<p>The online parameter identification of permanent magnet synchronous motors (PMSMs) has become a crucial technical requirement for ensuring the robustness and reliability of motor control systems. However, existing swarm intelligence algorithms suffer from low convergence accuracy and slow convergence speed. To address the challenges of slow convergence and low accuracy in parameter identification for PMSMs, this paper proposes an improved artificial lemming algorithm (IALA). The proposed IALA advances the standard artificial lemming algorithm (ALA) through three key modifications. First, a piecewise chaotic map is employed to initialize the population, enhancing its diversity. Second, a Gaussian walk strategy is integrated into the long-distance migration phase to bolster local search performance. Third, a fitness-based competitive attack strategy is introduced during the foraging phase to augment the global search capability. The performance of the IALA was compared with the standard ALA, particle swarm optimization (PSO), and the grey wolf optimizer (GWO) using four benchmark functions and a PMSM parameter identification model. Both simulation and experimental results demonstrate that the proposed IALA significantly outperforms the other algorithms in terms of convergence speed and identification accuracy, while achieving precise online identification of PMSM parameters. The IALA provides an effective and reliable method for the online identification of PMSM parameters, which is essential for developing high-performance motor control systems.</p>