Optimal control strategies are often combined with safety certificates to ensure both performance and safety in safety-critical systems. A prominent example is combining Model Predictive Control (MPC) with Control Barrier Functions (CBF). Yet, tuning MPC parameters and choosing an appropriate class kappa function in the CBF is challenging and problem dependent. This thesis introduces a safe model-based Reinforcement Learning (RL) framework where a parameterized MPC incorporates a CBF with a parameterized class kappa function and serves as a function approximator to learn improved safe control policies. Three variations are introduced, distinguished by the way the class kappa function is parameterized. The Learnable Optimal Decay CBF (LOPTD-CBF) extends the Optimal Decay CBF by allowing RL to tune the optimal decay parameters, improving performance while enhancing constraint feasibility and preserving safety guarantees. The Neural Network CBF (NN-CBF) parametrizes the decay term of a discrete exponential CBF with a neural network, enabling richer state-dependent safety conditions. Finally, the Recurrent Neural Network CBF (RNN-CBF) extends the NN-CBF with a recurrent architecture to handle time-varying CBF constraints, such as moving obstacles. Numerical experiments on a discrete double-integrator with static and dynamic obstacles demonstrate that the proposed methods improve performance while ensuring safety, each offering distinct trade-offs in performance, feasibility and complexity. ... Read more

The goal of this project was to develop a sensor system that detects Rapid Eye Movement Sleep Behaviour Disorder (RBD) episodes in an early stage and brings the patient to a lighter sleep stage at the episode onset. This project was completed in close collaboration with Momo Medical, a fast-growing start-up and developer of the BedSense - a device placed under mattresses that tracks the bed posture and restlessness of residents in nursing homes. To detect RBD episodes, the BedSense was used in combination with a sock that houses the biosensors EDA, PPG and an accelerometer. The sock also contains a vibrator module, that can bring the patient to a lighter sleep stage in case of an episode.

The outcomes of this project are promising, data has been recorded from both non-RBD test subjects and an RBD patient, and the vibration module has been tested.

The results of this project were integrated with another project seeking to develop an algorithm to detect RBD episodes in an early stage. This project used data from the Momo BedSense and from the sock that has been created in this project. ... Read more