Date of Award
Master of Science
Department of Electrical and Computer Engineering
Gilbert L. Peterson, PhD
Cooperative agent and robot systems are designed so that each is working toward the same common good. The problem is that the software systems are extremely complex and can be subverted by an adversary to either break the system or potentially worse, create sneaky agents who are willing to cooperate when the stakes are low and take selfish, greedy actions when the rewards rise. This research focuses on the ability of a group of agents to reason about the trustworthiness of each other and make decisions about whether to cooperate. A trust-based interactive partially observable Markov decision process (TI-POMDP) is developed to model the trust interactions between agents, enabling the agents to select the best course of action from the current state. The TI-POMDP is a novel approach to multiagent cooperation based on an interactive partially observable Markov decision process (I-POMDP) augmented with trust relationships. Experiments using the Defender simulation demonstrate the TI-POMDP's ability to accurately track the trust levels of agents with hidden agendas The TI-POMDP provides agents with the information needed to make decisions based on their level of trust and model of the environment. Testing demonstrates that agents quickly identify the hidden trust levels and mitigate the impact of a deceitful agent in comparison with a trust vector model. Agents using the TI-POMDP model achieved 3.8 times the average reward of agents using a trust vector model.
DTIC Accession Number
Seymour, Richard S., "The Trust-Based Interactive Partially Observable Markov Decision Process" (2019). Theses and Dissertations. 2562.