He obtained his Ph.D. (Dr. sc.) degree in Dynamic Systems and Control from ETH Zurich under the supervision of Raffaello D’Andrea in 2013. Before, he received a B.Sc. degree in General Engineering Science in 2005, a M.Sc. degree (Dipl.-Ing.) in Electrical Engineering in 2007, and an MBA degree in Technology Management in 2007, all from Hamburg University of Technology. In 2007, Sebastian was a research scholar at University of California at Berkeley.
Control Machine learning Networked systems Distributed Intelligence Intelligent physical systems Robotics
In Proceedings of the International Conference on Machine Learning (ICML), International Conference on Machine Learning (ICML), June 2019 (inproceedings) Accepted
Policy gradient methods are powerful reinforcement learning algorithms and have been demonstrated to solve many complex tasks. However, these methods are also data-inefficient, afflicted with high variance gradient estimates, and frequently get stuck in local optima. This work addresses these weaknesses by combining recent improvements in the reuse of off-policy data and exploration in parameter space with deterministic behavioral policies. The resulting objective is amenable to standard neural network optimization strategies like stochastic gradient descent or stochastic gradient Hamiltonian Monte Carlo. Incorporation of previous rollouts via importance sampling greatly improves data-efficiency, whilst stochastic optimization schemes facilitate the escape from local optima. We evaluate the proposed approach on a series of continuous control benchmark tasks. The results show that the proposed algorithm is able to successfully and reliably learn solutions using fewer system interactions than standard policy gradient methods.
IEEE Internet of Things Journal, 6(3):5013-5028, June 2019 (article)
The Internet of Things (IoT) interconnects multiple physical devices in large-scale networks. When the 'things' coordinate decisions and act collectively on shared information, feedback is introduced between them. Multiple feedback loops are thus closed over a shared, general-purpose network. Traditional feedback control is unsuitable for design of IoT control because it relies on high-rate periodic communication and is ignorant of the shared network resource. Therefore, recent event-based estimation methods are applied herein for resource-aware IoT control allowing agents to decide online whether communication with other agents is needed, or not. While this can reduce network traffic significantly, a severe limitation of typical event-based approaches is the need for instantaneous triggering decisions that leave no time to reallocate freed resources (e.g., communication slots), which hence remain unused. To address this problem, novel predictive and self triggering protocols are proposed herein. From a unified Bayesian decision framework, two schemes are developed: self triggers that predict, at the current triggering instant, the next one; and predictive triggers that check at every time step, whether communication will be needed at a given prediction horizon. The suitability of these triggers for feedback control is demonstrated in hardware experiments on a cart-pole, and scalability is discussed with a multi-vehicle simulation.
In Proceedings of the 10th ACM/IEEE International Conference on Cyber-Physical Systems, pages: 97-108, 10th ACM/IEEE International Conference on Cyber-Physical Systems, April 2019 (inproceedings)
Closing feedback loops fast and over long distances is key to emerging applications; for example, robot motion control and swarm coordination require update intervals below 100 ms. Low-power wireless is preferred for its flexibility, low cost, and small form factor, especially if the devices support multi-hop communication. Thus far, however, closed-loop control over multi-hop low-power wireless has only been demonstrated for update intervals on the order of multiple seconds. This paper presents a wireless embedded system that tames imperfections impairing control performance such as jitter or packet loss, and a control design that exploits the essential properties of this system to provably guarantee closed-loop stability for linear dynamic systems. Using experiments on a testbed with multiple cart-pole systems, we are the first to demonstrate the feasibility and to assess the performance of closed-loop control and coordination over multi-hop low-power wireless for update intervals from 20 ms to 50 ms.
