水下机器人T-S型模糊神经网络控制

T-S fuzzy neural network control for autonomous underwater vehicles

针对水下机器人模糊神经网络控制器运算量大和对强外界扰动的鲁棒性差及存在滞后性的问题,提出基于混合学习算法的水下机器人T-S型模糊神经网络控制方法。采用免疫遗传算法离线优化和神经网络自学习在线调整隶属函数的参数,从而减少神经网络的运算量,增强水下机器人对环境变化的反应能力。采用T-S模型,由后件网络动态调整模糊规则,提高控制系统的适应性。通过某微小型水下机器人的仿真和外场实验验证方法的可行性和优越性。实验结果表明,控制器对外界扰动具有较强的鲁棒性,保证即使在恶劣情况下,控制性能仍保持在较高水平。

Aiming at heavy calculation and low robustness and response hysteresis to strong disturbance of fuzzy neural network controller for autonomous underwater vehicles,T-S fuzzy neural network control based on hybrid learning algorithm was proposed.The parameters of the membership function are opti-mized by immune genetic algorithm off line and neural network on line to reduce the calculation of neural network and enhance the response ability to environment changing.Moreover,T-S model is used to adjust the dynamic fuzzy rules by the latter neural network,which can improve the adaptability of the control system.Finally,simulation and lake experiments are carried out to verify the feasibility and superiority.The results show that the control system has great robustness,which assures the high control performance even under formidable circumstance.