Deep reinforcement learning (DRL) has been successfully used to solve various robotic manipulation tasks. However, most of the existing works do not address the issue of control stability. This is in sharp contrast to the control theory community where the well-established norm is to prove stability whenever a control law is synthesized. What makes traditional stability analysis difficult for DRL are the uninterpretable nature of the neural network policies and unknown system dynamics. In this work, unconditional stability is obtained by deriving an interpretable deep policy structure based on the energy shaping control of Lagrangian systems. Then, stability during physical interaction with an unknown environment is established based on passivity. The result is a stability guaranteeing DRL in a model-free framework that is general enough for contact-rich manipulation tasks. With an experiment on a peg-in-hole task, we demonstrate, to the best of our knowledge, the first DRL with stability guarantee on a real robotic manipulator.

Reinforcement Learning (RL) of robotic manipu-lation skills, despite its impressive successes, stands to benefitfrom incorporating domain knowledge from control theory. Oneof the most important properties that is of interest is controlstability. Ideally, one would like to achieve stability guaranteeswhile staying within the framework of state-of-the-art deepRL algorithms. Such a solution does not exist in general,especially one that scales to complex manipulation tasks. Wecontribute towards closing this gap by introducing normalizing-flow control structure, that can be deployed in any latest deepRL algorithms. While stable exploration is not guaranteed,our method is designed to ultimately produce deterministiccontrollers with provable stability. In addition to demonstratingour method on challenging contact-rich manipulation tasks, wealso show that it is possible to achieve considerable explorationefficiency–reduced state space coverage and actuation efforts–without losing learning efficiency.

Given the need for modern researchers to produce open, reproducible scientific output, the lack of standards and best practices for sharing data and workflows used to produce and analyze molecular dynamics (MD) simulations has become an important issue in the field. There are now multiple well-established packages to perform molecular dynamics simulations, often highly tuned for exploiting specific classes of hardware, each with strong communities surrounding them, but with very limited interoperability/transferability options. Thus, the choice of the software package often dictates the workflow for both simulation production and analysis. The level of detail in documenting the workflows and analysis code varies greatly in published work, hindering reproducibility of the reported results and the ability for other researchers to build on these studies. An increasing number of researchers are motivated to make their data available, but many challenges remain in order to effectively share and reuse simulation data. To discuss these and other issues related to best practices in the field in general, we organized a workshop in November 2018 (https://bioexcel.eu/events/workshop-on-sharing-data-from-molecular-simulations/). Here, we present a brief overview of this workshop and topics discussed. We hope this effort will spark further conversation in the MD community to pave the way toward more open, interoperable, and reproducible outputs coming from research studies using MD simulations.

This paper addresses a formation problem for a network of autonomous agents with second-order dynamics and bounded disturbances. Coordination is achieved by having the agents asynchronously upload (download) data to (from) a shared repository, rather than directly exchanging data with other agents. Well-posedness of the closed-loop system is demonstrated by showing that there exists a lower bound for the time interval between two consecutive agent accesses to the repository. Numerical simulations corroborate the theoretical results.

We consider the problem of approximating flow functions of continuous-time dynamical systems with inputs. It is well-known that continuous-time recurrent neural networks are universal approximators of this type of system. In this paper, we prove that an architecture based on discrete-time recurrent neural networks universally approximates flows of continuous-time dynamical systems with inputs. The required assumptions are shown to hold for systems whose dynamics are well-behaved ordinary differential equations and with practically relevant classes of input signals. This enables the use of off-the-shelf solutions for learning such flow functions in continuous-time from sampled trajectory data.

High-Performance Computing (HPC) has become an important scientific driver. A wide variety of research ranging for example from drug design to climate modelling is nowadays performed in HPC systems. Furthermore, the tremendous computer power of such HPC systems allows scientists to simulate problems that were unimaginable a few years ago. However, the continuous increase in size and complexity of HPC systems is turning the development of efficient parallel software into a difficult task. Therefore, the use of per- formance monitoring and analysis is a must in order to unveil inefficiencies in parallel software. Nevertheless, performance tools also face challenges as a result of the size of HPC systems, for example, coping with huge amounts of performance data generated.

In this thesis, we propose a new model for performance characterisation of MPI applications that tackles the challenge of big performance data sets. Our approach uses Event Flow Graphs to balance the scalability of profiling techniques (generating performance reports with aggregated metrics) with the richness of information of tracing methods (generating files with sequences of time-stamped events). In other words, graphs allow to encode ordered se- quences of events without storing the whole sequence of such events, and therefore, they need much less memory and disk space, and are more scal- able. We demonstrate in this thesis how our Event Flow Graph model can be used as a trace compression method. Furthermore, we propose a method to automatically detect the structure of MPI applications using our Event Flow Graphs. This knowledge can afterwards be used to collect performance data in a smarter way, reducing for example the amount of redundant data collected. Finally, we demonstrate that our graphs can be used beyond trace compression and automatic analysis of performance data. We propose a new methodology to use Event Flow Graphs in the task of visual performance data exploration.

In addition to the Event Flow Graph model, we also explore in this thesis the design and use of performance data introspection frameworks. Future HPC systems will be very dynamic environments providing extreme levels of parallelism, but with energy constraints, considerable resource sharing, and heterogeneous hardware. Thus, the use of real-time performance data to or- chestrate program execution in such a complex and dynamic environment will be a necessity. This thesis presents two different performance data introspec- tion frameworks that we have implemented. These introspection frameworks are easy to use, and provide performance data in real time with very low overhead. We demonstrate, among other things, how our approach can be used to reduce in real time the energy consumed by the system.

