摘要
针对多旋翼飞行器在复杂风场环境下存在抖振和响应滞后的问题,提出一种基于PSoC硬件平台的神经网络增强型自适应滑模控制(ASMC)方法。该方法利用长短期记忆网络替换传统自适应滑模控制中的固定扰动上界估计模型,实现对滑模参数λ和切换增益ρ的动态优化调节。然后基于集成NPU的PSoC异构计算架构实现了算法的高效部署。硬件加速性能测试结果表明,PSoC硬件平台单样本处理时间仅需81.7μs,支持12.24 kHz的高频闭环控制。仿真结果表明,相较于ASMC方法,该方法将系统俯仰/滚转角收敛时间从0.62 s缩短至0.27 s,滚转角抖振幅值由±1.66°抑制至±0.14°,最大偏移速度仅为±1.1 m/s。
This paper addresses the issues of chattering and response lag in multi-rotor aircraft under complex wind field conditions by proposing a neural network-enhanced adaptive sliding mode control(ASMC)method based on a PSoC hardware platform.The method replaces the fixed disturbance upper-bound estimation model in traditional adaptive sliding mode control with a long short-term memory network,enabling dynamic optimization of the sliding mode parameterλand the switching gainρ.Furthermore,an efficient deployment of the algorithm is achieved using the heterogeneous computing architecture of PSoC with an integrated NPU.Hardware acceleration performance tests demonstrate that the PSoC hardware platform requires only 81.7μs for single-sample processing,supporting high-frequency closed-loop control at 12.24 kHz.Simulation results indicate that,compared to ASMC,the proposed method reduces the convergence time of the system pitch/roll angles from 0.62 s to 0.27 s,suppresses the roll angle chattering amplitude from±1.66°to±0.14°,and limits the maximum movement speed is to±1.1 m/s.
作者
杨茂林
王池
方明
刘晓萌
周明源
朱广胜
YANG Maolin;WANG Chi;FANG Ming;LIU Xiaomeng;ZHOU Mingyuan;ZHU Guangsheng(China Electronics Technology Group Corporation No.58 Research Institute,Wuxi 214035,China;China Electronics Technology Group Corporation No.32 Research Institute,Shanghai 201808,China;Wuxi Esiontech Co.,Ltd.,Wuxi214072,China)
出处
《电子与封装》
2026年第4期102-107,共6页
Electronics & Packaging
