# Sandra HircheTechnische Universität München | TUM · EI

Sandra Hirche

## About

392

Publications

35,670

Reads

**How we measure 'reads'**

A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more

5,854

Citations

Citations since 2016

## Publications

Publications (392)

Ensuring safety is of paramount importance in physical human-robot interaction applications. This requires both an adherence to safety constraints defined on the system state, as well as guaranteeing compliant behaviour of the robot. If the underlying dynamical system is known exactly, the former can be addressed with the help of control barrier fu...

When moving a piano or dancing tango with a partner, how should I control my arm muscles to best feel their movements and follow or guide them smoothly? Here we observe how physically connected pairs tracking a moving target with the arm modify muscle coactivation with their visual acuity and the partners performance. They coactivate muscles to sti...

For safe operation, a robot must be able to avoid collisions in uncertain environments. Existing approaches for motion planning with uncertainties often make conservative assumptions about Gaussianity and the obstacle geometry. While visual perception can deliver a more accurate representation of the environment, its use for safe motion planning is...

Ensuring safety is a crucial challenge when deploying reinforcement learning (RL) to real-world systems. We develop confidence-based safety filters, a control-theoretic approach for certifying state safety constraints for nominal policies learned via standard RL techniques, based on probabilistic dynamics models. Our approach is based on a reformul...

Reinforcement learning is a promising method for robotic grasping as it can learn effective reaching and grasping policies in difficult scenarios. However, achieving human-like manipulation capabilities with sophisticated robotic hands is challenging because of the problem's high dimensionality. Although remedies such as reward shaping or expert de...

This paper proposes a physically consistent Gaussian Process (GP) enabling the identification of uncertain Lagrangian systems. The function space is tailored according to the energy components of the Lagrangian and the differential equation structure, analytically guaranteeing physical and mathematical properties such as energy conservation and qua...

System representations inspired by the infinite-dimensional Koopman operator (generator) are increasingly considered for predictive modeling. Due to the operator’s linearity, a range of nonlinear systems admit linear predictor representations – allowing for simplified prediction, analysis and control. However, finding meaningful finite-dimensional...

In this letter, we investigate the problem of actuator scheduling for networked control systems. Given a stochastic linear system with a number of actuators, we consider the case that one actuator is activated at each time. This problem is combinatorial in nature and NP hard to solve. We propose a convex relaxation to the actuator scheduling proble...

Safety-critical technical systems operating in unknown environments require the ability to quickly adapt their behavior, which can be achieved in control by inferring a model online from the data stream generated during operation. Gaussian process-based learning is particularly well suited for safety-critical applications as it ensures bounded pred...

We propose a novel framework for learning linear time-invariant (LTI) models for a class of continuous-time non-autonomous nonlinear dynamics based on a representation of Koopman operators. In general, the operator is infinite-dimensional but, crucially, linear. To utilize it for efficient LTI control, we learn a finite representation of the Koopma...

During the last few decades, great research endeavors have been applied to healthcare robots, aiming to develop companions that extend the independent living of elderly people. To deploy such robots into the market, it is expected that certain applications should be addressed with repeatability and robustness. Such application is the assistance wit...

In this letter, we investigate the problem of actuator scheduling for networked control systems. Given a stochastic linear system with a number of actuators, we consider the case that one actuator is activated at each time. This problem is combinatorial in nature and NP-Hard to solve. We propose a convex relaxation to the actuator scheduling proble...

The rate-regulation tradeoff, defined between two objective functions, one penalizing the packet rate and one the regulation cost, can express the fundamental performance bound of networked control systems. However, the characterization of the set of globally optimal solutions in this tradeoff for multi-dimensional Gauss–Markov processes has been a...

Underactuated vehicles have gained much attention in the recent years due to the increasing amount of aerial and underwater vehicles as well as nanosatellites. The safe tracking control of these vehicles is a substantial aspect for an increasing range of application domains. However, external disturbances and parts of the internal dynamics are ofte...

For single agent systems, probabilistic machine learning techniques such as Gaussian process regression have been shown to be suitable methods for inferring models of unknown nonlinearities, which can be employed to improve the performance of control laws. While this approach can be extended to the cooperative control of multi-agent systems, it lea...

Underactuated vehicles have gained much attention in the recent years due to the increasing amount of aerial and underwater vehicles as well as nanosatellites. Trajectory tracking control of these vehicles is a substantial aspect for an increasing range of application domains. However, external disturbances and parts of the internal dynamics are of...

Gaussian process regression is often applied for learning unknown systems and specifying the uncertainty of the learned model. When using Gaussian process regression to learn unknown systems, a commonly considered approach consists of learning the residual dynamics after applying some standard discretization, which might however not be appropriate...

The modeling and simulation of dynamical systems is a necessary step for many control approaches. Using classical, parameter-based techniques for modeling of modern systems, e.g., soft robotics or human-robot interaction, is often challenging or even infeasible due to the complexity of the system dynamics. In contrast, data-driven approaches need o...