The approaches proposed in this thesis have been validated in different HPC systems using multiple scientific kernels as well as real scientific applica- tions. The experiments show that our approaches in performance character- isation and performance data introspection are not intrusive at all, and can be a valuable contribution to help in the performance monitoring of future HPC systems.

We design and develop a new Particle-in-Cell (PIC) method for plasma simulations using Deep-Learning (DL) to calculate the electric field from the electron phase space. We train a Multilayer Perceptron (MLP) and a Convolutional Neural Network (CNN) to solve the two-stream instability test. We verify that the DL-based MLP PIC method produces the correct results using the two-stream instability: the DL-based PIC provides the expected growth rate of the two-stream instability. The DL-based PIC does not conserve the total energy and momentum. However, the DL-based PIC method is stable against the cold-beam instability, affecting traditional PIC methods. This work shows that integrating DL technologies into traditional computational methods is a viable approach for developing next-generation PIC algorithms.

In this paper we consider a mobile platform controlled by two entities; an autonomousagent and a human user. The human aims for the mobile platform to complete a task, whichwe will denote as the human task, and will impose a control input accordingly, while not beingaware of any other tasks the system should or must execute. The autonomous agent will in turnplan its control input taking in consideration all safety requirements which must be met, sometask which should be completed as much as possible (denoted as the robot task), as well aswhat it believes the human task is based on previous human control input. A framework for theautonomous agent and a mixed initiative controller are designed to guarantee the satisfaction ofthe safety requirements while both the human and robot tasks are violated as little as possible.The framework includes an estimation algorithm of the human task which will improve witheach cycle, eventually converging to a task which is similar to the actual human task. Hence, theautonomous agent will eventually be able to find the optimal plan considering all tasks and thehuman will have no need to interfere again. The process is illustrated with a simulated example

The XENON experiment is looking for non-baryonic particle dark matter in the universe. The setup is a dual phase time projection chamber (TPC) filled with 3200 kg of ultra-pure liquid xenon. The setup is operated at the Laboratori Nazionali del Gran Sasso (LNGS) in Italy. We present a full overview of the computing scheme for data distribution and job management in XENON1T. The software package Rucio, which is developed by the ATLAS collaboration, facilitates data handling on Open Science Grid (OSG) and European Grid Infrastructure (EGI) storage systems. A tape copy at the Centre for High Performance Computing (PDC) is managed by the Tivoli Storage Manager (TSM). Data reduction and Monte Carlo production are handled by CI Connect which is integrated into the OSG network. The job submission system connects resources at the EGI, OSG, SDSC's Comet, and the campus HPC resources for distributed computing. The previous success in the XENON1T computing scheme is also the starting point for its successor experiment XENONnT, which starts to take data in autumn 2019.

Service assurance for cloud applications is a challenging task and is an active area of research for academia and industry. One promising approach is to utilize machine learning for service quality prediction and fault detection so that suitable mitigation actions can be executed. In our previous work, we have shown how to predict service-level metrics in real-time just from operational data gathered at the server side. This gives the service provider early indications on whether the platform can support the current load demand. This paper provides the logical next step where we extend our work by proposing an automated detection and diagnostic capability for the performance faults manifesting themselves in cloud and datacenter environments. This is a crucial task to maintain the smooth operation of running services and minimizing downtime. We demonstrate the effectiveness of our approach which exploits the interpretative capabilities of Self- Organizing Maps (SOMs) to automatically detect and localize different performance faults for cloud services.

Collective operations are commonly used in various parts of scientific applications. Especially in strong scaling scenarios collective operations can negatively impact the overall applications performance: while the load per rank here decreases with increasing core counts, time spent in e.g. barrier operations will increase logarithmically with the core count. In this article, we develop novel algorithmic solutions for collective operations such as Allreduce and Allgather(V)-by leveraging notified communication in shared windows. To this end, we have developed an extension of GASPI which enables all ranks participating in a shared window to observe the entire notified communication targeted at the window. By exploring benefits of this extension, we deliver high performing implementations of Allreduce and Allgather(V) on Intel and Cray clusters. These implementations clearly achieve 2x-4x performance improvements compared to the best performing MPI implementations for various data distributions.

