Unmanned Aerial Vehicles (UAVs) are integral to diverse applications, spanning surveillance, search and rescue operations, and precision agriculture. This paper addresses a critical challenge in these domains offering efficient and reliable target tracking, especially in the context of multiple UAV deployments. It presents an innovative approach, Volf-RL (Voronoi based-Grey Wolf Optimisation-Reinforcement Learning), leveraging Voronoi subspaces to optimize the spatial distribution of UAVs, coupled with the integration of the Grey Wolf Optimizer (GWO) for advanced target tracking capabilities. To enhance overall mission safety and efficiency, we further incorporate reinforcement learning techniques, ensuring collision avoidance among UAVs. This multifaceted approach aims to create a sophisticated UAV swarm optimization system. The collaboration of Voronoi subspaces, GWO and Deep-Q Networks not only refines the spatial arrangement of UAVs but also contributes to dynamic target tracking in complex environments. The primary objectives of this research include achieving optimal area coverage for target detection, addressing collision avoidance challenges, exploring moving swarm formations for comprehensive surveillance, and refining UAV-based target tracking accuracy using GWO. The amalgamation of computational geometry, meta-heuristic techniques, and reinforcement learning underscores the commitment of the work towards advancing UAV swarm intelligence. This research lays the foundation for an intelligent, coordinated, and adaptive UAV swarm optimization system, poised to excel in challenging environments with an accuracy of 96.77% in collision avoidance and 98.79% accuracy in target tracking. Through this successful realization of these objectives, we anticipate a paradigm shift in UAV swarm capabilities, fostering advancements in defense, transportation, and logistics.