedded hypersurfaces in $M$ representing $φ$. Its definition is akin to that of other classical complexes, such as the curve complex of a surface or the Kakimizu complex of a knot, w…
▽ More
For a compact oriented smooth $n$-manifold $M$ and a codimension-$1$ homology class $φ\in \operatorname{H}_{n-1}(M, \partial M)$, we investigate a simplicial complex $\mathcal{S}^\dagger(M, φ)$ relating the properly embedded hypersurfaces in $M$ representing $φ$. Its definition is akin to that of other classical complexes, such as the curve complex of a surface or the Kakimizu complex of a knot, with the difference that hypersurfaces are not taken up to isotopy.
We prove that $\mathcal{S}^\dagger(M, φ)$ is connected and simply connected in every dimension $n$. We also show connectedness of a similar complex $\mathcal{T}^\dagger(M, φ)$ adapted to the $3$-dimensional case, where only Thurston norm-realizing surfaces are considered. The connectedness results are transported to the complexes $\mathcal{S}(M, φ), \mathcal{T}(M, φ)$ where hypersurfaces are taken up to isotopy, and for $n=2$ the simple connectedness result carries over as well. We also briefly discuss extensions to a context studied by Turaev, where regular graphs in $2$-complexes are used to represent $1$-dimensional cohomology classes.
We finish with two applications: we give an alternative proof of the fact that all Seifert surfaces for a fixed knot in a rational homology sphere are tube-equivalent, and we use connectedness of $\mathcal{T}^\dagger(M, φ)$ to define a new $\ell^2$-invariant of $2$-dimensional homology classes in irreducible and boundary-irreducible oriented compact connected $3$-manifolds with empty or toroidal boundary.
△ Less
Submitted 22 February, 2022; v1 submitted 1 July, 2020;
originally announced July 2020.
-
The leading coefficient of the $L^2$-Alexander torsion
Authors:
Fathi Ben Aribi,
Stefan Friedl,
Gerrit Herrmann
Abstract:
We give upper and lower bounds on the leading coefficients of the $L^2$-Alexander torsions of a $3$-manifold $M$ in terms of hyperbolic volumes and of relative $L^2$-torsions of sutured manifolds obtained by cutting $M$ along certain surfaces.
We prove that for numerous families of knot exteriors the lower and upper bounds are equal, notably for exteriors of 2-bridge knots. In particular we comp…
▽ More
We give upper and lower bounds on the leading coefficients of the $L^2$-Alexander torsions of a $3$-manifold $M$ in terms of hyperbolic volumes and of relative $L^2$-torsions of sutured manifolds obtained by cutting $M$ along certain surfaces.
We prove that for numerous families of knot exteriors the lower and upper bounds are equal, notably for exteriors of 2-bridge knots. In particular we compute the leading coefficient explicitly for 2-bridge knots.
△ Less
Submitted 6 May, 2021; v1 submitted 28 June, 2018;
originally announced June 2018.
-
Sutured manifolds and $L^2$-Betti numbers
Authors:
Gerrit Herrmann
Abstract:
Using the virtual fibering theorem of Agol we show that a sutured 3-manifold $(M, R_+,R_-,γ)$ is taut if and only if the $\ell^2$-Betti numbers of the pair $(M,R_-)$ are zero. As an application we can characterize Thurston norm minimizing surfaces in a 3-manifold $N$ with empty or toroidal boundary by the vanishing of certain $\ell^2$-Betti numbers.
Using the virtual fibering theorem of Agol we show that a sutured 3-manifold $(M, R_+,R_-,γ)$ is taut if and only if the $\ell^2$-Betti numbers of the pair $(M,R_-)$ are zero. As an application we can characterize Thurston norm minimizing surfaces in a 3-manifold $N$ with empty or toroidal boundary by the vanishing of certain $\ell^2$-Betti numbers.
△ Less
Submitted 16 February, 2020; v1 submitted 25 April, 2018;
originally announced April 2018.
-
Torsion in the homology of finite covers of 3-manifolds
Authors:
Stefan Friedl,
Gerrit Herrmann
Abstract:
Let $N$ be a prime 3-manifold that is not a closed graph manifold. Building on a result of Hongbin Sun and using a result of Asaf Hadari we show that for every $k\in\Bbb{N}$ there exists a finite cover $\tilde{N}$ of $N$ such that $|\operatorname{Tor} H_1(\tilde{N};\Bbb{Z})|>k$.