SYCL is a cross-platform, royalty-free standard for programming a wide range of hardware accelerators. It is a powerful and convenient way to write standard C++ 17 code that can take full advantage of available devices. There are already multiple SYCL implementations targeting a wide range of platforms, from embedded to HPC clusters. Since several implementations can target the same hardware, application developers and users must know how to choose the most fitting runtime for their needs. In this talk, we will compare the runtime performance of two major SYCL runtimes targeting GPUs, oneAPI DPC++ and Open SYCL [3], to the native implementations for the purposes of GROMACS, a high-performance molecular dynamics engine.Molecular dynamics (MD) applications were one of the earliest adopters of GPU acceleration, with force calculations being an obvious target for offloading. It is an iterative algorithm where, in its most basic form, on each step, forces acting between particles are computed, and then the equations of motions are integrated. As the computational power of the GPUs grew, the strong scaling problem became apparent: the biophysical systems modeled with molecular dynamics typically have fixed sizes, and the goal is to perform more time steps, each taking less than a millisecond of wall time. This places high demands on the underlying GPU framework, requiring it to efficiently schedule multiple small tasks with minimal overhead, allowing to achieve overlap between CPU and GPU work for large systems and allowing to keep GPU occupied for smaller systems. Another requirement is the ability of application developers to have control over the scheduling to optimize for external dependencies, such as MPI communication.GROMACS is a widely-used MD engine, supporting a wide range of hardware and software platforms, from laptops to the largest supercomputers [1]. Portability and performance across multiple architectures have always been one of the primary goals of the project, necessary to keep the code not only efficient but also maintainable. The initial support for NVIDIA accelerators, using CUDA, was added to GROMACS in 2010. Since then, heterogeneous parallelization has been a major target for performance optimization, not limited to NVIDIA devices but later adding support for GPUs of other vendors, as well as Xeon Phi accelerators. GROMACS initially adopted SYCL in its 2021 release to replace its previous GPU portability layer, OpenCL [2]. In further releases, the number of offloading modes supported by the SYCL backend steadily increased. As of GROMACS 2023, SYCL support in GROMACS achieved near feature parity with CUDA while allowing the use of a single code to target the GPUs of all three major vendors with minimal specialization.While this clearly supports the portability promise of modern SYCL implementations, the performance of such portable code remains an open question, especially given the strict requirements of MD algorithms. In this talk, we compare the performance of GROMACS across a wide range of system sizes when using oneAPI DPC++ and Open SYCL runtimes on high-performance NVIDIA, AMD, and Intel GPUs. Besides the analysis of individual kernel performance, we focus on the runtime overhead and the efficiency of task scheduling when compared to a highly optimized implementation using the native frameworks and discuss the possible sources of suboptimal performance and the amount of vendor-specific code branches, such as intrinsics or workarounds for compiler bugs, required to achieve the optimal performance.

GROMACS is an open-source, high-performance molecular dynamics (MD) package primarily used for biomolecular simulations, accounting for 5% of HPC utilization worldwide. Due to the extreme computing needs of MD, significant efforts are invested in improving the performance and scalability of simulations. Target hardware ranges from supercomputers to laptops of individual researchers and volunteers of distributed computing projects such as Folding@Home. The code has been designed both for portability and performance by explicitly adapting algorithms to SIMD and data-parallel processors. A SIMD intrinsic abstraction layer provides high CPU performance. Explicit GPU acceleration has long used CUDA to target NVIDIA devices and OpenCL for AMD/Intel devices. In this talk, we discuss the experiences and challenges of adding support for the SYCL platform into the established GROMACS codebase and share experiences and considerations in porting and optimization. While OpenCL offers the benefits of using the same code to target different hardware, it suffers from several drawbacks that add significant development friction. Its separate-source model leads to code duplication and makes changes complicated. The need to use C99 for kernels, while the rest of the codebase uses C++17, exacerbates these issues. Another problem is that OpenCL, while supported by most GPU vendors, is never the main framework and thus is not getting the primary support or tuning efforts. SYCL alleviates many of these issues, employing a single-source model based on the modern C++ standard. In addition to being the primary platform for Intel GPUs, the possibility to target AMD and NVIDIA GPUs through other implementations (e.g., hipSYCL) might make it possible to reduce the number of separate GPU ports that have to be maintained. Some design differences from OpenCL, such as flow directed acyclic graphs (DAGs) instead of in-order queues, made it necessary to reconsider the GROMACS's task scheduling approach and architectural choices in the GPU backend. Additionally, supporting multiple GPU platforms presents a challenge of balancing performance (low-level and hardware-specific code) and maintainability (more generalization and code-reuse). We will discuss the limitations of the existing codebase and interoperability layers with regards to adding the new platform; the compute performance and latency comparisons; code quality considerations; and the issues we encountered with SYCL implementations tested. Finally, we will discuss our goals for the next release cycle for the SYCL backend and the overall architecture of GPU acceleration code in GROMACS.

GROMACS is a widely-used molecular dynamics software package with a focus on performance, portability, and maintainability across a broad range of platforms. Thanks to its early algorithmic redesign and flexible heterogeneous parallelization, GROMACS has successfully harnessed GPU accelerators for more than a decade.With the diversification of accelerator platforms in HPC and no obvious choice for a well-suited multi-vendor programming model, the GROMACS project found itself at a crossroads. The performance and portability requirements, as well as a strong preference for a standards-based programming model, motivated our choice to use SYCL for production on both new HPC GPU platforms: AMD and Intel.Since the GROMACS 2022 release, the SYCL backend has been the primary means to target AMD GPUs in preparation for exascale HPC architectures like LUMI and Frontier.SYCL is a cross-platform, royalty-free, C++17-based standard for programming hardware accelerators, from embedded to HPC.It allows using the same code to target GPUs from all three major vendors with minimal specialization, which offers major portability benefits.While SYCL implementations build on native compilers and runtimes, whether such an approach is performant is not immediately evident.Biomolecular simulations have challenging performance characteristics: latency sensitivity, the need for strong scaling, and typical iteration times as short as hundreds of microseconds. Hence, obtaining good performance across the range of problem sizes and scaling regimes is particularly challenging.Here, we share the results of our work on readying GROMACS for AMD GPU platforms using SYCL,and demonstrate performance on Cray EX235a machines with MI250X accelerators. Our findings illustrate that portability is possible without major performance compromises.We provide a detailed analysis of node-level kernel and runtime performance with the aim of sharing best practices with the HPC community on using SYCL as a performance-portable GPU framework.

