About
539
Publications
123,214
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
12,400
Citations
Introduction
Systems are becoming increasingly complex and interconnected, requiring autonomous responses to changes and disturbances. Our research group develops theoretically sound control methods integrated with machine learning to create safe autonomous systems for dynamic environments. We advance control theory while delivering solutions across robotics, autonomous driving, battery systems, energy management, synthetic biology and biomedicine. Our methods ensure safety and performance.
Current institution
Additional affiliations
May 2012 - October 2012
January 2012 - May 2012
July 1997 - October 1999
Publications
Publications (539)
Early battery life prediction models are most useful for R&D if they help us understand the early changes in battery electrochemical response that correspond with long-term degradation and failure. Linear regression models such as Fused lasso and Partial Least Squares can fit coefficients directly to high-dimensional electrochemical data like capac...
This tutorial paper focuses on safe physics-informed machine learning in the context of dynamics and control, providing a comprehensive overview of how to integrate physical models and safety guarantees. As machine learning techniques enhance the modeling and control of complex dynamical systems, ensuring safety and stability remains a critical cha...
This paper proposes a path planning algorithm for multi-agent unmanned aircraft systems (UASs) to autonomously cover a search area, while considering obstacle avoidance, as well as the capabilities and energy consumption of the employed unmanned aerial vehicles. The path planning is optimized in terms of energy efficiency to prefer low energy-consu...
Unmanned aerial vehicles (UAVs), especially multicopters, have recently gained popularity for use in surveillance, monitoring, inspection, and search and rescue missions. Their maneuverability and ability to operate in confined spaces make them particularly useful in cluttered environments. For advanced control and mission planning applications, ac...
Trajectory planning for automated vehicles commonly employs optimization over a moving horizon - Model Predictive Control - where the cost function critically influences the resulting driving style. However, finding a suitable cost function that results in a driving style preferred by passengers remains an ongoing challenge. We employ preferential...
This paper presents a framework for bounding the approximation error in imitation model predictive controllers utilizing neural networks. Leveraging the Lipschitz properties of these neural networks, we derive a bound that guides dataset design to ensure the approximation error remains at chosen limits. We discuss how this method can be used to des...
Cloud-based energy management systems (EMS) in smart grids face privacy challenges, as existing methods based on traditional homomorphic encryption support limited operations and are vulnerable to quantum attacks. We propose a privacy preservation method for smart grid energy management system, leveraging edge-assisted computing and ring learning w...
Closed-loop performance of sequential decision making algorithms, such as model predictive control, depends strongly on the parameters of cost functions, models, and constraints. Bayesian optimization is a common approach to learning these parameters based on closed-loop experiments. However, traditional Bayesian optimization approaches treat the l...
Navigating autonomous vehicles within a partially known environment to achieve a specific goal is an important yet challenging problem. It necessitates ensuring the safety of the vehicle along its trajectory, accounting for potentially unknown obstacles while maintaining the vehicle's ability to navigate the path at all times. Conventionally , a sa...
Linear regression is often deemed inherently interpretable; however, challenges arise for high-dimensional data. We focus on further understanding how linear regression approximates nonlinear responses from high-dimensional functional data, motivated by predicting cycle life for lithium-ion batteries. We develop a linearization method to derive fea...
Employing model predictive control to systems with unbounded, stochastic disturbances poses the challenge of guaranteeing safety, i.e., repeated feasibility and stability of the closed-loop system. Especially, there are no strict repeated feasibility guarantees for standard stochastic MPC formulations. Thus, traditional stability proofs are not str...
We present a method, which allows efficient and safe approximation of model predictive controllers using kernel interpolation. Since the computational complexity of the approximating function scales linearly with the number of data points, we propose to use a scoring function which chooses the most promising data. To further reduce the complexity o...
Model predictive control (MPC) is a powerful tool for controlling complex nonlinear systems under constraints, but often struggles with model uncertainties and the design of suitable cost functions. To address these challenges, we discuss an approach that integrates MPC with safe Bayesian optimization to optimize long-term closed-loop performance d...
Safe learning of control policies remains challenging, both in optimal control and reinforcement learning. In this article, we consider safe learning of parametrized predictive controllers that operate with incomplete information about the underlying process. To this end, we employ Bayesian optimization for learning the best parameters from closed-...
Safety and maintaining high performance are key considerations during the operation of lithium-ion batteries. Battery degradation, in particular lithium plating and loss of active material, is often accelerated by fast charging. This study explores a strategy for the design of fast charging protocols that takes into account the influence of the var...
The expansion in automation of increasingly fast applications and low-power edge devices poses a particular challenge for optimization based control algorithms, like model predictive control. Our proposed machine-learning supported approach addresses this by utilizing a feed-forward neural network to reduce the computation load of the online-optimi...
