基于改进样本卷积交互网络的车辆组合导航系统研究

Research on vehicle integrated navigation system based on improved sample convolutional interaction network

KUANGXinghong，YANBiyun

（SchoolofEngineering，ShanghaiOceanUniversity，Shanghai20l3o6，China）

Abstract：Global Navigation Satelite System/Inertial NavigationSystem （GNSS/INS）integrated navigation system in vehicles is prone tosignal loss inobstructed environments，leading todivergent positioning results andcompromising the efficiencyand safetyof unmanned vehicles.Toaddressthis issue，thisstudy proposed anartificial inteligence solution based onanimproved Sample Convolutionand Interaction Network （SClNet）， which incorporated strategies such as principal component analysis，trend decomposition，and linear convolutional interactive learningonalow-layerSCINetarchitecture，enhancing thestabilityandaccuracyof the model under such operating conditions.The results show thatthe proposed model reduces positioning errors by 80.9% and 67.6% compared to Long Short-Term Memory （LSTM）and SCINet， respectively，effectively improving theoutdoor positioning accuracyof unmanned vehiclesduring GNSS signal lossandensuringthereliabilityand safety of unmanned vehicle positioning.

Keywords： unmanned vehicles； integrated navigation； inertial navigation system （INS）outage；sample convolutional interaction network （SCINet）； trend seasonal separating

随着时代的进步和科技的飞速发展，无人车在日常生活等多个领域得到了广泛运用。（剩余14308字）