In this research, recognition sensors are used for tracking the outside world along with a steering assistance model that reduces delays inherent in the human cognition, decision, and operation decision-making chain, thereby minimizing vehicle behavior delay with respect to the vehicle position and steering wheel angle based on a read-ahead effect. Additionally, the creation of an emergency avoidance assistance program based on obstacle detection and a cornering assistance program based on white line detection are reported. The effectiveness of these programs using a driving simulator and a model vehicle was investigated. It was found that when the emergency avoidance assist provided by the obstacle detection program was used, the automatic override function was capable of successfully intervening to prevent accidents in situations where it determined that manual steering operations by the driver would be too late. Additionally, in experiments involving cornering assistance by white line detection, it is found that smoother steering around curves was facilitated by the system's ability to set up a optimal approach earlier than could be expected by the curve recognition processes used by human drivers, and that the vehicle was more stable at the curve exit.
[1]
Koji Suzuki,et al.
Development of Pre-Crash Safety System with Pedestrian Collision Avoidance Assist
,
2013
.
[2]
Julius Ziegler,et al.
Video based localization for Bertha
,
2014,
2014 IEEE Intelligent Vehicles Symposium Proceedings.
[3]
Sebastian Thrun,et al.
Towards fully autonomous driving: Systems and algorithms
,
2011,
2011 IEEE Intelligent Vehicles Symposium (IV).
[4]
T. Ogawa,et al.
Lane Recognition Using On-vehicle LIDAR
,
2006,
2006 IEEE Intelligent Vehicles Symposium.
[5]
John McDonald,et al.
Application of the Hough Transform to Lane Detection and Following on High Speed Roads
,
2001
.