Unmanned Aerial Vehicles (UAVs) are utilized extensively across multiple domains. UAVs offer a range of services to users, such as monitoring, logistics and sensing due to their adaptable deployment as well as dynamic reconfigurability. Nonetheless, UAV networks have become increasingly vulnerable to malicious threats due to their multi-connectivity and openness. Significant effort has been expended to create an efficient Intrusion Detection System (IDS) utilizing machine-learning methodologies for UAVs. Unfortunately, limitations of standalone IDS make them inadequate for protecting UAV networks from potential security threats. The lack of accurate identification for compromised UAV nodes in UAV networks is a significant security vulnerability, jeopardizing the integrity of the entire network due to the compromise of a single node. In this chapter, an autonomous collaborative UAV-CIDS utilizing a Convolutional Neural Network (CNN) is presented. This system effectively finds zero-day vulnerabilities with high precision. The suggested approach considers encoded Wi-Fi traffic logs from three prevalent UAV types: DBPower UDI, Parrot Bebop and DJI Spark. The evaluation findings demonstrate that proposed FFCNN model has achieved exceptional performance on the UAVIDS dataset, with an accuracy of 97.21% in comparison to previous models. Following the identification of attacks, their mitigation is equally crucial. Furthermore, real-time incident response protocols are created and executed for cyber-attacks on UAV networks. The incident response management will aid in mitigating the impact of a security breach, addressing vulnerabilities and thoroughly securing the entire UAV networks.