Data-Driven Optimal Control: An Inverse Optimization Model and Algorithm

In Inverse Optimization (IO), it is hypothesized that experts, when making decisions, implicitly engage in solving an optimization problem. If we can reconstruct this optimization problem using the decision data of the expert, then the behavior of the expert can be emulated. In this thesis, a novel inverse optimization model, Kernel Inverse...

Modelling Flow-based Market Coupling in the CORE region: A Data-driven and Interpretable Approach

Integrating renewables in the electricity system in a cost-efficient way requires massive transmission system investments and the efficient use of available transmission capacity. Markets are pivotal in the latter, especially in coordinating flows between countries. In the European Union, flow-based market coupling (FBMC) arose as the preferred...

Integrated Forecasting and Scheduling of Implicit Demand Response in Balancing Markets Using Inverse Optimization

Demand Response (DR) programs offer flexibility that is considered to hold significant potential for enhancing power system reliability and promoting the integration of renewable energy sources. Nevertheless, the distributed nature of DR resources presents challenges in developing scalable optimization tools. This paper explores a novel data...

Online and offline learning of player objectives from partial observations in dynamic games

Robots deployed to the real world must be able to interact with other agents in their environment. Dynamic game theory provides a powerful mathematical framework for modeling scenarios in which agents have individual objectives and interactions evolve over time. However, a key limitation of such techniques is that they require a priori...

Learning Drivers’ Preferences in Delivery Route Planning: an Inverse Optimization Approach

Optimizing delivery routes is a well-researched topic, however, most of the classical approaches do not incorporate preferences of drivers, as those approaches focus on minimizing the time or distance of the routes. As a result, the actual driven route of an experienced driver often deviates from the proposed route since the drivers have tacit...

Learning to Make Routing Decisions Based on Expert Examples

Logistics and mobility services play a major role in our society, and efficient routing is a crucial part of this. However, even though routing problems have been widely researched, the solutions provided by algorithms do not always match drivers' expectations. Routing costs used by these algorithms are often based on one or a few parameters,...

Learning for Control: An Inverse Optimization Approach

We present a learning method to learn the mapping from an input space to an action space, which is particularly suitable when the action is an optimal decision with respect to a certain unknown cost function. We use an inverse optimization approach to retrieve the cost function by introducing a new loss function and a new hypothesis class of...

From Supervised to Reinforcement Learning: an Inverse Optimization Approach

We propose a novel method combining elements of supervised- and Q-learning for the control of dynamical systems subject to unknown disturbances. By using the Inverse Optimization framework and in-hindsight information we can derive a causal parametric optimization policy that approximates a non-causal MPC expert. Furthermore, we propose a new...

Learning Parametric Integer Programming via Inverse Optimization

Inverse Learning is implemented in order to learn a control/decision policy (in the integer space) from an Expert Agent. The Learner Agent assumes that the Expert is acting minimizing an unknown cost function and tries to approximate it, through its own parametrized version of it. Learning can be performed in two different ways: offline ...

IOC based trajectory generation to increase human acceptance of robot motions in collaborative tasks

Collaboration between humans and robots is an important aspect of Industry 4.0. It can be improved by incorporating human-like characteristics into robot motion planning. It is assumed that humans move optimal with respect to a certain objective or cost function. To find this function, also for a robot, we use an inverse optimal control...

Learning for Control: An Inverse Optimization Approach

Data driven learning has become a common practice in many decision making applications. Machine learning techniques are the dominant methods employed but have limitations with respect to constraints satisfaction. A learning method is presented to learn the mapping from an input space to an action space, which is particularly suitable when the...

Human-like overtaking maneuvers using Inverse Optimal Control

In recent years, rapid progress has been made towards automated highway driving. To increase comfort and trust in these automated cars, their motion planners, such as Model Predictive Controllers (MPCs), should exhibit a human-like driving style. However the effect of surrounding traffic on overtaking maneuvers has not yet been investigated....

Inverse Optimal Control for robot arm motions: Improving human-robot collaboration by incorporating human characteristics into optimal motion generation

Human-robot collaboration can be improved if the motions of the robot are more legible and predictable. This can be achieved by making the motions more human-like. It is assumed that humans move optimal with respect to a certain objective or cost function. To find this function an inverse optimal control approach is developed. It uses a bilevel...

Variable Circular Collimator in Robotic Radiosurgery: A Time-Efficient Alternative to a Mini-Multileaf Collimator?

urpose Compared with many small circular beams used in CyberKnife treatments, beam’s eye view-shaped fields are generally more time-efficient for dose delivery. However, beam’s eye view-shaping devices, such as a mini-multileaf collimator (mMLC), are not presently available for CyberKnife, although a variable-aperture collimator (Iris, 12 field...