Efficient sensor deployment in Wireless Sensor Networks (WSNs), a cornerstone of the Internet of Things (IoT), is critical for optimal operation, particularly in environments with abundant obstacles. This paper proposes GAMGWO, a Multi-strategy Grey Wolf Optimizer inspired by Generative Adversarial Networks, to address the challenges of node deployment in such complex physical environments. By incorporating a GAN-inspired mechanism for generating candidate solutions, GAMGWO enhances population diversity. It is guided by a multi-objective fitness function that holistically balances coverage, network connectivity, and node distribution uniformity. To overcome the typical limitations of GWO, such as premature convergence, GAMGWO integrates a Virtual Force Algorithm to prevent node clustering, a non-linear convergence factor to balance exploration and exploitation, and a memory pool to preserve elite solutions. Simulation results demonstrate that GAMGWO significantly outperforms existing GWO variants, achieving a superior average coverage rate of 95.66% in obstacle-free conditions and 89.21% in complex obstacle-rich scenarios, while ensuring 93.33% full network connectivity. These findings highlight GAMGWO’s robustness and reliability, positioning it as an effective solution for practical sensor deployment in challenging real-world IoT applications.