<![CDATA[Effectively pursuing dynamically moving targets in the domain of multi-robot systems (MRS) poses a significant challenge. This paper proposes an innovative leader-follower strategy within the MRS framework, enabling robots to dynamically adjust their roles based on target proximity. This approach fosters coordination, allowing robots to act cohesively when pursuing diverse targets, from other robots to mobile objects. The centralized architecture of the MRS facilitates wireless communication, enabling robots to share sensor-derived data providing proximity cues rather than precise location information. However, data anomalies arising from sensor errors, transmission glitches, or encoding issues pose challenges, compromising the reliability of target-related information. To mitigate this, the paper introduces an advanced methodology integrating the leader-follower strategy with Discriminant Analysis (DA)-based anomaly detection. This novel approach validates and filters data, enhancing data integrity and supporting decision-making processes. The integration of DA methods within the leader-follower strategy is detailed, emphasizing steps in anomaly detection implementation, showcasing robustness in selecting high-quality information for decision-making in dynamic environments. The research's real-world relevance addresses the problem of the impact of sensor anomalies on the performance and reliability of MRS in dynamic environments. By integrating machine learning-based anomaly detection, this methodology enhances MRS adaptability and robustness, particularly in scenarios requiring precise target tracking and coordination. Numerical experiments and simulations demonstrate the efficacy of the DA-based anomaly detection and collaborative hunting strategy in MRS. This method contributes to improved target tracking, enhanced system coordination, and streamlined pursuit of dynamic targets, affirming its practical applicability in surveillance, search and rescue operations, and industrial automation.]]>
