A Case Study of Human-AI Interactions Using Transparent AI-Driven Autonomous Systems for Improved Human-AI Trust Factors

Human interactions with Artificial Intelligence (AI) have become a ubiquitous phenomenon in our society. It plays a major role in defense and espionage in the context of AI-based autonomous systems such as drones and ground robots that are sent to the front lines to navigate terrains, map the surroundings, and identify adversaries in the environment. In these types of scenarios, hostages may be spread out over a larger geographical area. Reliance on ground vision alone may not only be obscured and insufficient but also can expose soldiers or human rescuers to danger. Since time is of great essence, such an intense situation calls for decisions to be made swiftly and efficiently as the lives of hostages are at stake. This makes such a scenario ideal to evaluate the effectiveness of an AI-driven drone autonomous system for combat search and rescue (CSAR) missions. Here the focus is to gather information about the operational environment (detect hostages and other targets-of-interest) in a short amount of time that spans a varying geographic terrain. This allows for targeted rescue operations with AI-informed decision-making outcomes to ensure that the rescuers at ground level can be directed along the most optimal path, and casualties will be either minimized or entirely averted. In this work, we study the feasibility of our drone-based AI target detection autonomous system to evaluate the classification accuracy of multi-view targets-of-interest in the CSAR environment. The overarching goal of this study is to understand the role of AI-driven autonomous systems as an assistive support tool for efficient, swift and accurate decision making, navigating a complex and ever-changing environment, and identification of targets.