The energy sector is going through a large transformation due to public demands of renewable energy sources. However, a major issue is that these energy sources are intermittent. If designed correctly, district heating systems can naturally contain energy storing units, for example by storing heat in the isolated pipes that make up the heat grid. Additionally, this makes it easier to reuse and transport already generated heat to other users. This paper proposes a mathematical model of such a grid, where excess energy can be retracted from one user and distributed to other users using a network of heat pumps. In some cases, one can balance residual heat production with the heat consumption, temporarily eliminating the need for a centralized heating plant. Existence conditions for stable steady states of such a network with general topology are given. Finally, energy optimal stable steady states are obtained through convex optimization. 

Consider challenging sim-to-real cases lacking high-fidelity simulators and allowing only 10-20 hardware trials. This work shows that even imprecise simulation can be beneficial if used to build transfer-aware representations.

First, the thesis introduces an informed kernel that embeds the space of simulated trajectories into a lower-dimensional space of latent paths. It uses a sequential variational autoencoder (sVAE) to handle large-scale training from simulated data. Its modular design enables quick adaptation when used for Bayesian optimization (BO) on hardware. The thesis and the included publications demonstrate that this approach works for different areas of robotics: locomotion and manipulation. Furthermore, a variant of BO that ensures recovery from negative transfer when using corrupted kernels is introduced. An application to task-oriented grasping validates its performance on hardware.

For the case of parametric learning, simulators can serve as priors or regularizers. This work describes how to use simulation to regularize a VAE's decoder to bind the VAE's latent space to simulator parameter posterior. With that, training on a small number of real trajectories can quickly shift the posterior to reflect reality. The included publication demonstrates that this approach can also help reinforcement learning (RL) quickly overcome the sim-to-real gap on a manipulation task on hardware.

A longer-term vision is to shape latent spaces without needing to mandate a particular simulation scenario. A first step is to learn general relations that hold on sequences of states from a set of related domains. This work introduces a unifying mathematical formulation for learning independent analytic relations. Relations are learned from source domains, then used to help structure the latent space when learning on target domains. This formulation enables a more general, flexible and principled way of shaping the latent space. It formalizes the notion of learning independent relations, without imposing restrictive simplifying assumptions or requiring domain-specific information. This work presents mathematical properties, concrete algorithms and experimental validation of successful learning and transfer of latent relations.

We develop an approach that benefits from large simulated datasets and takes full advantage of the limited online data that is most relevant. We propose a variant of Bayesian optimization that alternates between using informed and uninformed kernels. With this Bernoulli Alternation Kernel we ensure that discrepancies between simulation and reality do not hinder adapting robot control policies online. The proposed approach is applied to a challenging real-world problem of task-oriented grasping with novel objects. Our further contribution is a neural network architecture and training pipeline that use experience from grasping objects in simulation to learn grasp stability scores. We learn task scores from a labeled dataset with a convolutional network, which is used to construct an informed kernel for our variant of Bayesian optimization. Experiments on an ABB Yumi robot with real sensor data demonstrate success of our approach, despite the challenge of fulfilling task requirements and high uncertainty over physical properties of objects.

Gaussian Processes (GPs) have been widely used in robotics as models, and more recently as key structures in active learning algorithms, such as Bayesian optimization. GPs consist of two main components: the mean function and the kernel. Specifying a prior mean function has been a common way to incorporate prior knowledge. When a prior mean function could not be constructed manually, the next default has been to incorporate prior (simulated) observations into a GP as 'fake' data. Then, this GP would be used to further learn from true data on the target (real) domain. We argue that embedding prior knowledge into GP kernels instead provides a more flexible way to capture simulation-based information. We give examples of recent works that demonstrate the wide applicability of such kernel-centric treatment when using GPs as part of Bayesian optimization. We also provide discussion that helps to build intuition for why such 'kernels as priors' view is beneficial.

Data-efficiency is crucial for autonomous robots to adapt to new tasks and environments. In this work, we focus on robotics problems with a budget of only 10-20 trials. This is a very challenging setting even for data-efficient approaches like Bayesian optimization (BO), especially when optimizing higher-dimensional controllers. Previous work extracted expert-designed low-dimensional features from simulation trajectories to construct informed kernels and run ultra sample-efficient BO on hardware. We remove the need for expert-designed features by proposing a model and architecture for a sequential variational autoencoder that embeds the space of simulated trajectories into a lower-dimensional space of latent paths in an unsupervised way. We further compress the search space for BO by reducing exploration in parts of the state space that are undesirable, without requiring explicit constraints on controller parameters. We validate our approach with hardware experiments on a Daisy hexapod robot and an ABB Yumi manipulator. We also present simulation experiments with further comparisons to several baselines on Daisy and two manipulators. Our experiments indicate the proposed trajectory-based kernel with dynamic compression can offer ultra data-efficient optimization.

