Simplified Design of Dual Quaternion Strapdown Inertial Navigation Integration Algorithms

2012 ◽  
Vol 239-240 ◽  
pp. 1421-1427
Author(s):  
Yu Rong Lin ◽  
Liang Chen ◽  
Zhen Xian Fu

Dual quaternion navigation algorithm gain higher accuracy than traditional strapdown inertial navigation algorithm at the cost of real-time performance. In order to reduce tremendous computation amount of the former, a simplified design scheme for navigation integration algorithms is presented in this paper. First, based on update principle and computation rules of dual quaternion we separate rotational and translational increment information from dual quaternion increment, and deduce exact solutions defined by the spiral vector for thrust velocity increment, gravitational velocity increment and displacement increment. Then, considering characteristics of a strapdown inertial navigation system, implementation schemes of simplified integration algorithms for dual quaternion differential equations in three frames, including thrust velocity coordinates, gravitational velocity coordinates and position coordinates, are designed separately. Under the premise of ensuring the accuracy advantage of the original dual quaternion inertial navigation algorithm, the proposed simplified algorithm significantly improve the computational efficiency. This will lay favorable foundation for engineering realization of the dual quaternion strapdown inertial navigation algorithm.

2012 ◽  
Vol 479-481 ◽  
pp. 2610-2615
Author(s):  
Kai Yao ◽  
Qi Dan Zhu ◽  
Bo Zhang

This paper addresses a practical problem arising in the calibration of bottom-lock doppler velocity log for the navigation of surface ships. Firstly, a dead reckoning navigation algorithm and briefly error analyze are proposed. Then, employing ship’s true trajectory and calculated trajectory, the rotational alignment offset between a bottom-lock doppler velocity log and a strapdown inertial navigation system as well as the scale factor error of the doppler velocity log can be experimentally determined using sensors commonly deployed with a vehicle in the field. It requires velocity values from the vehicle's doppler log and strapdown inertial navigation system, and absolute vehicle position fixes from a GPS receiver. Lake experiment results show that the calibration algorithm can calibrate the error parameters effectively, thus the position error decreases significantly after compensating the error parameters.


Author(s):  
Seong Yun Cho ◽  
Hyung Keun Lee ◽  
Hung Kyu Lee

In this paper, performance of the initial fine alignment for the stationary nonleveling strapdown inertial navigation system (SDINS) containing low-grade gyros is analyzed. First, the observability is analyzed by conducting a rank test of an observability matrix and by investigating the normalized error covariance of the extended Kalman filter based on the ten-state model. The results show that the accelerometer biases on horizontal axes are unobservable. Second, the steady-state estimation errors of the state variables are derived using the observability equation. It is verified that the estimates of the state variables have errors due to the unobservable state variables and nonleveling attitude angles of a vehicle containing the SDINS. Especially, this paper shows that the larger the attitude angles of the vehicle are, the greater the estimation errors are. Finally, it is shown that the performance of the eight-state model excluding the two unobservable state variables is better than that of the ten-state model in the fine alignment by a Monte Carlo simulation.


2012 ◽  
Vol 229-231 ◽  
pp. 1671-1674
Author(s):  
Jian Feng Chen ◽  
Xi Yuan Chen ◽  
Xue Fen Zhu

Recent dramatic progress in strapdown inertial navigation system (SINS) algorithm is the design of SINS principle based on screw algorithm, utilizing dual quaternion. In this paper, the screw algorithm consisting of angular rate and specific force is optimized under a special screw motion. The special screw motion is derived from classical screw motion and can be taken as a complicated sculling motion including classical coning motion. Subsequently, the coefficients in the multi-sample screw algorithms and the corresponding algorithm drifts are determined by minimizing the error on direct component. The simulation results of attitude and velocity errors agree with the optimization goals, except when the number of subinterval is greater than 2. An explanation of this phenomenon is delivered.


Sign in / Sign up

Export Citation Format

Share Document