Proceedings of the 18th ACM/IEEE Conference on Information Processing in Sensor Networks (IPSN), pages: 340-341, 18th ACM/IEEE Conference on Information Processing in Sensor Networks (IPSN), April 2019 (poster)
IEEE Transactions on Control Systems Technology, 2019 (article) Accepted
Bayesian optimization is proposed for automatic
learning of optimal controller parameters from experimental
data. A probabilistic description (a Gaussian process) is used
to model the unknown function from controller parameters to
a user-defined cost. The probabilistic model is updated with
data, which is obtained by testing a set of parameters on the
physical system and evaluating the cost. In order to learn fast,
the Bayesian optimization algorithm selects the next parameters
to evaluate in a systematic way, for example, by maximizing
information gain about the optimum. The algorithm thus iteratively
finds the globally optimal parameters with only few
experiments. Taking throttle valve control as a representative
industrial control example, the proposed auto-tuning method is
shown to outperform manual calibration: it consistently achieves
better performance with a low number of experiments. The
proposed auto-tuning framework is flexible and can handle
different control structures and objectives.
In Proceeding of the 15th Latin American Robotics Symposium, João Pessoa, Brazil, 15th Latin American Robotics Symposium, November 2018 (inproceedings)
The development of accurate control systems for underwater robotic vehicles relies on the adequate compensation for hydrodynamic effects. In this work, a new robust control scheme is presented for remotely operated underwater vehicles. In order to meet both robustness and tracking requirements, sliding mode control is combined with Gaussian process regression. The convergence properties of the closed-loop signals are analytically proven. Numerical results confirm the stronger improved performance of the proposed control scheme.
In International Conference on Intelligent Robots and Systems (IROS) 2018, pages: 6199-6206, International Conference on Intelligent Robots and Systems 2018, October 2018 (inproceedings)
Soft microrobots based on photoresponsive materials and controlled by light fields can generate a variety of different gaits. This inherent flexibility can be exploited to maximize their locomotion performance in a given environment and used to adapt them to changing environments. However, because of the lack of accurate locomotion models, and given the intrinsic variability among microrobots, analytical control design is not possible. Common data-driven approaches, on the other hand, require running prohibitive numbers of experiments and lead to very sample-specific results. Here we propose a probabilistic learning approach for light-controlled soft microrobots based on Bayesian Optimization (BO) and Gaussian Processes (GPs). The proposed approach results in a learning scheme that is highly data-efficient, enabling gait optimization with a limited experimental budget, and robust against differences among microrobot samples. These features are obtained by designing the learning scheme through the comparison of different GP priors and BO settings on a semisynthetic data set. The developed learning scheme is validated in microrobot experiments, resulting in a 115% improvement in a microrobot’s locomotion performance with an experimental budget of only 20 tests. These encouraging results lead the way toward self-adaptive microrobotic systems based on lightcontrolled soft microrobots and probabilistic learning control.
Hertneck, M., Koehler, J., Trimpe, S., Allgöwer, F.
IEEE Control Systems Letters, 2(3):543-548, July 2018 (article)
A supervised learning framework is proposed to approximate a model predictive controller (MPC) with reduced computational complexity and guarantees on stability and constraint satisfaction. The framework can be used for a wide class of nonlinear systems. Any standard supervised learning technique (e.g. neural networks) can be employed to approximate the MPC from samples. In order to obtain closed-loop guarantees for the learned MPC, a robust MPC design is combined with statistical
learning bounds. The MPC design ensures robustness to inaccurate inputs within given bounds, and Hoeffding’s Inequality is used to validate that the learned MPC satisfies these bounds with high confidence. The result is a closed-loop statistical guarantee on stability and constraint satisfaction for the learned MPC. The proposed learning-based MPC framework is illustrated on a nonlinear benchmark problem, for which we learn a neural network controller with guarantees.
In Proceedings of the International Conference on Machine Learning (ICML), International Conference on Machine Learning (ICML), July 2018 (inproceedings)
State-space models (SSMs) are a highly expressive model class for learning patterns in time series data and for system identification. Deterministic versions of SSMs (e.g., LSTMs) proved extremely successful in modeling complex time-series data. Fully probabilistic SSMs, however, unfortunately often prove hard to train, even for smaller problems. To overcome this limitation, we propose a scalable initialization and training algorithm based on doubly stochastic variational inference and Gaussian processes. In the variational approximation we propose in contrast to related approaches to fully capture the latent state temporal correlations to allow for robust training.
