KIT | KIT-Bibliothek | Impressum | Datenschutz

Deep Decentralized Reinforcement Learning for Cooperative Control

Köpf, Florian; Tesfazgi, Samuel; Flad, Michael; Hohmann, Sören

Abstract:
In order to collaborate effciently with unknown partners in cooperative control settings, adaptation of the partners based on online experience is required. The rather general and widely applicable control setting, where each cooperation partner might strive for individual goals while the control laws and objectives of the partners are unknown, entails various challenges such as the non-stationarity of the environment, the multi-agent credit assignment problem, the alter-exploration problem and the coordination problem. We propose new, modular deep decentralized Multi-Agent Reinforcement Learning mechanisms to account for these challenges. Therefore, our method uses a time-dependent prioritization of samples, incorporates a model of the system dynamics and utilizes variable, accountability-driven learning rates and simulated, artificial experiences in order to guide the learning process. The effectiveness of our method is demonstrated by means of a simulated, nonlinear cooperative control task.



Zugehörige Institution(en) am KIT Institut für Regelungs- und Steuerungssysteme (IRS)
Publikationstyp Proceedingsbeitrag
Publikationsjahr 2020
Sprache Englisch
Identifikator KITopen-ID: 1000117640
Erschienen in IFAC-PapersOnLine
Veranstaltung 21st IFAC World Congress (2020), Berlin, Deutschland, 12.07.2020 – 17.07.2020
Schlagwörter Reinforcement Learning, Deep Learning, Learning Control, Shared Control, Decentralized Control, Machine Learning, Non-stationary Systems, Nonlinear Control
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft
KITopen Landing Page