Resilient multi-UAV path planning for effective and reliable information collection

Abstract

UAV technology provides opportunities for regional information collection in complex and dynamic environments. In many large-scale or intricate scenarios, such as widespread natural disasters, UAVs often encounter challenges in wirelessly transmitting collected data in real time, which may compromise the timeliness and relevance of the information. Additionally, UAV malfunctions or failures can result in the loss of critical data from specific target points. Therefore, when planning flight paths of multiple UAVs, it becomes essential to coordinate their flight paths to maintain data freshness while minimizing the risk of data loss. In this paper, we introduce a resilient path planning strategy to proactively respond to potential UAV failures. Our approach seeks to reduce discrepancies in the total volume of information carried by individual UAVs, therefore reducing the risk of substantial information loss. Given that traditional weighting methods often heavily rely on subjective coefficient settings, we have established a multi-objective resilient path planning model for multi-UAV information collection scenarios. To optimize UAV flight paths, we propose an improved non-dominated sorting whale optimization algorithm, which provides a more robust and adaptive solution to UAV coordination. Experimental results validated the effectiveness of the constructed mathematical model. Comparative analysis demonstrates that the Pareto front solutions generated by the proposed algorithm hold significant advantages in terms of distribution and uniformity.