PROJECT TITLE :

Real-Time Trajectory Planning for Autonomous Urban Driving: Framework, Algorithms, and Verifications

ABSTRACT:

This paper focuses on the real-time trajectory planning downside for autonomous vehicles driving in realistic urban environments. To resolve the complicated navigation downside, we adopt a hierarchical motion coming up with framework. First, a rough reference path is extracted from the digital map using commands from the high-level behavioral planner. The conjugate gradient nonlinear optimization algorithm and therefore the cubic B-spline curve are used to smoothen and interpolate the reference path sequentially. To follow the refined reference path in addition to handle both static and moving objects, the trajectory designing task is decoupled into lateral and longitudinal designing problems inside the curvilinear coordinate framework. A made set of kinematically possible path candidates are generated to accommodate the dynamic traffic each deliberatively and reactively. In the meanwhile, the velocity profile generation is performed to boost driving safety and luxury. After that, the generated trajectories are fastidiously evaluated by an objective operate, which combines behavioral selections by reasoning concerning the traffic things. The optimal collision-free, smooth, and dynamically possible trajectory is chosen and reworked into commands executed by the low-level lateral and longitudinal controllers. Field experiments have been carried out with our test autonomous vehicle on the realistic inner-town roads. The experimental results demonstrated capabilities and effectiveness of the proposed trajectory designing framework and algorithms to safely handle a selection of typical driving scenarios, like static and moving objects avoidance, lane keeping, and vehicle following, whereas respecting the traffic rules.