Target fast recognition and risk identification based on dynamic programming

Target fast recognition directly affects the performance of driving assistance systems (DAS). In order to improve the accuracy of target recognition and real-time of risk identification in complex driving conditions based on vehicular radar, a dynamic programming method is proposed. Before target validity examination, the horizontal and longitudinal searching area is estimated real-time to recognize the target fast and accurately. Tests show that the method can recognize the risk target faster and identify hazard more efficiently in three typical dangerous conditions: lane changing, vehicle cutting-in and following on curve road. © 2011 IEEE.