Paper Image

Lane change control for autonomous vehicles

Published on:

28 March 2024

Primary Category:

Systems and Control

Paper Authors:

Sangjae Bae,

David Isele,

Alireza Nakhaei,

Peng Xu,

Alexandre Miranda Anon,

Chiho Choi,

Kikuo Fujimura,

Scott Moura

Bullets

Key Details

Presents a two-stage control approach combining model predictive control and neural networks

Neural networks predict interactive motions of surrounding drivers

Generates smooth lane change trajectories for passenger comfort

Incorporates adaptive safety and robustness enhancements

Simulation results validate effectiveness across traffic scenarios

AI generated summary

Lane change control for autonomous vehicles

This paper presents a method for autonomous vehicles to smoothly and safely change lanes in dense traffic, where gaps between vehicles are small. It uses a two-stage approach combining model predictive control and generative adversarial networks to plan smooth trajectories while predicting how surrounding drivers will react. The system is made more robust with adaptive safety boundaries and filters to reduce sensor noise. Simulations demonstrate the approach is effective, comfortable, and safe across varying traffic densities and cooperativeness levels.

Answers from this paper

Comments

No comments yet, be the first to start the conversation...

Sign up to comment on this paper

Sign Up