During the 21st IFAC World Congress 2020 one of the many contributions of our lab was publishing PoCET, a Polynomial Chaos Expansion Toolbox for Matlab. It offers a powerful set of routines that allow for fast uncertainty propagation through linear and nonlinear ODE systems and can therefore be used for all kinds of stochastic dynamic optimization problems like parameter estimation, experimental design, or MPC. Moreover, it is available completely open-source and free of charge on GitHub. More information on PoCET can be found on the corresponding research page.
In a new project, the group will contribute its expertise in set-based methods, dynamical models of the infectious disease and optimal control to design and analyse mitigation actions. The project results will hopefully contribute to mitigate the spread of the Corona virus. Further information can be found in this press release from TU Chemnitz. An overview of our projects can be found here.
Our ACSD lab is part of the new research project SOPRANN (synthesis of optimal controls and adaptive neural networks for mobility applications). The project focuses on (proveable) safe learning-based control using neural networks in mobility applications. Further information can be found in this press release from TU Chemnitz. An overview of our projects can be found here.
The IFAC World Congress 2020 is one of the most important control conferences. The ACSD lab will publish and present 16 contributions covering a variety of topics, including theoretical contributions in the areas of learning-based control, model predictive control and set-based system analysis as well as application oriented contributions in the area of energy networks and agricultural systems:
- Maintaining Hard Infection Caps in Epidemics via the Theory of Barriers
- Sustainability Analysis of Interconnected Food Production Systems via Theory of Barriers
- Model Predictive Control of a Food Production Unit: A Case Study for Lettuce Production
- Nonsmooth stabilization and its computational aspects
- Online Traction Parameter Identification and Mapping
- Optimal control of centrifugal spreader
- A reinforcement learning method with closed-loop stability guarantee
- Hierarchical control of a quadcopter under stuck actuator fault
- Half-Gain Tuning for Active Disturbance Rejection Control
- Stabilising the Light Spectrum of LED Solar Simulators using LQG Control
- Prototypical Description and Controller Design for a Set of Systems Using v-gap Based Clustering
- A Hierarchical Architecture for the Coordination of an Ensemble of Steam Generators
- PoCET: a Polynomial Chaos Expansion Toolbox for Matlab
- Reserve Balancing in a Microgrid System for Safety Analysis
- A Hierarchical MPC Scheme for Ensembles of Hammerstein Systems
- A reinforcement learning method with closed-loop stability guarantee for systems with unknown parameters
Prof. Dr. Florin Stoican (Politehnica University of Bucharest) will give a talk on About the use of B-spline functions in motion planning at 4:00 pm in 2/W065. Everybody is heartily invited. All past and future seminars can be found here.
An insight how bio-economically optimised resource and energy cycles help in the production of sustainable food can be found in this press release, which also describes the group's research activities in the area of the production of insect larvae (Hermetia illucens).
Prof. Dr.-Ing. Johann Reger (TU Ilmenau) will give a talk on Exact Backstepping Control for Systems in Pure Feedback Form at 5:30 pm in 2/W014. Everybody is heartily invited. All past and future seminars can be found here.
Starting in 2019, several members of the ACSD lab will be working on Control of closed energy and mass flow of a high complex production system. The lab will closely collaborate with several industrial and academic partners. Further information can be found in this press release from TU Chemnitz, at the project's homepage and in this press release. An overview of our projects can be found here.
The ACSD Lab presents two papers on the Conference on Decision and Control (CDC), which takes place in Nice on December 11th-13th. Dr. Osinenko presents two papers, namely Adaptive Dynamic Programming Using Lyapunov Function Constraints (Göhrt, Osinenko, Streif) and Model Predictive Control with Stage Cost Shaping Inspired by Reinforcement Learning (Beckenbach, Osinenko, Streif). Furthermore, Prof. Streif is co-chair of the Optimal Control V session.
On constructive extractability of measurable selectors of set-valued maps
IEEE Transactions on Automatic Control, 2021 [DOI]
A comprehensive dynamic growth and development model of Hermetia illucens larvae
PLOS ONE, 2020 [DOI]
Optimization and Stabilization of Hierarchical Electrical Networks
Mathematics in Industry: Mathematical MSO for Power Engineering and Management, 2020
A reinforcement learning method with closed-loop stability guarantee for systems with unknown parameters
21st IFAC World Congress, 2020
Recursive feasibility of continuous-time model predictive control without stabilising constraints
IEEE Control Systems Letters, 2021 [DOI]
Control Design with Guaranteed Transient Performance: an Approach with Polyhedral Target Tubes
Automatica, 2020 [DOI]
On Maximal Robust Positively Invariant Sets in Constrained Nonlinear Systems
Automatica, 2020 [DOI]
Nonsmooth stabilization and its computational aspects
21st IFAC World Congress, 2020
Converse optimality for discrete-time systems
IEEE Transactions on Automatic Control, 2020 [DOI]
On closed-loop stability of model predictive controllers with learning costs
European Control Conference (ECC), 2020