Improving Traffic Safety and Efficiency with Multi-Objective Lane Change Assistance

Agile lane changes pose a high risk of collisions, leading to increased incident severity and injuries. This paper aims to develop a safety-based lane changing framework that generates safe strategies and actions for high-risk lane change scenarios. The framework effectively balances the safety and mobility of all vehicles involved in a lane change. The study employs a two-stage approach using the dual-objective Non-dominated Sorting Genetic Algorithm II (NSGA-II) to generate optimal lane change strategies. This method searches for strategies under various dynamic and static conditions by utilizing both risk and mobility scores to enhance traffic safety and mobility efficiency. Using real-world data from the HighD dataset, the study demonstrates the framework’s effectiveness in simultaneously improving traffic safety and mobility. On average, the frame can reduce the safety score by up to 42% and improve the mobility score by 50% for all vehicles inovolving in agile lane changing events.