In Proceedings of the IEEE Workshop on Benchmarking Cyber-Physical Networks and Systems (CPSBench), pages: 13-18, IEEE Workshop on Benchmarking Cyber-Physical Networks and Systems (CPSBench), April 2018 (inproceedings)
Proceedings of the 56th IEEE Annual Conference on Decision and Control (CDC), pages: 5193-5200, IEEE, IEEE Conference on Decision and Control, December 2017 (conference)
Finding optimal feedback controllers for nonlinear dynamic systems from data
is hard. Recently, Bayesian optimization (BO) has been proposed as a powerful
framework for direct controller tuning from experimental trials. For selecting
the next query point and finding the global optimum, BO relies on a
probabilistic description of the latent objective function, typically a
Gaussian process (GP). As is shown herein, GPs with a common kernel choice can,
however, lead to poor learning outcomes on standard quadratic control problems.
For a first-order system, we construct two kernels that specifically leverage
the structure of the well-known Linear Quadratic Regulator (LQR), yet retain
the flexibility of Bayesian nonparametric learning. Simulations of uncertain
linear and nonlinear systems demonstrate that the LQR kernels yield superior
Proceedings of 1st Annual Conference on Robot Learning (CoRL), 78, pages: 227-238, (Editors: Sergey Levine and Vincent Vanhoucke and Ken Goldberg), 1st Annual Conference on Robot Learning, November 2017 (conference)
We propose a novel long-term optimization criterion to improve the robustness of model-based reinforcement learning in real-world scenarios. Learning a dynamics model to derive a solution promises much greater data-efficiency and reusability compared to model-free alternatives. In practice, however, modelbased RL suffers from various imperfections such as noisy input and output data, delays and unmeasured (latent) states. To achieve higher resilience against such effects, we propose to optimize a generative long-term prediction model directly with respect to the likelihood of observed trajectories as opposed to the common approach of optimizing a dynamics model for one-step-ahead predictions. We evaluate the proposed method on several artificial and real-world benchmark problems and compare it to PILCO, a model-based RL framework, in experiments on a manipulation robot. The results show that the proposed method is competitive compared to state-of-the-art model learning methods. In contrast to these more involved models, our model can directly be employed for policy search and outperforms a baseline method in the robot experiment.
IET Control Theory & Applications, 11(11):1684-1693, July 2017 (article)
An event-based state estimation approach for reducing communication in a networked control system is proposed. Multiple distributed sensor agents observe a dynamic process and sporadically transmit their measurements to estimator agents over a shared bus network. Local event-triggering protocols ensure that data is transmitted only when necessary to meet a desired estimation accuracy. The event-based design is shown to emulate the performance of a centralised state observer design up to guaranteed bounds, but with reduced communication. The stability results for state estimation are extended to the distributed control system that results when the local estimates are used for feedback control. Results from numerical simulations and hardware experiments illustrate the effectiveness of the proposed approach in reducing network communication.
In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), pages: 5295-5301, IEEE, Piscataway, NJ, USA, IEEE International Conference on Robotics and Automation (ICRA), May 2017 (inproceedings)
In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), pages: 1557-1563, IEEE, Piscataway, NJ, USA, IEEE International Conference on Robotics and Automation (ICRA), May 2017 (inproceedings)
The International Journal of Robotics Research, 35(14):1731-1749, December 2016 (article)
The Gaussian Filter (GF) is one of the most widely used filtering algorithms; instances are the Extended Kalman Filter, the Unscented Kalman Filter and the Divided Difference Filter. The GF represents the belief of the current state by a Gaussian distribution, whose mean is an affine function of the measurement. We show that this representation can be too restrictive to accurately capture the dependences in systems with nonlinear observation models, and we investigate how the GF can be generalized to alleviate this problem. To this end, we view the GF as the solution to a constrained optimization problem. From this new perspective, the GF is seen as a special case of a much broader class of filters, obtained by relaxing the constraint on the form of the approximate posterior. On this basis, we outline some conditions which potential generalizations have to satisfy in order to maintain the computational efficiency of the GF. We propose one concrete generalization which corresponds to the standard GF using a pseudo measurement instead of the actual measurement. Extending an existing GF implementation in this manner is trivial. Nevertheless, we show that this small change can have a major impact on the estimation accuracy.
