Dynamic Planning of Optimally-safe Lane-change Trajectory by Fuzzy Logic-based Collision Estimator

What is it about?

The collision avoidance system within autonomous vehicles plays a pivotal role in enhancing traffic safety, employing a critical function known as collision estimation. This involves the identification of potential hazards and timely notification to drivers or autonomous control to navigate safely. This research introduces an innovative methodology for generating and selecting a lane change trajectory in scenarios where two vehicles are concurrently executing lane changes on highways and approaching the same target lane. Additionally, a novel fuzzy logic estimator, based on time-to-collision (TTC) and time-to-gap (TTG), is developed to assess collision risk. During the collision avoidance process, the proposed estimator evaluates the risk of collision through polynomial function-based generation of feasible lane change trajectories. The safest trajectory is then communicated to the motion controller, enabling secure navigation through challenging lane change scenarios. The study also explores the application of Stanley and Pure Pursuit controllers to follow the optimized trajectory.

Featured Image

Photo by Axel Antas-Bergkvist on Unsplash

Photo by Axel Antas-Bergkvist on Unsplash

Why is it important?

The research introduces a pivotal advancement in autonomous vehicle safety by presenting an innovative method for generating lane change trajectories. This method proves particularly crucial in navigating scenarios involving simultaneous lane changes on highways, addressing a key challenge in autonomous vehicle operation. The inclusion of a fuzzy logic estimator enhances collision risk assessment by considering time-to-collision and time-to-gap. The approach utilizes polynomial function-based trajectory generation, providing the safest path to the motion controller. Simulation experiments confirm the effectiveness of the methodology in preventing potential collisions in dynamic scenarios, emphasizing its contribution to advancing overall traffic safety in autonomous vehicles.

Perspectives