Unmanned Aerial Vehicles (UAVs) or drones gained major attention in different application areas such as law enforcement, military, industrial automation, etc. The hostile operating environment integrated with the UAV’s dependency on wireless communication protocols poses an enhanced threat level. Several attacks against UAVs are developing conventional as they can be easy to conduct with low-cost hardware like spoofing and jamming. Unfortunately, many vulnerabilities occur in essential technologies, so securing the UAV develops a complex task. A promising method to identify and mitigate these attacks is the progress of intelligent intrusion detection systems (IDS). With this inspiration, this research projects an Artificial intelligence Driven Intelligent Intrusion Detection for Cybersecure UAV Networks (AIDID-CUAVN) technique. The main aim of the AIDID-CUAVN model is to safe the UAV network communication via the intrusion detection process. In the presented AIDID-CUAVN technique, Perceptive Craving Game Search (PCGS) technique was utilized for feature selection. For intrusion detection, a deep sparse autoencoder (DSAE) classifier is exploited. At the last phase, the coati optimization algorithm (COA) was applied for an optimal parameter selection process. The experimental result of the AIDID-CUAVN system was performed on a benchmark database. The wide comparative study revealed that the AIDID-CUAVN technique outperformed the other models regarding different evaluation metrics.