This article proposes a distributed control methodology for the coordination of multiple connected and automated vehicles (CAVs). Each vehicle computes its local trajectory based on a model predictive control (MPC) law and communicates the result with other relevant vehicles. In an iterative negotiation process, planned trajectories are optimized w...

When signals are measured through physical sensors, they are perturbed by noise. To reduce noise, low-pass filters are commonly employed in order to attenuate high frequency components in the incoming signal, regardless if they come from noise or the actual signal. Therefore, low-pass filters must be carefully tuned in order to avoid significant de...

The Koopman operator allows for handling nonlinear systems through a globally linear representation. In general, the operator is infinite-dimensional – necessitating finite approximations – for which there is no overarching framework. Although there are principled ways of learning such finite approximations, they are in many instances overlooked in...

We present a novel data-driven approach for learning linear representations of a class of stable nonlinear systems using Koopman eigenfunctions. By learning the conjugacy map between a nonlinear system and its Jacobian linearization through a Normalizing Flow one can guarantee the learned function is a diffeomorphism. Using this diffeomorphism, we...

The use of rehabilitation robotics in clinical applications gains increasing importance, due to therapeutic benefits and the ability to alleviate labor-intensive works. However, their practical utility is dependent on the deployment of appropriate control algorithms, which adapt the level of task-assistance according to each individual patient's ne...

Although transmission of a data packet containing sensory information in a networked control system improves the quality of regulation, it has indeed a price from the communication perspective. It is, therefore, rational that such a data packet be transmitted only if it is valuable in the sense of a cost-benefit analysis. Yet, the fact is that litt...

Simulation of contact and friction dynamics is an important basis for control- and learning-based algorithms. However, the numerical difficulties of contact interactions pose a challenge for robust and efficient simulators. A maximal-coordinate representation of the dynamics enables efficient solving algorithms, but current methods in maximal coord...

Gaussian processes have become a promising tool for various safety-critical settings, since the posterior variance can be used to directly estimate the model error and quantify risk. However, state-of-the-art techniques for safety-critical settings hinge on the assumption that the kernel hyperparameters are known, which does not apply in general. T...

In this article, we study an energy-regulation trade-off that delineates the fundamental performance bound of a feedback control system over a noisy channel in an unreliable communication regime. The channel and process are modeled by an additive white Gaussian noise channel with fading and a partially observable Gauss-Markov process, respectively....

This paper is concerned with a special case of stochastic distributed optimal control, where the objective is to design a structurally constrained controller for a system subject to state and input power constraints. The structural constraints are induced by the directed communication between local controllers over a strongly connected graph. Based...

Voronoi coverage control is a particular problem of importance in the area of multi-robot systems, which considers a network of multiple autonomous robots, tasked with optimally covering a large area. This is a common task for fleets of fixed-wing Unmanned Aerial Vehicles (UAVs), which are described in this work by a unicycle model with constant fo...

Recent works in the domain of networked control systems have demonstrated that the joint design of medium access control strategies and control strategies for the closed-loop system is beneficial. However, several metrics introduced so far fail in either appropriately representing the network requirements or in capturing how valuable the data is. I...

In this paper, we propose an observer-based visual pursuit control integrating three-dimensional target motion learning by Gaussian Process Regression (GPR). We consider a situation where a visual sensor equipped rigid body pursuits a target rigid body whose velocity is unknown but dependent on the target's pose. We estimate the pose from visual in...

Transmission delay and packet dropout are inevitable network-induced phenomena that severely compromise the control performance of network control systems. The real-time network traffic is a major dynamic parameter that directly influences delay and reliability of transmission channels, and thus, acts as an unavoidable source of induced coupling am...

Federated learning is the distributed machine learning framework that enables collaborative training across multiple parties while ensuring data privacy. Practical adaptation of XGBoost, the state-of-the-art tree boosting framework, to federated learning remains limited due to high cost incurred by conventional privacy-preserving methods. To addres...

A fundamental aspect of racing is overtaking other race cars. Whereas previous research on autonomous racing has majorly focused on lap-time optimization, here, we propose a method to plan overtaking maneuvers in autonomous racing. A Gaussian process is used to learn the behavior of the leading vehicle. Based on the outputs of the Gaussian process,...

In the operation of networked control systems, where multiple processes share a resource-limited and time-varying cost-sensitive network, communication delay is inevitable and primarily induced by, first, intermittent sensor sampling to restrict non-urgent transmissions, and second, resource management to avoid contentions, excessive traffic, and d...

For tasks where the dynamics of multiple agents are physically coupled, e.g., in cooperative manipulation, the coordination between the individual agents becomes crucial, which requires exact knowledge of the interaction dynamics. This problem is typically addressed using centralized estimators, which can negatively impact the flexibility and robus...

Inferring the intent of an intelligent agent from demonstrations and subsequently predicting its behavior, is a critical task in many collaborative settings. A common approach to solve this problem is the framework of inverse reinforcement learning (IRL), where the observed agent, e.g., a human demonstrator, is assumed to behave according to an int...