Manipulation of deformable objects has given rise to an important set of open problems in the field of robotics. Application areas include robotic surgery, household robotics, manufacturing, logistics, and agriculture, to name a few. Related research problems span modeling and estimation of an object's shape, estimation of an object's material properties, such as elasticity and plasticity, object tracking and state estimation during manipulation, and manipulation planning and control. In this survey article, we start by providing a tutorial on foundational aspects of models of shape and shape dynamics. We then use this as the basis for a review of existing work on learning and estimation of these models and on motion planning and control to achieve desired deformations. We also discuss potential future lines of work.

Unmanned Aerial Vehicles (UAV) are a potential solution to fast and cost efficient package delivery services. There are two types of UAVs, namely fixed wing (UAV-FW) and rotor wing (UAV-RW), which have their own advantages and drawbacks. In this paper we aim at providing different solutions to a collaborating multi-agent scenario combining both UAVs types. We show the problem can be reduced to the facility location problem (FLP) and propose two local search algorithms to solve it: Tabu search and simulated annealing.

Fibrin polymerization involves thrombin-mediated exposure of knobs on one monomer that bind to holes available on another, leading to the formation of fibers. In silico evidence has suggested that the classical A:a knob-hole interaction is enhanced by surrounding residues not directly involved in the binding pocket of hole a, via noncovalent interactions with knob A. We assessed the importance of extended knob-hole interactions by performing biochemical, biophysical, and in silico modeling studies on recombinant human fibrinogen variants with mutations at residues responsible for the extended interactions. Three single fibrinogen variants, yD297N, yE323Q, and yK356Q, and a triple variant yDEK (yD297N/yE323Q/yK356Q) were produced in a CHO (Chinese Hamster Ovary) cell expression system. Longitudinal protofibril growth probed by atomic force microscopy was disrupted for yD297N and enhanced for the yK356Q mutation. Initial polymerization rates were reduced for all variants in turbidimetric studies. Laser scanning confocal microscopy showed that yDEK and yE323Q produced denser clots, whereas yD297N and yK356Q were similar to wild type. Scanning electron microscopy and light scattering studies showed that fiber thickness and protofibril packing of the fibers were reduced for all variants. Clot viscoelastic analysis showed that only yDEK was more readily deformable. In silico modeling suggested that most variants displayed only slip-bond dissociation kinetics compared with biphasic catch-slip kinetics characteristics of wild type. These data provide new evidence for the role of extended interactions in supporting the classical knob-hole bonds involving catch-slip behavior in fibrin formation, clot structure, and clot mechanics.

In situ visualization on high-performance computing systems allows us to analyze simulation results that would otherwise be impossible, given the size of the simulation data sets and offline post-processing execution time. We develop an in situ adaptor for Paraview Catalyst and Nek5000, a massively parallel Fortran and C code for computational fluid dynamics. We perform a strong scalability test up to 2048 cores on KTH’s Beskow Cray XC40 supercomputer and assess in situ visualization’s impact on the Nek5000 performance. In our study case, a high-fidelity simulation of turbulent flow, we observe that in situ operations significantly limit the strong scalability of the code, reducing the relative parallel efficiency to only ≈ 21 % on 2048 cores (the relative efficiency of Nek5000 without in situ operations is ≈ 99 %). Through profiling with Arm MAP, we identified a bottleneck in the image composition step (that uses the Radix-kr algorithm) where a majority of the time is spent on MPI communication. We also identified an imbalance of in situ processing time between rank 0 and all other ranks. In our case, better scaling and load-balancing in the parallel image composition would considerably improve the performance of Nek5000 with in situ capabilities. In general, the result of this study highlights the technical challenges posed by the integration of high-performance simulation codes and data-analysis libraries and their practical use in complex cases, even when efficient algorithms already exist for a certain application scenario.

Electric trucks usually need to charge their batteries during long-range delivery missions, and the charging times are often nontrivial. As charging resources are limited, waiting times for some trucks can be prolonged at certain stations. To facilitate the efficient operation of electric trucks, we propose a distributed charging coordination framework. Within the scheme, the charging stations provide waiting estimates to incoming trucks upon request and assign charging ports according to the first-come, first-served rule. Based on the updated information, the individual trucks compute where and how long to charge whenever approaching a charging station in order to complete their delivery missions timely and cost-effectively. We perform empirical studies for trucks traveling over the Swedish road network and compare our scheme with the one where charging plans are computed offline, assuming unlimited charging facilities. It is shown that the proposed scheme outperforms the offline approach at the expense of little communication overhead.

Recent trends envisage robots being deployed in areas deemed dangerous to humans, such as buildings with gas and radiation leaks. In such situations, the model of the underlying hazardous process might be unknown to the agent a priori, giving rise to the problem of planning for safe behaviour in partially known environments. We employ Gaussian process regression to create a probabilistic model of the hazardous process from local noisy samples. The result of this regression is then used by a risk metric, such as the Conditional Value-at-Risk, to reason about the safety at a certain state. The outcome is a risk function that can be employed in optimal motion planning problems. We demonstrate the use of the proposed function in two approaches. First is a sampling-based motion planning algorithm with an event-based trigger for online replanning. Second is an adaptation to the incremental Gaussian Process motion planner (iGPMP2), allowing it to quickly react and adapt to the environment. Both algorithms are evaluated in representative simulation scenarios, where they demonstrate the ability of avoiding high-risk areas.