Let $N$ be a prime 3-manifold that is not a closed graph manifold. Building on a result of Hongbin Sun and using a result of Asaf Hadari we show that for every $k\in\Bbb{N}$ there exists a finite cover $\tilde{N}$ of $N$ such that $|\operatorname{Tor} H_1(\tilde{N};\Bbb{Z})|>k$.
△ Less
Submitted 24 October, 2017;
originally announced October 2017.
-
Linking forms revisited
Authors:
Anthony Conway,
Stefan Friedl,
Gerrit Herrmann
Abstract:
We show that the $\mathbb{Q}/\mathbb{Z}$-valued linking forms on rational homology spheres are (anti-) symmetric and we compute the linking form of a 3-dimensional rational homology sphere in terms of a Heegaard splitting. Both results have been known to a larger or lesser degree, but it is difficult to find rigorous down-to-earth proofs in the literature.
We show that the $\mathbb{Q}/\mathbb{Z}$-valued linking forms on rational homology spheres are (anti-) symmetric and we compute the linking form of a 3-dimensional rational homology sphere in terms of a Heegaard splitting. Both results have been known to a larger or lesser degree, but it is difficult to find rigorous down-to-earth proofs in the literature.
△ Less
Submitted 26 February, 2018; v1 submitted 12 August, 2017;
originally announced August 2017.
-
Net Shape 3D Printed NdFeB Permanent Magnet
Authors:
J. Jacimovic,
F. Binda,
L. G. Herrmann,
F. Greuter,
J. Genta,
M. Calvo,
T. Tomse,
R. A. Simon
Abstract:
For two decades, NdFeB based magnets have been a critical component in a range of electrical devices engaged in energy production and conversion. The magnet shape and the internal microstructure of the selected NdFeB grade govern their efficiency and size. However, stricter requirements on device efficiency call for better performing magnets preferably with novel functionality not achievable today…
▽ More
For two decades, NdFeB based magnets have been a critical component in a range of electrical devices engaged in energy production and conversion. The magnet shape and the internal microstructure of the selected NdFeB grade govern their efficiency and size. However, stricter requirements on device efficiency call for better performing magnets preferably with novel functionality not achievable today. Here we use 3D metal printing by Selective Laser Melting to fabricate dense net shape permanent magnets based on NdFeB that exhibit high magnetic performance. Evidence is provided that the internal microstructure, not achievable by traditional manufacturing means, is the origin of the solid magnetic properties. The freedom in magnet body shape and size that ranges from the millimeter to tens of centimeter scale opens up a design freedom that could be a catalyzer for the next generation of electrical devices.
△ Less
Submitted 15 November, 2016;
originally announced November 2016.
-
The $L^2$-Alexander torsion for Seifert fiber spaces
Authors:
Gerrit Herrmann
Abstract:
We calculate the $L^2$-Alexander torsion for Seifert fiber spaces and graph manifolds in terms of the Thurston norm.
We calculate the $L^2$-Alexander torsion for Seifert fiber spaces and graph manifolds in terms of the Thurston norm.
△ Less
Submitted 30 March, 2016; v1 submitted 28 February, 2016;
originally announced February 2016.
-
Self-organization of weighted networks for optimal synchronizability
Authors:
Louis Kempton,
Guido Herrmann,
Mario di Bernardo
Abstract:
We show that a network can self-organize its structure in a completely distributed manner in order to optimize its synchronizability whilst satisfying the local constraints: non-negativity of edge weights, and maximum weighted degree of nodes. A novel multilayer approach is presented which uses a distributed strategy to estimate two spectral functions of the graph Laplacian, the algebraic connecti…
▽ More
We show that a network can self-organize its structure in a completely distributed manner in order to optimize its synchronizability whilst satisfying the local constraints: non-negativity of edge weights, and maximum weighted degree of nodes. A novel multilayer approach is presented which uses a distributed strategy to estimate two spectral functions of the graph Laplacian, the algebraic connectivity $λ_2$ and the eigenratio $r = λ_n / λ_2$ . These local estimates are then used to evolve the edge weights so as to maximize $λ_2$, or minimize $r$ and, hence, achieve an optimal structure.
△ Less
Submitted 27 May, 2015;
originally announced May 2015.