Health monitoring, fault analysis, and detection are critical for the safe and sustainable operation of battery systems. We apply Gaussian process resistance models on lithium iron phosphate battery field data to effectively separate the time-dependent and operating point-dependent resistance. The data set contains 29 battery systems returned to th...
This article concerns nonlinear model predictive control (MPC) with guaranteed feasibility of inequality path constraints (PCs). For MPC with PCs, the existing methods, such as direct multiple shooting, cannot guarantee feasibility of PCs because the PCs are enforced at finitely many time points only. Therefore, this article presents a novel MPC fr...
Secure information exchange of the devices among different domains for cyber-physical power systems (CPPSs) is important yet challenging. Conventional blockchain-based authentication schemes generally adopt single blockchain and signature algorithm, only achieving intradomain or interdomain authentication with lower efficiency, and always failing t...
Optogenetic modulation of adenosine triphosphatase (ATPase) expression represents a novel approach to maximize bioprocess efficiency by leveraging enforced adenosine triphosphate (ATP) turnover. In this study, we experimentally implement a model-based open-loop optimization scheme for optogenetic modulation of the expression of ATPase. Increasing t...
The discrete and charge-separated nature of matter — electrons and nuclei — results in local electrostatic fields that are ubiquitous in nanoscale structures and relevant in catalysis, nanoelectronics and quantum nanoscience. Surface-averaging techniques provide only limited experimental access to these potentials, which are determined by the shape...
Model‐based optimization approaches for monitoring and control, such as model predictive control and optimal state and parameter estimation, have been used successfully for decades in many engineering applications. Models describing the dynamics, constraints, and desired performance criteria are fundamental to model‐based approaches. Thanks to rece...
Optogenetic modulation of adenosine triphosphatase (ATPase) expression represents a novel approach to maximize bioprocess efficiency by leveraging enforced adenosine triphosphate (ATP) turnover. In this study, we experimentally implement a model-based open-loop optimization scheme for optogenetic modulation of the expression of ATPase. Increasing t...
The control of manufacturing processes must satisfy high-quality and efficiency requirements while meeting safety requirements. A broad spectrum of monitoring and control strategies, such as model and optimization-based controllers, is utilized to address these issues. Driven by rising demand for flexible yet energy and resource-efficient operation...
Biotechnology offers many opportunities for the sustainable manufacturing of valuable products. The toolbox to optimize bioprocesses includes extracellular process elements such as the bioreactor design and mode of operation, medium formulation, culture conditions, feeding rates, and so on. However, these elements are frequently insufficient for ac...
Many robotic tasks, such as human-robot interactions or the handling of fragile objects, require tight control and limitation of appearing forces and moments alongside sensible motion control to achieve safe yet high-performance operation. We propose a learning-supported model predictive force and motion control scheme that provides stochastic safe...
Microbial consortia are promising biotechnological production systems with the potential to divide complex metabolic pathways into smaller submodules, as well as make products and consume substrates that monocultures cannot. Maintaining optimal cell population levels and preventing mono culture formation challenge bioproduction by microbial consort...
High-dimensional linear regression is important in many scientific fields. This article considers discrete measured data of underlying smooth latent processes, as is often obtained from chemical or biological systems. Interpretation in high dimensions is challenging because the nullspace and its interplay with regularization shapes regression coeff...
Dynamic optimization problems are pervasive in various fields, ranging from chemical process control to aerospace, autonomous driving, physics, robotics, and beyond. These problems involve optimizing a dynamic system considering inputs, parameters, constraints, and a cost function. For dynamic optimization, two broad classes of strategies emerge: d...
High-dimensional linear regression is important in many scientific fields. This article considers discrete measured data of underlying smooth latent processes, as is often obtained from chemical or biological systems. Interpretation in high dimensions is challenging because the nullspace and its interplay with regularization shapes regression coeff...
A bold vision in nanofabrication is the assembly of functional molecular structures using a scanning probe microscope (SPM). This approach requires continuous monitoring of the molecular configuration during manipulation. Until now, this has been impossible because the SPM tip cannot simultaneously act as an actuator and an imaging probe. Here, we...
This paper tackles the challenge of operating autonomous systems safely and efficiently in cluttered environments with uncertainties. Traditional planning algorithms often rely on simplified, static models, which can result in collisions and under utilization of the system's dynamic capabilities. To address this, we propose a moving horizon plannin...
Planning and control for autonomous vehicles usually are hierarchically separated. However, increasing performance demands and operating in highly dynamic environments requires a frequent re‐evaluation of the planning and tight integration of control and planning to guarantee safety, performance, and reliability. We propose an integrated hierarchic...
Analysis of electrochemical impedance spectroscopy (EIS) data for electrochemical systems often consists of defining an equivalent circuit model (ECM) using expert knowledge and then optimizing the model parameters to deconvolute various resistance, capacitive, inductive, or diffusion responses. For small data sets, this procedure can be conducted...
