Performance evaluation of decision making strategies for an embedded lane departure warning system

An on-board lane departure warning system embedded in a vehicle is composed of a localization module and a decision making module. The decision making module detects unintended lane departure so as to warn the driver of the danger. The performance of a decision making module is crucial to the performance of the total embedded system. This article proposes two heuristic decision making strategies: a lateral offset (LO) based strategy and a time-to-lane crossing (TLC) based strategy. The performance criteria of decision making strategies are proposed as: (1) false alarm rate, and (2) alarm triggering time (ATT). Numerical parameters of both strategies are optimized through numerical simulation, taking the performance criteria into consideration. The proposed strategies are incorporated into the prototype system and evaluated in real expressway experiments. The comparative study of both methods with experimental results shows the applicability of the on-board lane departure warning system. © 2002 Wiley Periodicals, Inc.