Research on Key Technologies of Robot Target Recognition and Intelligent Localization Based on Improved PSO-BP Algorithm

With the widespread application of robot technology in target recognition and intelligent localization, issues such as slow convergence and low accuracy in traditional BP neural networks and Particle Swarm Optimization (PSO) algorithms have gradually become prominent. To address these problems, an improved PSO-BP algorithm is proposed, which optimizes PSO by incorporating chaotic sequences and adaptive strategies. Simultaneously, PSO is utilized to optimize the initial weights and biases of the BP neural network, enhancing recognition accuracy and localization precision. Experimental results demonstrate that the improved PSO-BP algorithm exhibits higher robustness and stability under various lighting and noise conditions. In an environment with 300 lux lighting and low noise, the recognition accuracy of this algorithm reaches 92.5%, with a localization error of 1.12 cm, significantly outperforming traditional algorithms. Furthermore, the algorithm reduces the error to 0.05 within 100 iterations, showcasing rapid convergence. In summary, the improved PSO-BP algorithm demonstrates high accuracy, stability, and efficiency in robot target recognition and localization, providing an effective solution for robot applications in complex environments.