Multi-human-machine Collaborative Virtual-Real Fusion Carrier-based Aircraft Aviation Support Operations Electronic Sand Table Decision System

The aviation support operational environment of aircraft carriers is characterized by high dynamism, with diverse and complex sources of operational situation information. Existing sand table systems have significant limitations in terms of the comprehensiveness of situational information display and the efficiency of operational plan simulations. This constrains commanders’ ability to intuitively understand and make real-time adjustments to aviation support operational plans. In response to these issues, this study designs and implements a multi-human and machine collaborative virtual-real fusion carrier-based aircraft aviation support operations electronic sand table decision system. First, we establish a multi-source situational information fusion environment oriented toward cross-device collaborative interaction analysis. Through fusing situation awareness data obtained from multiple channels to achieve dynamic and complete situational presentation. Furthermore, we propose a mixed reality (MR) based multi-human-machine collaborative interactive visualization method for operational plan simulation and correction. This method optimizes the machine-recommended carrier-based aircraft operational plan through multi-human-machine collaboration while utilizing virtual-real mapping technology to provide global visualized situational evolution process of the deck for adjustments. Additionally, by leveraging commanders’ intuitive reasoning capabilities, it relies on group collaboration mechanisms to integrate the experiences of multiple commanders, mitigates information asymmetry, and ultimately computes more optimized simulation plans. Finally, we validate the system using a multi-human-machine collaborative simulation of aviation support operations on the deck of a large naval vessel as a test scenario. The experimental results demonstrate the system’s validity and advantages, showing that compared to current mainstream machine intelligent algorithms, the planned operational paths achieved an average 2% reduction in length, an 80% decrease in total station traversals, and a 60.8% improvement in path smoothness. This provides scientific and reliable reference support for on-site decision-making in aviation support tasks.