Deep Learning Based Data Fusion for Sensor Fault Diagnosis and Tolerance in Autonomous Vehicles

Chinese Journal of Mechanical Engineering

Table 1 Feature extraction network structure

	Type	Channel	Filter size	Output
	Input	3		416 \(\times\) 416
	Convolutional	16	3 \(\times\) 3/4	104 \(\times\) 104
	Convolutional	32	3 \(\times\) 3/2	52 \(\times\) 52
2\(\times\)	Convolutional	32	3 \(\times\) 3
	Convolutional	32	3 \(\times\) 3
	Residual			52 \(\times\) 52
	Convolutional	64	3 \(\times\) 3/2	26 \(\times\) 26
2\(\times\)	Convolutional	64	3 \(\times\) 3
	Convolutional	64	3 \(\times\) 3
	Residual			26 \(\times\) 26
	Convolutional	128	3 \(\times\) 3/2	13 \(\times\) 13
2\(\times\)	Convolutional	128	3 \(\times\) 3
	Convolutional	128	3 \(\times\) 3
	Residual			13 \(\times\) 13
	Required computing power: 0.686 \(\times\) 10⁹ FLOPs