Lattice QCD calculations of the nucleon electromagnetic form factors are of interest at both the high and low momentum transfer regions. For high momentum transfers especially there are open questions which require more intense study, such as the potential zero crossing in the proton's electric form factor. We will present recent progress from the QCDSF/UKQCD/CSSM collaboration on the calculation of these form factors using the Feynman-Hellmann method in lattice QCD. The Feynman-Hellmann method allows for greater control over excited states which we take advantage of by going to high values of the momentum transfer. In this proceeding we present results of the form factors up to 6 GeV2, using Nf = 2 + 1 flavour fermions for three different pion masses in the range 310-470 MeV. The results are extrapolated to the physical pion mass through the use of a flavour breaking expansion. 

The recent growth in cellular communications has rapidly created a need for more radio channels. In order to make better use of the available frequency bands new access techniques such as TDMA (Time Division Multiple Access), FDMA (Frequency DMA) and CDMA (Code DMA) have been introduced. Still, the need for new channels is strong and new frequency bands have therefore been allocated for future wireless communication systems. These new communication systems, e.g. the European Personal Communication Network (PCN: 1710-1880MHz) and the North American Personal Communication System (PCS: 1850-1990MHz), use frequencies about twice as high as their predecessors (e.g. AMPS: 824-894MHz and GSM: 880-960MHz). However, at a time when our downtown areas already are littered with basestation antennas, operators are not keen to install more. Therefore, dual-band antennas have gained an increased interest. They would allow the operators to replace two antennas for separate frequency bands with one only, thus, reducing the windload of the towers and perhaps most importantly reducing the cost of installation.

KTH, Skolan för elektroteknik och datavetenskap (EECS) har beretts möjlighet att ge sin syn på PTS ”Konsultation inför planerad tilldelning av frekvensutrymme i 2,3- och 3,5 GHz-banden samt tilldelning av frekvensutrymme för lokala tillstånd”, PTS DNR 18-8496.

KTH anser att det är olyckligt att PTS frångår principen om teknikneutrala band. Vi förstår att i detta fall finns inga alternativ men råder PTS att arbeta för att framtida spektrumtilldelning sker på teknikneutral basis då inlåsning av frekvensband till specifika tekniker inte bidrar till innovation.

TDD medger ett mer effektivt utnyttjande av spektrum i fallet då nedlänkstrafik dominerar. Denna vinst måste dock vägas mot det faktum att co-ordinering och synkronisering av och mellan operatörernas nät hämmar konkurrensen. En möjlig framtida lösning är att 3.5 bandet enbart används för nedlänk och att upplänk sker i annat band (tex 1800 bandet). 

Slutligen så bör spektrumvillkoren vara oberoende av de nu existerande aktörerna. För att även öppna upp för flera och ev nya operatörer så föreslår vi att hela bandet 3,4 - 3,8GHz bjuds ut vid denna auktion.

Using an SU(3)-flavour symmetry breaking expansion between the strange and light quark masses, we determine how this constrains the extrapolation of baryon octet matrix elements and form factors. In particular we can construct certain combinations, which fan out from the symmetric point (when all the quark masses are degenerate) to the point where the light and strange quarks take their physical values. As a further example we consider the vector amplitude at zero momentum transfer for flavour changing currents.

Since the 1950s, robotics research has sought to build a general-purpose agent capable of autonomous, open-ended interaction with realistic, unconstrained environments. Cognition is perceived to be at the core of this process, yet understan#ding has been challenged because cognition is referred to differently within and across research areas, and is not clearly defined. The classic robotics approach is decomposition into functional modules which perform planning, reasoning, and problem solving or provide input to these mechanisms. Although advancements have been made and numerous success stories reported in specific niches, this systems-engineering approach has not succeeded in building such a cognitive agent. The emergence of an action-oriented paradigm oilers a new approach: action and perception are no longer separable into functional modules but must be considered in a complete loop. This chapter reviews work on different mechanisms for action-perception learning and discusses the role of embodiment in the design of the underlying representations and learning. It discusses the evaluation of agents and suggests the development of a new embodied Turing lest. Appropriate scenarios need to be devised in addition to current competitions, so that abilities can be tested over long time periods.

The task of exploration does not end when the robot has covered the entire environment. The world is dynamic and to model this property and to keep the map up to date the robot needs to re-explore. In this work, we present an approach to long-term exploration that builds on prior work on dynamic mapping, volumetric representations of space, and exploration planning. The main contribution of our work is a novel formulation of the information gain function that controls the exploration so that it trades off revisiting highly dynamic areas where changes are very likely with covering the rest of the environment to ensure both coverage and up-to-date estimates of the dynamics. We provide experimental validation of our approach in three different simulated environments.

Skyrmion stabilization in novel magnetic systems with the B20 crystal structure is reported here, primarily based on theoretical results. The focus is on the effect of alloying on the 3d sublattice of the B20 structure by substitution of heavier 4d and 5d elements, with the ambition to tune the spin-orbit coupling and its influence on magnetic interactions. State-of-the-art methods based on density functional theory are used to calculate both isotropic and anisotropic exchange interactions. Significant enhancement of the Dzyaloshinskii-Moriya interaction is reported for 5d-doped FeSi and CoSi, accompanied by a large modification of the spin stiffness and spiralization. Micromagnetic simulations coupled to atomistic spin-dynamics and ab initio magnetic interactions reveal the spin-spiral nature of the magnetic ground state and field-induced skyrmions for all these systems. Especially small skyrmions similar to 50 nm are predicted for Co0.75Os0.25Si, compared to similar to 148 nm for Fe0.75Co0.25Si. Convex-hull analysis suggests that all B20 compounds considered here are structurally stable at elevated temperatures and should be possible to synthesize. This prediction is confirmed experimentally by synthesis and structural analysis of the Ru-doped CoSi systems discussed here, both in powder and in single-crystal forms.