In Proceedings of the American Control Conference (ACC), Boston, MA, USA, July 2016 (inproceedings)
Most widely-used state estimation algorithms, such as the Extended Kalman Filter and the Unscented Kalman Filter, belong to the family of Gaussian Filters (GF). Unfortunately, GFs fail if the measurement process is modelled by a fat-tailed distribution. This is a severe limitation, because thin-tailed measurement models, such as the analytically-convenient and therefore widely-used Gaussian distribution, are sensitive to outliers. In this paper, we show that mapping the measurements into a specific feature space enables any existing GF algorithm to work with fat-tailed measurement models. We find a feature function which is optimal under certain conditions. Simulation results show that the proposed method allows for robust filtering in both linear and nonlinear systems with measurements contaminated by fat-tailed noise.
In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), pages: 270-277, IEEE, IEEE International Conference on Robotics and Automation, May 2016 (inproceedings)
This paper proposes an automatic controller tuning framework based on linear optimal control combined with Bayesian optimization. With this framework, an initial set of controller gains is automatically improved according to a pre-defined performance objective evaluated from experimental data. The underlying Bayesian optimization algorithm is Entropy Search, which represents the latent objective as a Gaussian process and constructs an explicit belief over the location of the objective minimum. This is used to maximize the information gain from each experimental evaluation. Thus, this framework shall yield improved controllers with fewer evaluations compared to alternative approaches. A seven-degree- of-freedom robot arm balancing an inverted pole is used as the experimental demonstrator. Results of a two- and four- dimensional tuning problems highlight the method’s potential for automatic controller tuning on robotic platforms.
In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA) 2016, IEEE, IEEE International Conference on Robotics and Automation, May 2016 (inproceedings)
We consider the problem of model-based 3D- tracking of objects given dense depth images as input. Two difficulties preclude the application of a standard Gaussian filter to this problem. First of all, depth sensors are characterized by fat-tailed measurement noise. To address this issue, we show how a recently published robustification method for Gaussian filters can be applied to the problem at hand. Thereby, we avoid using heuristic outlier detection methods that simply reject measurements if they do not match the model. Secondly, the computational cost of the standard Gaussian filter is prohibitive due to the high-dimensional measurement, i.e. the depth image. To address this problem, we propose an approximation to reduce the computational complexity of the filter. In quantitative experiments on real data we show how our method clearly outperforms the standard Gaussian filter. Furthermore, we compare its performance to a particle-filter-based tracking method, and observe comparable computational efficiency and improved accuracy and smoothness of the estimates.
Max Planck Institute for Intelligent Systems, November 2015 (techreport)
An event-based state estimation approach for reducing communication in a networked control system is proposed. Multiple distributed sensor-actuator-agents observe a dynamic process and sporadically exchange their measurements and inputs over a bus network. Based on these data, each agent estimates the full state of the dynamic system, which may exhibit arbitrary inter-agent couplings. Local event-based protocols ensure that data is transmitted only when necessary to meet a desired estimation accuracy. This event-based scheme is shown to mimic a centralized Luenberger observer design up to guaranteed bounds, and stability is proven in the sense of bounded estimation errors for bounded disturbances. The stability result extends to the distributed control system that results when the local state estimates are used for distributed feedback control. Simulation results highlight the benefit of the event-based approach over classical periodic ones in reducing communication requirements.
Machine Learning in Planning and Control of Robot Motion Workshop at the IEEE/RSJ International Conference on Intelligent Robots and Systems (iROS), pages: , , Machine Learning in Planning and Control of Robot Motion Workshop, October 2015 (conference)
This paper proposes an automatic controller tuning framework based on linear optimal control combined with Bayesian optimization. With this framework, an initial set of controller gains is automatically improved according to a pre-defined performance objective evaluated from experimental data. The underlying Bayesian optimization algorithm is Entropy Search, which represents the latent objective as a Gaussian process and constructs an explicit belief over the location of the objective minimum. This is used to maximize the information gain from each experimental evaluation. Thus, this framework shall yield improved controllers with fewer evaluations compared to alternative approaches. A seven-degree-of-freedom robot arm balancing an inverted pole is used as the experimental demonstrator. Preliminary results of a low-dimensional tuning problem highlight the method’s potential for automatic controller tuning on robotic platforms.
