This paper presents a novel graph-based method for adapting control system architectures at runtime. We use a service-oriented architecture as a basis for its formulation. In our method, adaptation is achieved by selecting the most suitable elements, such as filters and controllers, for a control system architecture to improve control systems objective based on a predefined cost function. Traditional configuration methods, such as state machines, lack flexibility and depend on a predefined control system architecture during runtime. Our graph-based method allows for dynamic changes in the control system architecture, as well as a change in its objective depending on the given system state. Our approach uses a weighted, directed graph to model the control system elements and their interaction. In a case-study with a three-tank system, we show that by using our graph-based method for architecture adaptation, the control system is more flexible, has lower computation time, and higher accuracy than traditional configuration methods.

We propose a learning-based Control Barrier Function (CBF) to reduce conservatism in collision avoidance of car-like robots. Traditional CBFs often use Euclidean distance between robots' centers as safety margin, neglecting headings and simplifying geometries to circles. While this ensures smooth, differentiable safety functions required by CBFs, it can be overly conservative in tight environments. To address this limitation, we design a heading-aware safety margin that accounts for the robots' orientations, enabling a less conservative and more accurate estimation of safe regions. Since the function computing this safety margin is non-differentiable, we approximate it with a neural network to ensure differentiability and facilitate integration with CBFs. We describe how we achieve bounded learning error and incorporate the upper bound into the CBF to provide formal safety guarantees through forward invariance. We show that our CBF is a high-order CBF with relative degree two for a system with two robots whose dynamics are modeled by the nonlinear kinematic bicycle model. Experimental results in overtaking and bypassing scenarios reveal a 33.5 % reduction in conservatism compared to traditional methods, while maintaining safety. Code: https://github.com/bassamlab/sigmarl

Non-stationarity poses a fundamental challenge in Multi-Agent Reinforcement Learning (MARL), arising from agents simultaneously learning and altering their policies. This creates a non-stationary environment from the perspective of each individual agent, often leading to suboptimal or even unconverged learning outcomes. We propose an open-source framework named XP-MARL, which augments MARL with auxiliary prioritization to address this challenge in cooperative settings. XP-MARL is 1) founded upon our hypothesis that prioritizing agents and letting higher-priority agents establish their actions first would stabilize the learning process and thus mitigate non-stationarity and 2) enabled by our proposed mechanism called action propagation, where higher-priority agents act first and communicate their actions, providing a more stationary environment for others. Moreover, instead of using a predefined or heuristic priority assignment, XP-MARL learns priority-assignment policies with an auxiliary MARL problem, leading to a joint learning scheme. Experiments in a motion-planning scenario involving Connected and Automated Vehicles (CAVs) demonstrate that XP-MARL improves the safety of a baseline model by 84.4% and outperforms a state-of-the-art approach, which improves the baseline by only 12.8%. Code: github.com/cas-lab-munich/sigmarl

Distributed control algorithms are known to reduce overall computation time compared to centralized control algorithms. However, they can result in inconsistent solutions leading to the violation of safety-critical constraints. Inconsistent solutions can arise when two or more agents compute concurrently while making predictions on each others control actions. To address this issue, we propose an iterative algorithm called Synchronization-Based Cooperative Distributed Model Predictive Control, which we presented in [1]. The algorithm consists of two steps: 1. computing the optimal control inputs for each agent and 2. synchronizing the predicted states across all agents. We demonstrate the efficacy of our algorithm in the control of multiple small-scale vehicles in our Cyber-Physical Mobility Lab.

The computation time required to solve nonconvex, nonlinear optimization problems increases rapidly with their size. This poses a challenge in trajectory planning for multiple networked vehicles with collision avoidance. In the centralized formulation, the optimization problem size increases with the number of vehicles in the networked control system (NCS), rendering the formulation unusable for experiments. We investigate two methods to decrease the complexity of networked trajectory planning. First, we approximate the optimization problem by discretizing the vehicle dynamics with an automaton, which turns it into a graph-search problem. Our search-based trajectory planning algorithm has a limited horizon to further decrease computation complexity. We achieve recursive feasibility by design of the automaton which models the vehicle dynamics. Second, we distribute the optimization problem to the vehicles with prioritized distributed model predictive control (P-DMPC), which reduces the problem size. To counter the incompleteness of P-DMPC, we propose a framework for time-variant priority assignment. The framework expands recursive feasibility to every vehicle in the NCS. We present two time-variant priority assignment algorithms for road vehicles, one to improve vehicle progress and one to improve computation time of the NCS. We evaluate our approach for online trajectory planning of multiple networked vehicles in simulations and experiments.