Generative software development has paved the way for the creation of multiple code generators that serve as a basis for automatically generating code to different software and hardware platforms. In this context, the software quality becomes highly correlated to the quality of code generators used during software development. Eventual failures may result in a loss of confidence for the developers, who will unlikely continue to use these generators. It is then crucial to verify the correct behaviour of code generators in order to preserve software quality and reliability. In this paper, we leverage the metamorphic testing approach to automatically detect inconsistencies in code generators via so-called “metamorphic relations”. We define the metamorphic relation (i.e., test oracle) as a comparison between the variations of performance and resource usage of test suites running on different versions of generated code. We rely on statistical methods to find the threshold value from which an unexpected variation is detected. We evaluate our approach by testing a family of code generators with respect to resource usage and performance metrics for five different target software platforms. The experimental results show that our approach is able to detect, among 95 executed test suites, 11 performance and 15 memory usage inconsistencies.

While the anomalous non-additive size-dependencies of static dipole polarizabilities and van der Waals C-6 dispersion coefficients of carbon fullerenes are well established, the widespread reported scalings for the latter (ranging from N-2.2 to N-2.8) call for a comprehensive first-principles investigation. With a highly efficient implementation of the linear complex polarization propagator, we have performed Hartree-Fock and Kohn-Sham density functional theory calculations of the frequency-dependent polarizabilities for fullerenes consisting of up to 540 carbon atoms. Our results for the static polarizabilities and C-6 coefficients show scalings of N-1.2 and N-2.2, respectively, thereby deviating significantly from the previously reported values obtained with the use of semi-classical/empirical methods. Arguably, our reported values are the most accurate to date as they represent the first ab initio or first-principles treatment of fullerenes up to a convincing system size.

We present the implementation of an efficient matrix-folded formalism for the evaluation of complex response functions and the calculation of transition properties at the level of the second-order algebraic-diagrammatic construction (ADC(2)) scheme. The underlying algorithms, in combination with the adopted hybrid MPI/OpenMP parallelization strategy, enabled calculations of the UV/vis spectra of a guanine oligomer series ranging up to 1032 contracted basis functions, thereby utilizing vast computational resources from up to 32,768 CPU cores. Further analysis of the convergence behavior of the involved iterative subspace algorithms revealed the superiority of a frequency-separated treatment of response equations even for a large spectral window, including 101 frequencies. We demonstrate the applicability to general quantum mechanical operators by the first reported electronic circular dichroism spectrum calculated with a complex polarization propagator approach at the ADC(2) level of theory.

So-called SIMD instructions, which trigger operations that process in each clock cycle a data tuple, have become widespread in modern processor architectures. In particular, processors for high-performance computing (HPC) systems rely on this additional level of parallelism to reach a high throughput of arithmetic operations. Leveraging these SIMD instructions can still be challenging for application software developers. This challenge has become simpler due to a compiler technique called auto-vectorization. In this paper, we explore the current state of auto-vectorization capabilities using state-of-the-art compilers using a recent extension of the Arm instruction set architecture, called SVE. We measure the performance gains on a recent processor architecture supporting SVE, namely the Fujitsu A64FX processor.

We present an approach to automatically assign semantic labels to rooms reconstructed from 3D RGB maps of apartments. Evidence for the room types is generated using state-of-the-art deep-learning techniques for scene classification and object detection based on automatically generated virtual RGB views, as well as from a geometric analysis of the map's 3D structure. The evidence is merged in a conditional random field, using statistics mined from different datasets of indoor environments. We evaluate our approach qualitatively and quantitatively and compare it to related methods.

In this paper, we propose a new way of doing formation obstacle avoidance using a combination of Constraint Based Programming (CBP) and Rapidly Exploring Random Trees (RRTs). RRT is used to select waypoint nodes, and CBP is used to move the formation between those nodes, reactively rotating and translating the formation to pass the obstacles on the way. Thus, the CBP includes constraints for both formation keeping and obstacle avoidance, while striving to move the formation towards the next waypoint. The proposed approach is compared to a pure RRT approach where the motion between the RRT waypoints is done following linear interpolation trajectories, which are less computationally expensive than the CBP ones. The results of a number of challenging simulations show that the proposed approach is more efficient for scenarios with high obstacle densities.

We study the temperature behavior of the first four peaks of the oxygen-oxygen radial distribution function of water, simulated by the TIP4P/2005, MB-pol, TIP5P, and SPC/E models and compare to experimental X-ray diffraction data, including a new measurement which extends down to 235 K [H. Pathak et al., J. Chem. Phys. 150, 224506 (2019)]. We find the overall best agreement using the MB-pol and TIP4P/2005 models. We observe, upon cooling, a minimum in the position of the second shell simulated with TIP4P/2005 and SPC/E potentials, located close to the temperature of maximum density. We also calculated the two-body entropy and the contributions coming from the first, second, and outer shells to this quantity. We show that, even if the main contribution comes from the first shell, the contribution of the second shell can become important at low temperature. While real water appears to be less ordered at short distance than obtained by any of the potentials, the different water potentials show more or less order compared to the experiments depending on the considered length-scale.

