Efficient Communication in Robust Multi-agent Reinforcement Learning: Trading Observational Robustness for Fewer Communications

Reinforcement learning, especially deep reinforcement learning, has made many advances in the last decade. Similarly, great strides have been made in multi-agent reinforcement learning. Systems of cooperative autonomous robots are increasingly being used, for which multi-agent reinforcement learning can be used as a training method. However, the...

Distributed multi-agent magnetic field norm SLAM with Gaussian processes

Accurately estimating the positions of multi-agent systems in indoor environments is challenging due to the lack of Global Navigation Satelite System (GNSS) signals. Noisy measurements of position and orientation can cause the integrated position estimate to drift without bound. Previous research has proposed using magnetic field simultaneous...

Sample-efficient multi-agent reinforcement learning using learned world models

Multi-agent robotic systems could benefit from reinforcement learning algorithms that are able to learn behaviours in a small number trials, a property known as sample efficiency. This research investigates the use of learned world models to create more sample-efficient algorithms. We present a novel multi-agent model-based reinforcement...

Exploring the effects of conditioning Independent Q-Learners on the sufficient plan-time statistic for Dec-POMDPs

The Decentralized Partially Observable Markov Decision Process is a commonly used framework to formally model scenarios in which multiple agents must collaborate using local information. A key difficulty in a Dec-POMDP is that in order to coordinate successfully, an agent must decide on actions not only using its own information, but also by...

A Guide to Solving Pathfinding Problems with Multiple Agents

Currently, literature regarding Multiagent Path Finding (MAPF) does not give a broad enough overview of all the different approaches. Many papers are hard to read and require proper knowledge of MAPF. The goal of this report is to give a global overview of MAPF. To achieve this goal, we provide a detailed explanation of what MAPF problems look...

A dynamic multi-agent approach for modeling the evolution of multi-hazard accident scenarios in chemical plants

In the chemical industry, multi-hazard (toxic, flammable, and explosive) materials such as acrylonitrile are stored, transported, and processed in large quantities. A release of multi-hazard materials can simultaneously or sequentially lead to acute toxicity, fire and explosion. The spatial-temporal evolution of hazards may also result in...

Modelling MECA scenarios using Brahms and KAoS

In this report we present a simulation of a MECA scenario, powered by Brahms and KAoS. To achieve this, the KAoS/Brahms bridge is developed, which allows agents in a Brahms multi-agent system to query KAoS policies. With the KAoS/Brahms bridge at hand, the MECA project is used as context for a scenario modelled in Brahms. The scenario is...