In Proceedings of the American Control Conference, July 2015 (inproceedings)
This paper presents an LMI-based synthesis procedure for distributed event-based state estimation. Multiple agents observe and control a dynamic process by sporadically exchanging data over a broadcast network according to an event-based protocol. In previous work , the synthesis of event-based state estimators is based on a centralized design. In that case three different types of communication are required: event-based communication of measurements, periodic reset of all estimates to their joint average, and communication of inputs. The proposed synthesis problem eliminates the communication of inputs as well as the periodic resets (under favorable circumstances) by accounting explicitly for the distributed structure of the control system.
In Proceedings of the IEEE International Conference on Robotics and Automation, May 2015 (inproceedings)
An event-based communication framework for remote operation of a robot via a bandwidth-limited network is proposed. The robot sends state and environment estimation data to the operator, and the operator transmits updated control commands or policies to the robot. Event-based communication protocols are designed to ensure that data is transmitted only when required: the robot sends new estimation data only if this yields a significant information gain at the operator, and the operator transmits an updated control policy only if this comes with a significant improvement in control performance. The developed framework is modular and can be used with any standard estimation and control algorithms. Simulation results of a robotic arm highlight its potential for an efficient use of limited communication resources, for example, in disaster response scenarios such as the DARPA Robotics Challenge.
An exploded power plant, collapsed buildings after an earthquake, a burning vehicle loaded with hazardous goods – all of these are dangerous situations for human emergency responders. What if we could send robots instead of humans? Researchers at the Autonomous Motion Department work on fundamental principles required to build intelligent robots which one day can help us in dangerous situations. A key requirement for making this happen is that robots must be enabled to learn.
In Robotics: Science and Systems, 2015 (inproceedings)
The Gaussian Filter (GF) is one of the most widely used filtering algorithms; instances are the Extended Kalman Filter, the Unscented Kalman Filter and the Divided Difference Filter. GFs represent the belief of the current state by a Gaussian with the mean being an affine function of the measurement. We show that this representation can be too restrictive to accurately capture the dependencies in systems with nonlinear observation models, and we investigate how the GF can be generalized to alleviate this problem. To this end we view the GF from a variational-inference perspective, and analyze how restrictions on the form of the belief can be relaxed while maintaining simplicity and efficiency. This analysis provides a basis for generalizations of the GF. We propose one such generalization which coincides with a GF using a virtual measurement, obtained by applying a nonlinear function to the actual measurement. Numerical experiments show that the proposed Feature Gaussian Filter (FGF) can have a substantial
performance advantage over the standard GF for systems with nonlinear observation models.
In Proceedings of the 53rd IEEE Conference on Decision and Control, Los Angeles, CA, 2014 (inproceedings)
An approach for distributed and event-based state estimation that was proposed in previous work  is analyzed and extended to practical networked systems in this paper. Multiple sensor-actuator-agents observe a dynamic process, sporadically exchange their measurements over a broadcast network according to an event-based protocol, and estimate the process state from the received data. The event-based approach was shown in  to mimic a centralized Luenberger observer up to guaranteed bounds, under the assumption of identical estimates on all agents. This assumption, however, is unrealistic (it is violated by a single packet drop or slight numerical inaccuracy) and removed herein. By means of a simulation example, it is shown that non-identical estimates can actually destabilize the overall system. To achieve stability, the event-based communication scheme is supplemented by periodic (but infrequent) exchange of the agentsâ?? estimates and reset to their joint average. When the local estimates are used for feedback control, the stability guarantee for the estimation problem extends to the event-based control system.
Our goal is to understand the principles of Perception, Action and Learning in autonomous systems that successfully interact with complex environments and to use this understanding to design future systems