Agent-based simulations are powerful tools for simulating emergent mobility modes, but they often require significant memory and computing power. To address this issue, researchers have previously used sampling techniques, where only a fraction of agents are explicitly simulated while others are simulated through teleportation. However, recent stud...
In this paper, nonlinear model predictive control (NMPC) is proposed for autonomous vehicle drifting, that is, stabilizing the vehicle at a desired unstable equilibrium point. Firstly, a three‐degree‐of‐freedom vehicle model with a nonlinear tire model is introduced, and the equilibrium points are calculated. The relationship between the desired un...
Integrating measurements and historical data can enhance control systems through learning-based techniques, but ensuring performance and safety is challenging. Robust model predictive control strategies, like stochastic model predictive control, can address this by accounting for uncertainty. Gaussian processes are often used but have limitations w...
Predictive control, which is based on a model of the system to compute the applied input optimizing the future system behavior, is by now widely used. If the nominal models are not given or are very uncertain, data-driven model predictive control approaches can be employed, where the system model or input is directly obtained from past measured tra...
Many robotic tasks, such as human-robot interactions or the handling of fragile objects, require tight control and limitation of appearing forces and moments alongside sensible motion control to achieve safe yet high-performance operation. We propose a learning-supported model predictive force and motion control scheme that provides stochastic safe...
We propose a model predictive control approach for autonomous vehicles that exploits learned Gaussian processes for predicting human driving behavior. The proposed approach employs the uncertainty about the GP's prediction to achieve safety. A multi-mode predictive control approach considers the possible intentions of the human drivers. While the i...
Analysis of Electrochemical Impedance Spectroscopy (EIS) data for electrochemical systems often consists of defining an Equivalent Circuit Model (ECM) using expert knowledge and then optimizing the model parameters to deconvolute various resistance, capacitive, inductive, or diffusion responses. For small data sets, this procedure can be conducted...
Biotechnology offers many opportunities for the sustainable manufacturing of valuable products. The toolbox to optimize bioprocesses includes extracellular process elements such as the bioreactor design and mode of operation, medium formulation, culture conditions, feeding rates, etc. However, these elements are frequently insufficient for achievin...
The control of manufacturing processes must satisfy high quality and efficiency requirements while meeting safety requirements. A broad spectrum of monitoring and control strategies, such as model- and optimization-based controllers, are utilized to address these issues. Driven by rising demand for flexible yet energy and resource-efficient operati...
Model predictive control presents remarkable potential for the optimal control of dynamic systems. However, the necessity for an online solution to an optimal control problem often renders it impractical for control systems with limited computational capabilities. To address this issue, specialized dimensionality reduction techniques designed for o...
Synthetic microbial communities are promising production strategies that can circumvent, via division of labor, many challenges associated with monocultures in biotechnology. Here, we consider microbial communities as lumped metabolic pathways where their members catalyze different metabolic submodules. We outline a machine learning-supported cyber...
The biotechnology industry can significantly benefit from new paradigms such as smart manufacturing, digitalization and quality-by-design to render more competitive and robust processes. Real-time monitoring of key process parameters and performance indicators can facilitate the transition toward smart biomanufacturing. Since cells are typically us...
Machine learning techniques such as neural networks bear the potential to improve the performance and applicability of model predictive control to real-world systems.
However, they also bear the danger of erratic-unpredictable behavior and malfunctioning of machine learning approaches.
Neural networks might fail to predict system behavior as it i...
Safe collision-free operation of autonomous systems, such as mobile robots in crowded, uncertain, only partially known environments, is challenging. We propose learning a collision-free corridor from demonstration via heteroscedastic Gaussian processes. We incorporate available deterministic obstacle information in the learning procedure to derive...
Collision-free control and planning for autonomous vehicles in an only partially known environment is challenging. Often this problem is tackled passively - without active exploration, i.e., planning and control are performed based on the information obtained during operation without explicitly considering that reducing the environment uncertainty...
Agent-based simulations have become a popular and powerful tool for simulating emergent mobility modes. Often times, the memory and computing requirements are daunting. Scaling down agent populations by simulating only a fraction of all agents is a frequently used option to reduce these burdens. However, recent studies have pointed out the difficul...
Machine learning methods, like Gaussian process regression, allow improving the performance of model-based control methods, such as model predictive control. They can, for example, be used to improve the prediction quality of the used model of the system, learning the uncertain system part. However, fusing model-based approaches with machine learni...
Environmental and economic pressure leads to an increasing desire to operate processes over various operational conditions, adapting to changing conditions such as feed quality, available energy, customer demand, or product prices. This requires frequent changes in the process setpoints, involving transitions between those. We focus on the explicit...
Planning and control for autonomous vehicles usually are hierarchical separated. However, increasing performance demands and operating in highly dynamic environments requires an frequent re-evaluation of the planning and tight integration of control and planning to guarantee safety. We propose an integrated hierarchical predictive control and plann...
