Spatially-Aware Adaptive Trajectory Optimization with Controller-Guided Feedback for Autonomous Racing
Alexander Wachter, Alexander Willert, Marc-Philip Ecker, Christian Hartl-Nesic
AI summary
Problem
Conventional autonomous racing methods decouple offline trajectory planning from online control, treating tracking errors as local deficiencies rather than spatial signals, which limits global trajectory adaptation to repetitive track characteristics.
Approach
The method uses a Kalman-inspired spatial update to convert tracking errors into an adaptive acceleration constraint map, which guides a global NURBS-based trajectory optimizer to iteratively refine the raceline.
Key results
- 17.38% lap time reduction in simulation versus static acceleration baselines
- 7.60% lap time improvement on real hardware across varying tire compounds
- Implicit learning of spatial friction variations without explicit environmental sensing
- Open-source implementation with integrated GUI for reproducible research
Why it matters
Enables autonomous racing platforms to continuously adapt to spatially varying grip conditions and improve lap times without heavy online computation or explicit friction modeling.
Abstract
We present a closed-loop framework for au- tonomous raceline optimization that combines NURBS-based trajectory representation, CMA-ES global trajectory optimiza- tion, 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 itera- tively refines trajectories toward near-optimal performance un- der 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 low to high friction, we obtain a 7.60 % lap time improvement without explicitly parametrizing friction. This demonstrates robustness to changing grip conditions in real- world scenarios.