AI-Enhanced Mimic Defense for Securing Microservices in Consumer Autonomous Platforms

Consumer unmanned systems are increasingly deployed in emerging applications including logistics, automotive, and smart homes. These platforms depend on microservice architectures to support flexible service scheduling and scalable software deployment. However, the adopted container technologies and the shift to microservice architectures also bring new security challenges. Traditional solutions rely mainly on passive detection for known vulnerabilities and remain ineffective against emerging attacks. This paper introduces a novel AI-enhanced mimic defense system that utilizes random Heterogeneous Redundant Systems (HRSs) to rebalance the dynamics of attack and defense in these crucial consumer electronics sectors. The cornerstone of our approach lies in the trustworthy arbitration of these systems. Through the analysis of historical system log data, we develop arbitration strategies using offline reinforcement learning, capable of distinguishing reliable and trustworthy messages within consumer electronics IoT environments. Our approach significantly enhances security through proactive defense mechanisms that dynamically manage the asymmetry between attack and defense.