01626380

Component

This study addresses the problem of curved roads obstacle avoidance for autonomous vehicles. An optimal control strategy is proposed. The critical conditions of collision between a vehicle and an obstacle are determined while considering the shape and the size of both the vehicle and the obstacle. The reference path is generated by fuzzy potential method (FPM) considering the shape and the size of the vehicle. The vehicle tracks the path while modifying it by optimization using model predictive control (MPC) which computes an optimal input under prediction based on the dynamics until finite-time future in order to achieve smooth and efficient obstacle avoidance. Coordinates transformation is conducted to transform curved road coordinates into straight road coordinates and then into ego-vehicle-based coordinates to make the proposed strategy realized by only the ego-vehicle instead of global operation. Simulation results showed that the proposed strategy is effective in curved roads obstacle avoidance in the case of both a large and a small radius of curvature. The efforts were made to improve curved road safety and enrich the research on obstacle avoidance.

This paper was sponsored by TRB committee AHB30 Standing Committee on Vehicle-Highway Automation.

01618707

17-05528

English

18p

2017

Transportation Research Board 96th Annual Meeting

Digital/other

Figures; References (27)

Automated vehicle control for ground vehicles; Autonomous land vehicles; Highway curves; Obstructions (Navigation); Proximity detectors; Vehicle trajectories

Highways; Safety and Human Factors; Vehicles and Equipment

Transportation Research Board Annual Meeting 2017 Paper #17-05528

TRIS, TRB, ATRI

Dec 8 2016 12:11PM