Fault tolerant adaptive neuro-fuzzy based automated cruise controller on FPGA

The objective of the paper is to implement a system which effectively reduces vehicle collisions by using a combination of image processing and neuro-fuzzy based techniques. It uses an image processing module in order to determine the distance of the front vehicle. The distance measurements are given to a sugeno type adaptive neuro-fuzzy system. Since the operation of the system is mission critical in nature, a fault tolerant field programmable gate array (FPGA) implementation of the system is proposed to balance the reliability, availability and performance criteria. Finally, a comparison is made with respect to difference in errors and speed, between neural network and adaptive neuro-fuzzy based methods.