Track-centric Iterative Learning for Global Trajectory Optimization in Autonomous Racing

TL;DR

The paper tackles the challenge of global, full-horizon trajectory optimization for autonomous racing under uncertain dynamics. It proposes a track-centric paradigm that directly learns and optimizes the full-horizon trajectory using a wavelet-based parameterization and Bayesian optimization, with simulation-based evaluation under a learned residual Gaussian Process model. An iterative loop collects real-world data to update the residual dynamics, progressively refining the trajectory and closing the gap between planned and executed lap times. Theoretical guarantees plus extensive simulations and 1/10-scale hardware experiments show up to 20.7% improvement over a nominal baseline and consistent outperformance of baselines.

Abstract

This paper presents a global trajectory optimization framework for minimizing lap time in autonomous racing under uncertain vehicle dynamics. Optimizing the trajectory over the full racing horizon is computationally expensive, and tracking such a trajectory in the real world hardly assures global optimality due to uncertain dynamics. Yet, existing work mostly focuses on dynamics learning at the tracking level, without updating the trajectory itself to account for the learned dynamics. To address these challenges, we propose a track-centric approach that directly learns and optimizes the full-horizon trajectory. We first represent trajectories through a track-agnostic parametric space in light of the wavelet transform. This space is then efficiently explored using Bayesian optimization, where the lap time of each candidate is evaluated by running simulations with the learned dynamics. This optimization is embedded in an iterative learning framework, where the optimized trajectory is deployed to collect real-world data for updating the dynamics, progressively refining the trajectory over the iterations. The effectiveness of the proposed framework is validated through simulations and real-world experiments, demonstrating lap time improvement of up to 20.7% over a nominal baseline and consistently outperforming state-of-the-art methods.

Figures (8)

• Figure 1: (a) Trajectory optimization with an inaccurate nominal model yields a suboptimal trajectory; (b) Standard MPC exhibits large tracking errors due to model mismatch; (c) MPC with learned dynamics reduces tracking error but follows a suboptimal reference; and (d) Our proposed framework iteratively refines the trajectory to improve the lap time.
• Figure 2: Schematic of the bicycle model described in the Frenet frame.
• Figure 3: Overall architecture of iterative track-centric learning framework.
• Figure 4: Expressivity gap in conventional vs. wavelet-based trajectory parameterizations: under a limited parameter number, (a) wavelet-based parameterization can accurately capture different trajectory trends; and (b) can represent rich samples near the optimal trajectory.
• Figure 5: Comparison after 10 iterations for each method. The solid colored line indicates the planned trajectories (colored by speed), the black dotted line shows the trajectories evaluated by the simulation-enabled BO using the learned dynamics, and the red solid line displays the actual trajectories on the real track.

Theorems & Definitions (4)

• Proposition 1
• proof : Proof
• Theorem 1
• proof