We use molecular dynamics simulations using TIP4P/2005 to investigate the self- and distinct-van Hove functions for different local environments of water, classified using the local structure index as an order parameter. The orientational dynamics were studied through the calculation of the time-correlation functions of different-order Legendre polynomials in the OH-bond unit vector. We found that the translational and orientational dynamics are slower for molecules in a low-density local environment and correspondingly the mobility is enhanced upon increasing the local density, consistent with some previous works, but opposite to a recent study on the van Hove function. From the analysis of the distinct dynamics, we find that the second and fourth peaks of the radial distribution function, previously identified as low density-like arrangements, show long persistence in time. The analysis of the time-dependent interparticle distance between the central molecule and the first coordination shell shows that particle identity persists longer than distinct van Hove correlations. The motion of two first-nearest-neighbor molecules thus remains coupled even when this correlation function has been completely decayed. With respect to the orientational dynamics, we show that correlation functions of molecules in a low-density environment decay exponentially, while molecules in a local high-density environment exhibit bi-exponential decay, indicating that dynamic heterogeneity of water is associated with the heterogeneity among high-density and between high-density and low-density species. This bi-exponential behavior is associated with the existence of interstitial waters and the collapse of the second coordination sphere in high-density arrangements, but not with H-bond strength.

Learning state representations enables robotic planning directly from raw observations such as images. Several methods learn state representations by utilizing losses based on the reconstruction of the raw observations from a lower-dimensional latent space. The similarity between observations in the space of images is often assumed and used as a proxy for estimating similarity between the underlying states of the system. However, observations commonly contain task-irrelevant factors of variation which are nonetheless important for reconstruction, such as varying lighting and different camera viewpoints. In this work, we define relevant evaluation metrics and perform a thorough study of different loss functions for state representation learning. We show that models exploiting task priors, such as Siamese networks with a simple contrastive loss, outperform reconstruction-based representations in visual task planning in case of task-irrelevant factors of variations.

Artificial Intelligence (AI) and Machine Learning (ML) approaches have emerged in the networking domain with great expectation. They can be broadly divided into AI/ML techniques for network engineering and management, network designs for AI/ML applications, and system concepts. AI/ML techniques for networking and management improve the way we address networking. They support efficient, rapid, and trustworthy engineering, operations, and management. As such, they meet the current interest in softwarization and network programmability that fuels the need for improved network automation in agile infrastructures, including edge and fog environments. Network design and optimization for AI/ML applications addresses the complementary topic of supporting AI/ML-based systems through novel networking techniques, including new architectures and algorithms. The third topic area is system implementation and open-source software development.

We study the computation of the global generalized Nash equilibrium (GNE) for a class of non-convex multi-player games, where players' actions are subject to both local and coupling constraints. Due to the non-convex payoff functions, we employ canonical duality to reformulate the setting as a complementary problem. Under given conditions, we reveal the relation between the stationary point and the global GNE. According to the convex-concave properties within the complementary function, we propose a continuous-time mirror descent to compute GNE by generating functions in the Bregman divergence and the damping-based design. Then, we devise several Lyapunov functions to prove that the trajectory along the dynamics is bounded and convergent.

Undetectable attacks are an important class of malicious attacks threatening the security of cyber-physical systems, which can modify a system's state but leave the system output measurements unaffected and hence cannot be detected from the output. This paper studies undetectable attacks on cone-invariant systems and multi-agent systems. We first provide a general characterization of zero-dynamics attacks, which characterizes fully undetectable attacks targeting the nonminimum phase zeros of a system. This geometrical characterization makes it possible to develop a defense strategy seeking to place a minimal number of sensors to detect and counter the zero-dynamics attacks on the system's actuators. The detect and defense scheme amounts to computing a set containing potentially vulnerable actuator locations and nodes and a defense union for feasible placement of sensors based on the geometrical properties of the cones under consideration.

The performance of Deep-Learning (DL) computing frameworks rely on the rformance of data ingestion and checkpointing. In fact, during the aining, a considerable high number of relatively small files are first aded and pre-processed on CPUs and then moved to accelerator for mputation. In addition, checkpointing and restart operations are rried out to allow DL computing frameworks to restart quickly from a eckpoint. Because of this, I/O affects the performance of DL plications. this work, we characterize the I/O performance and scaling of nsorFlow, an open-source programming framework developed by Google and ecifically designed for solving DL problems. To measure TensorFlow I/O rformance, we first design a micro-benchmark to measure TensorFlow ads, and then use a TensorFlow mini-application based on AlexNet to asure the performance cost of I/O and checkpointing in TensorFlow. To prove the checkpointing performance, we design and implement a burst ffer. find that increasing the number of threads increases TensorFlow ndwidth by a maximum of 2.3 x and 7.8 x on our benchmark environments. e use of the tensorFlow prefetcher results in a complete overlap of mputation on accelerator and input pipeline on CPU eliminating the fective cost of I/O on the overall performance. The use of a burst ffer to checkpoint to a fast small capacity storage and copy ynchronously the checkpoints to a slower large capacity storage sulted in a performance improvement of 2.6x with respect to eckpointing directly to slower storage on our benchmark environment.