-
Spectroscopy of non-local superconducting correlations in a double quantum dot
Authors:
L. G. Herrmann,
P. Burset,
W. J. Herrera,
F. Portier,
P. Roche,
C. Strunk,
A. Levy Yeyati,
T. Kontos
Abstract:
We investigate non-linear transport in a double quantum dot connected to two normal electrodes and a central superconducting finger. By this means, we perform a transport spectroscopy of such a system which implements a Cooper pair splitter. The non-linear conductance exhibits strong subgap features which can be associated with the coherence of the injected Cooper pairs. Our findings are well acco…
▽ More
We investigate non-linear transport in a double quantum dot connected to two normal electrodes and a central superconducting finger. By this means, we perform a transport spectroscopy of such a system which implements a Cooper pair splitter. The non-linear conductance exhibits strong subgap features which can be associated with the coherence of the injected Cooper pairs. Our findings are well accounted for by the recently developed microscopic theory of Cooper pairs splitters made in SWNTs.
△ Less
Submitted 9 May, 2012;
originally announced May 2012.
-
Noisy Kondo impurities
Authors:
T. Delattre,
C. Feuillet-Palma,
L. G. Herrmann,
P. Morfin,
J. -M. Berroir,
G. Fève,
B. Plaçais,
D. C. Glattli,
M. -S. Choi,
C. Mora,
T. Kontos
Abstract:
The anti-ferromagnetic coupling of a magnetic impurity carrying a spin with the conduction electrons spins of a host metal is the basic mechanism responsible for the increase of the resistance of an alloy such as Cu${}_{0.998}$Fe${}_{0.002}$ at low temperature, as originally suggested by Kondo . This coupling has emerged as a very generic property of localized electronic states coupled to a contin…
▽ More
The anti-ferromagnetic coupling of a magnetic impurity carrying a spin with the conduction electrons spins of a host metal is the basic mechanism responsible for the increase of the resistance of an alloy such as Cu${}_{0.998}$Fe${}_{0.002}$ at low temperature, as originally suggested by Kondo . This coupling has emerged as a very generic property of localized electronic states coupled to a continuum . The possibility to design artificial controllable magnetic impurities in nanoscopic conductors has opened a path to study this many body phenomenon in unusual situations as compared to the initial one and, in particular, in out of equilibrium situations. So far, measurements have focused on the average current. Here, we report on \textit{current fluctuations} (noise) measurements in artificial Kondo impurities made in carbon nanotube devices. We find a striking enhancement of the current noise within the Kondo resonance, in contradiction with simple non-interacting theories. Our findings provide a test bench for one of the most important many-body theories of condensed matter in out of equilibrium situations and shed light on the noise properties of highly conductive molecular devices.
△ Less
Submitted 22 October, 2010;
originally announced October 2010.
-
Carbon Nanotubes as Cooper Pair Beam Splitters
Authors:
L. G. Herrmann,
F. Portier,
P. Roche,
A. Levy Yeyati,
T. Kontos,
C. Strunk
Abstract:
We report on conductance measurements in carbon nanotube based double quantum dots connected to two normal electrodes and a central superconducting finger. By operating our devices as beam splitters, we provide evidence for Crossed Andreev Reflections \textit{tunable in situ}. This opens an avenue to more sophisticated quantum optics-like experiments with spin entangled electrons.
We report on conductance measurements in carbon nanotube based double quantum dots connected to two normal electrodes and a central superconducting finger. By operating our devices as beam splitters, we provide evidence for Crossed Andreev Reflections \textit{tunable in situ}. This opens an avenue to more sophisticated quantum optics-like experiments with spin entangled electrons.
△ Less
Submitted 17 September, 2009;
originally announced September 2009.
-
Shot noise in carbon nanotube based Fabry-Perot interferometers
Authors:
L. G. Herrmann,
T. Delattre,
P. Morfin,
J. -M. Berroir,
B. Plaçais,
D. C. Glattli,
T. Kontos
Abstract:
We report on shot noise measurements in carbon nanotube based Fabry-Perot electronic interferometers. As a consequence of quantum interferences, the noise power spectral density oscillates as a function of the voltage applied to the gate electrode. The quantum shot noise theory accounts for the data quantitatively. It allows to confirm the existence of two nearly degenerate orbitals. At resonanc…
▽ More
We report on shot noise measurements in carbon nanotube based Fabry-Perot electronic interferometers. As a consequence of quantum interferences, the noise power spectral density oscillates as a function of the voltage applied to the gate electrode. The quantum shot noise theory accounts for the data quantitatively. It allows to confirm the existence of two nearly degenerate orbitals. At resonance, the transmission of the nanotube approaches unity, and the nanotube becomes noiseless, as observed in quantum point contacts. In this weak backscattering regime, the dependence of the noise on the backscattering current is found weaker than expected, pointing either to electron-electron interactions or to weak decoherence.
△ Less
Submitted 5 March, 2007;
originally announced March 2007.
