The Modeling and Prediction of Strapdown Inertial Measurement Unit based on Support Vector Regression and Particle Swarm Optimization

Strapdown inertial measurement unit (SIMU) installed in a vehicle determines the navigation accuracy of the vehicle. However, the stability of SIMU changes with the storage time passing by, which influences the navigation accuracy of the vehicle. Thus, it is very necessary to establish a predication model which can predict the stability of SIMU and evaluate the performance of SIMU. This paper proposes a method to predict the SIMU stability. This algorithm is based on support vector regression and particle swarm optimization. The results expressed that support vector regression and particle swarm optimization can forecast the changing trend of SIMU accurately, which can provide the theoretical basis for evaluating the SIMU’s performance.

[1]  Youxian Sun,et al.  Gas flow-rate measurement using a transit-time multi-path ultrasonic flow meter based on PSO-SVM , 2017, 2017 IEEE International Instrumentation and Measurement Technology Conference (I2MTC).

[2]  Hong Bao,et al.  A method for people counting using feature fusion based on SVR with PSO optimization , 2017, 2017 IEEE 9th International Conference on Communication Software and Networks (ICCSN).

[3]  Yating Zhang,et al.  Comparative study on several PSO algorithms , 2014, The 26th Chinese Control and Decision Conference (2014 CCDC).

[4]  Huan Wang,et al.  PSO-SVR: A Hybrid Short-term Traffic Flow Forecasting Method , 2015, 2015 IEEE 21st International Conference on Parallel and Distributed Systems (ICPADS).

[5]  Hao Tang,et al.  Reliable Prediction System Based on Support Vector Regression with Genetic Algorithms , 2009, 2009 Fifth International Conference on Natural Computation.