We present a closed-loop framework for autonomous raceline optimization that combines NURBS-based trajectory representation, CMA-ES global trajectory optimization, and controller-guided spatial feedback. Instead of treating tracking errors as transient disturbances, our method exploits them as informative signals of local track characteristics via a Kalman-inspired spatial update. This enables the construction of an adaptive, acceleration-based constraint map that iteratively refines trajectories toward near-optimal performance under spatially varying track and vehicle behavior. In simulation, our approach achieves a 17.38% lap time reduction compared to a controller parametrized with maximum static acceleration. On real hardware, tested with different tire compounds ranging from high to low friction, we obtain a 7.60% lap time improvement without explicitly parametrizing friction. This demonstrates robustness to changing grip conditions in real-world scenarios.
Fig. 3: Track segmentation into acceleration (red) and deceleration (blue) zones for blame region identification. Since the trajectory does not contain any neutral phases, these are not represented in the figure.
Fig. 4: F1Tenth experimental platform with Asus NUC and Hokuyo LiDAR.
Map
Ours
Ours
Min Curvature
Homeomorphism
Static
Adaptive
Static
Static
F1Aut
20.02
16.54
22.12
21.83
Wall1
16.82
15.71
17.32
16.92
Levine
11.08
10.42
11.39
12.18
Operngasse
7.49
6.24
8.37
7.58
TABLE I: Lap times in s for different trajectories and racetracks.
Fig. 5: Optimizer convergence on F1Aut track. Red line: Error feedback activation. Convergence is achieved within 8 laps, with 500 iterations/lap.
Fig. 6: llustration of the two artificially introduced low-friction regions (highlighted in gray) and the corresponding evolution of the optimized trajectory across three laps, showing the nominal planned trajectory (red-dotted line), the executed trajectory (blue line), and updated constraint map (shaded blue and red regions).
Fig. 7: Lap times for three tire configurations (Low/Medium/High friction). Error feedback activated at lap 0.
Fig. 8: Optimized trajectories under different global friction conditions showing the lateral acceleration a⊥ , the longitudinal acceleration a∥ , and the acceleration constraints amax from the constraint map. Higher-friction setups produce trajectories with increased average velocity and slightly longer path lengths, while lower-friction setups result in reduced speeds and more circular trajectories.
Autonomous racecars operate at the limits of vehicle dynamics, where small control errors translate into safety-critical behavior and lost performance. Trajectory planners assume perfect tracking and remain blind to execution errors. To guarantee safety, trajectory planners therefore restrict themselves to conservative spatial margins, leaving usable track space untapped. To overcome these issues, we introduce a control-informed online trajectory planning framework that learns from its own execution errors. By measuring systematic tracking deviations during runtime, we dynamically adapt spatial track constraints and iteratively expand the free-space planning area. The planner remains time-optimal while compensating for accumulated execution errors. This method was analyzed in a high-fidelity closed-loop simulation environment with autonomous racecars. The results demonstrate that our approach reduces lap time by 1.8,s without increasing computational burden, maintaining a median runtime of 25 ms. Our finding indicates that feeding control-induced deviations back into the planning layer unlocks performance previously inaccessible to modular architectures and enables autonomous vehicles to exploit track limits systematically.
Ann-Kathrin Schwehn, Alexander Langmann, Mattia Piccinini +1
Professorship of Autonomous Vehicle Systems, TUM School of Engineering and Design, Technical University of Munich, 85748 Garching, Germany; Munich Institute of Robotics and Machine Intelligence (MIRMI)
Trajectory optimization is a central component of fast and efficient autonomous racing. However practical optimization pipelines remain highly sensitive to initialization and may converge slowly or to suboptimal local solutions when seeded with heuristic trajectories such as the centerline or minimum-curvature paths. To address this limitation, we leverage expert driving behavior as a initialization prior and propose a learning-informed initialization strategy based on real-world Formula1 telemetry. To this end, we first construct a multi-track Formula1 trajectory dataset by reconstructing and aligning noisy GPS telemetry to a standardized reference-line representation across 17 tracks. Building on this, we present a neural network that predicts an expert-like raceline offset directly from local track geometry, without explicitly modeling vehicle dynamics or forces. The predicted raceline is then used as an informed seed for a minimum-time optimal control solver. Experiments on all 17 tracks demonstrate that the learned initialization accelerates solver convergence and significantly reduces runtime compared to traditional geometric baselines, while preserving the final optimized lap time.
Samir Shehadeh, Lukas Kutsch, Nils Dengler +2
University of Bonn, Germany · Lamarr Institute for Machine Learning and Artificial Intelligence and the Center for Robotics, Bonn, Germany
Autonomous racing requires accurate trajectory tracking near the handling limits of a vehicle while maintaining low computational latency. Geometric controllers are computationally efficient, but the Ackermann steering geometry becomes invalid under limit-handling conditions. Model- and Acceleration-based Pursuit (MAP) preserves the simplicity of geometric approaches while leveraging tire dynamics. Yet MAP remains fundamentally a geometric controller that relies on Ackermann steering geometry, and its tire model may not fully capture the vehicle's actual dynamic response. This paper presents MAP2, a model-based pursuit controller that combines a curvature-based kinematic MPC and a Sparse Gaussian Process (SGP) residual correction. The proposed algorithm uses MPC to optimize kinematic control inputs over a prediction horizon and maps them to steering commands through a tire dynamics model augmented with SGP residual correction. Real-world vehicle experiments demonstrate substantial reductions in lateral tracking error and lap time. Compared with MAP and Pure Pursuit (PP), MAP2 reduces the average lateral tracking error by 37.99% and 44.65%, respectively, while reducing average lap time by at least 1.5%.
Weizhan Huang, Cheng Hu, Edoardo Ghignone +5
Zhejiang University · Center for Project-Based Learning, D-ITET, ETH Zurich