In this paper, we study the trade-off between the transmission cost and the control performance of the multi-loop networked control system subject to network-induced delay. Within the linear-quadratic-Gaussian (LQG) framework, the joint design of control policy and networking strategy is decomposed into separation optimization problems. Based on th...

Formation control algorithms for multi-agent systems have gained much attention in the recent years due to the increasing amount of mobile and aerial robotic swarms. The design of safe controllers for these vehicles is a substantial aspect for an increasing range of application domains. However, parts of the vehicle's dynamics and external disturba...

In this paper, a distributed learning leader-follower consensus protocol based on Gaussian process regression for a class of nonlinear multi-agent systems with unknown dynamics is designed. We propose a distributed learning approach to predict the residual dynamics for each agent. The stability of the consensus protocol using the data-driven model...

The rate-regulation trade-off defined between two objective functions, one penalizing the packet rate and one the state deviation and control effort, can express the performance bound of a networked control system. However, the characterization of the set of globally optimal solutions in this trade-off for multi-dimensional controlled Gauss-Markov...

This paper deals with data-driven output synchronization for heterogeneous leader-follower linear multi-agent systems. Given a multi-agent system that consists of one autonomous leader and a number of heterogeneous followers with external disturbances, we provide necessary and sufficient data-based conditions for output synchronization. We also pro...

Consider a network of multiple independent stochastic linear systems where, for each system, a scheduler collocated with the sensors arbitrates data transmissions to a corresponding remote controller through a shared contention-based communication network. While the systems are physically independent, their optimal controller design problems may, i...

The Koopman operator allows for handling nonlinear systems through a (globally) linear representation. In general, the operator is infinite-dimensional - necessitating finite approximations - for which there is no overarching framework. Although there are principled ways of learning such finite approximations, they are in many instances overlooked...

In application areas where data generation is expensive, Gaussian processes are a preferred supervised learning model due to their high data-efficiency. Particularly in model-based control, Gaussian processes allow the derivation of performance guarantees using probabilistic model error bounds. To make these approaches applicable in practice, two o...

Despite the existence of formal guarantees for learning-based control approaches, the relationship between data and control performance is still poorly understood. In this paper, we present a measure to quantify the value of data within the context of a predefined control task. Our approach is applicable to a wide variety of unknown nonlinear syste...

In this brief note we compute the Fisher information of a family of generalized normal distributions. Fisher information is usually defined for regular distributions, i.e. continuously differentiable (log) density functions whose support does not depend on the family parameter $\theta$. Although the uniform distribution in $[-\theta, + \theta]$ doe...

Safety-critical decisions based on machine learning models require a clear understanding of the involved uncertainties to avoid hazardous or risky situations. While aleatoric uncertainty can be explicitly modeled given a parametric description, epistemic uncertainty rather describes the presence or absence of training data. This paper proposes a no...

Underactuated vehicles have gained much attention in the recent years due to the increasing amount of aerial and underwater vehicles as well as nanosatellites. The safe tracking control of these vehicles is a substantial aspect for an increasing range of application domains. However, external disturbances and parts of the internal dynamics are ofte...

Networked control system (NCS) refer to a set of control loops that are closed over a communication network. In this article, the joint operation of control and networking for NCS is investigated wherein the network serves the sensor-to-controller communication links for multiple stochastic linear time-invariant (LTI) sub-systems. The sensors sampl...

Control schemes that learn using measurement data collected online are increasingly promising for the control of complex and uncertain systems. However, in most approaches of this kind, learning is viewed as a side effect that passively improves control performance, e.g., by updating a model of the system dynamics. Determining how improvements in c...

In the operation of networked control systems, where multiple processes share a resource-limited and time-varying cost-sensitive network, communication delay is inevitable and primarily influenced by, first, the control systems deploying intermittent sensor sampling to reduce the communication cost by restricting non-urgent transmissions, and secon...

When first principle models cannot be derived due to the complexity of the real system, data-driven methods allow us to build models from system observations. As these models are employed in learning-based control, the quality of the data plays a crucial role for the performance of the resulting control law. Nevertheless, there hardly exist measure...

Learning-based techniques are increasingly effective at controlling complex systems. However, most work done so far has focused on learning control laws for individual tasks. Simultaneously learning multiple tasks on the same system is still a largely unaddressed research question. In particular, no efficient state space exploration schemes have be...

In the operation of networked control systems , where multiple processes share a resource-limited and time-varying cost-sensitive network, communication delay is inevitable and primarily influenced by, first, the control systems deploying intermittent sensor sampling to reduce the communication cost by restricting non-urgent transmissions, and seco...

The modeling and simulation of dynamical systems is a necessary step for many control approaches. Using classical, parameter-based techniques for modeling of modern systems, e.g., soft robotics or human-robot interaction, is often challenging or even infeasible due to the complexity of the system dynamics. In contrast, data-driven approaches need o...

The increased demand for online prediction and the growing availability of large data sets drives the need for computationally efficient models. While exact Gaussian process regression shows various favorable theoretical properties (uncertainty estimate, unlimited expressive power), the poor scaling with respect to the training set size prohibits i...