Robust Path Tracking for Vehicles via Continuous-Time Residual Learning: An ICODE-MPPI Approach
Shugen Song, Wenjie Mei, Chengyan Zhao
TLDR
ICODE-MPPI uses continuous-time residual learning with Neural ODEs to significantly improve robust path tracking for vehicles.
Key contributions
- Introduces ICODE-MPPI, a robust framework for vehicle path tracking.
- Leverages Input Concomitant Neural ODEs (ICODEs) to learn unmodeled residual dynamics.
- ICODEs ensure physical consistency and temporal continuity during the MPPI prediction horizon.
- Achieves up to 69% reduction in cross-tracking error and suppresses control chattering.
Why it matters
MPPI control is powerful but struggles with unmodeled dynamics. ICODE-MPPI addresses this by learning residuals, significantly improving robust path tracking and control smoothness for autonomous vehicles. This is crucial for reliable real-world performance in complex environments.
Original Abstract
Model Predictive Path Integral (MPPI) control is a powerful sampling-based strategy for nonlinear autonomous systems. However, its performance is often bottlenecked by the fidelity of nominal dynamics. We propose ICODE-MPPI, a robust framework that leverages Input Concomitant Neural Ordinary Differential Equations (ICODEs) to learn and compensate for unmodeled residual dynamics. Unlike discrete-time learners, ICODEs maintain physical consistency and temporal continuity during the MPPI prediction horizon. High-fidelity simulations on complex trajectories demonstrate that ICODE-MPPI achieves up to a 69\% reduction in cross-tracking error under persistent disturbances compared to standard MPPI control. Furthermore, our analysis confirms that ICODE-MPPI significantly suppresses control chattering, yielding smoother steering commands and superior robust performance.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.