Time-Optimal Trajectory Planning and Tracking for Autonomous Vehicles

Area of Science:

• Robotics
• Control Systems
• Autonomous Driving

Background:

• Accurate trajectory tracking is crucial for autonomous vehicles, especially during high-speed, dynamic maneuvers.
• Existing methods often struggle to balance trajectory optimization with real-time tracking performance.

Purpose of the Study:

• To develop a hierarchical control framework for autonomous vehicle trajectory planning and tracking.
• To enable accurate following of time-optimal trajectories during aggressive driving scenarios.

Main Methods:

• A hierarchical framework combining offline trajectory optimization (TRO) and online nonlinear model predictive control (NMPC).
• TRO uses direct collocation for minimum-lap-time trajectory generation.
• NMPC with a preview algorithm ensures real-time tracking accuracy.

Main Results:

• The framework successfully tracked time-optimal racelines at an average speed of 116 km/h on the Catalunya circuit.
• Maximum lateral error was limited to 0.32 m.
• The NMPC module achieved millisecond-level computation times using an acados solver with an RTI scheme.

Conclusions:

• The proposed hierarchical control framework effectively addresses autonomous vehicle trajectory planning and tracking challenges.
• Real-time implementation is feasible due to efficient NMPC computation.
• This approach enhances the performance of autonomous vehicles in dynamic driving conditions.