Air Combat Maneuver Trajectory Prediction Model of Target Based on Chaotic Theory and IGA-VNN

Table 2

Target X-coordinate multistep prediction performance comparison of four different prediction methods.

Prediction step	Algorithm	RMSE	MAE	MAPE	COR
1	BP	31.2054	5.2515	0.0011	0.9987
	IGA-VNN	25.0554	4.2594		0.9984
	GA-VNN	46.6161	6.2959	0.0016	0.9983
	Volterra	88.0949	9.2494	0.0035	0.9985
	RBF	65.5883	7.4046	0.0023	0.9987
2	BP	58.2029	7.1255	0.0021	0.9972
	IGA-VNN	43.1102	5.6694	0.0013	0.9969
	GA-VNN	51.5107	6.6589	0.0018	0.9961
	Volterra	103.4790	9.4192	0.0036	0.9920
	RBF	103.9374	8.9800	0.0033	0.9888
4	BP	100.8075	9.0122	0.0032	0.9983
	IGA-VNN	64.8278	7.2519	0.0021	0.9952
	GA-VNN	130.0170	10.6018	0.0046	0.9935
	Volterra	227.3807	12.1373	0.0059	0.9816
	RBF	171.4466	11.3985	0.0053	0.9636
6	BP	151.8142	11.9705	0.0058	0.9978
	IGA-VNN	115.1146	9.7368	0.0038	0.9866
	Volterra	200.2440	13.0592	0.0070	0.9980
	RBF	238.1826	13.9321	0.0079	0.9531