Real-time states estimation of a farm tractor using dynamic mode decomposition

• DOI: 10.1007/s10291-020-01051-5
• Publication Date: 2020-11-13T07:02:44Z
• Authors: Hao Wang; Noboru Noguchi
• Keywords: Dynamic Mode Decomposition; Yaw; GPS/INS; Least-squares function approximation

Abstract We present a pure data-driven method to estimate vehicle dynamics from the measurements of sideslip and yaw rate in use GPS inertial navigation system. The INS configuration provides position, velocity vector, orientation, observations. A new dynamic mode decomposition with control (DMDc) denoises state observations by adopting total least-squares algorithm. DMD (tlsDMDc) helps discover underlying time-dependent states external control. experiments simulated linear model synthetic Gaussian noise illustrate that solutions tlsDMDc are more accurate than standard DMDc characterize imperfect measurements. additionally investigate how algorithm performs on motion deduction sensor bias correction. It has been shown tlsDMDc-based estimator couple robust observation presence error measurement noise.