New High-Precision Strapdown Navigation Attitude Algorithm under Angular-Rate Input Condition

Coning motion is a standard test input to evaluate the performance of the strapdown attitude algorithms. Angular-rate coning algorithms error consists of two parts: drift error and approximation error. Traditional angular-rate coning algorithms usually improve the algorithm performance by increasing the sampling number in one update period. However the increase of the sampling number can only reduce the drift error, it has few effects on reducing approximation error. And the approximation error compensation is neglected in traditional angular-rate coning algorithms. In this paper the calculation result shows that the approximation error is comparable with drift error for the most general case, which means the approximation error can not be neglected in high-precision strapdown navigation systems. A new angular-rate coning algorithm with an additional second-order noncommutativity error compensation term is developed. Without increasing sampling number, the new angular-rate coning algorithm can reduce the approximation error greatly. Theoretical analyses and digital simulations indicate that the new algorithm has advantages over the traditional coning algorithms for the general case.