Modeling FOG Drift Using Back-Propagation Neural Network Optimized by Artificial Fish Swarm Algorithm

Table 1

Allan variance analysis of FOG’s output.

	Original data	After denoising data	Compensation result
			GABP	AFSA
𝑄/𝜇rad	149.37	129.58	1.48	1.17
𝑁/(°/h^1/2)	0.94	0.79	0.009	0.007
𝐵/(°/h)	28.09	24.52	0.14	0.09
𝐾/(°/h^3/2)	176.34	154.15	3.64	1.72
𝑅/(°/h^2)	229	197.57	5.12	2.45