Performance enhancement of INS/CNS integration navigation system based on particle swarm optimization back propagation neural network

• Published in: Ocean engineering Vol. 108; pp. 33 - 45
• Main Authors: Wang, Qiuying, Li, Yibing, Diao, Ming, Gao, Wei, Qi, Zhao
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.11.2015
• Subjects: Artificial neural networks; Back propagation; BP neural network; Computer simulation; Inertial navigation; Integrated navigation; Marine; Navigation; Neural networks; Particle swarm optimization; Sensors; Star Sensor; Stars; Swarm intelligence; BP neural network; Star Sensor; Integrated navigation; Particle swarm optimization; Inertial navigation
• ISSN: 0029-8018; 1873-5258
• DOI: 10.1016/j.oceaneng.2015.07.062

For the surface ship integrated navigation system of INS/CNS, the Star Sensor may invalidity due to the cloud weather, which leads to the integrated system cannot work anymore. To resolve the problem, an INS/CNS integrated navigation method based on particle swarm optimization back propagation neural network (PSO BPNN) is proposed in this paper. During the effective Star Sensor navigation process, the INS positioning error can be obtained and used for training PSO BPNN; when the Star Sensor is in invalid state, the already trained BPNN is used for forecasting the INS positioning error. The effectiveness of this approach was demonstrates by simulation and experimental study. The results showed that the INS/CNS integrated navigation method based on PSO BPNN can effectively estimate and compensate the INS navigation error under the star senor invalid state. •The Marine INS/CNS integrated navigation method is proposed.•The BPNN is introduced in the integration navigation method.•PSO algorithm is adopted to set BPNN weights and thresholds.•The effectiveness of this approach was demonstrated by simulation and experimental.