Fuzzy control of an autonomous car using a smart phone

This paper presents preliminary results on the development of a vision-based controller with a smart phone in the loop for autonomous vehicles. Our approach involves the use of the low cost vision sensors available on a smart phone to implement speed control for a lead-follower application, under stop-and-go traffic conditions. The controller uses fuzzy logic to determine commands based on estimated car position and speed with respect to the target vehicle being followed. We define constraints for system parameters associated with the minimum resolution and distance required for the controller to operate efficiently. The proposed design was successfully demonstrated on the University of Arizona CAT Vehicle by having it identify and follow a chosen human-driven lead vehicle.