Friction uncertainty is a major obstacle for autonomous vehicle control, because it can make planned trajectories infeasible and degrade tracking performance. The paper proposes a two-layer framework that separates planning from control to address this challenge.

The upper layer solves a constrained optimal control problem to produce a time-optimal trajectory along the desired path while respecting all constraints. The lower layer uses indirect adaptive robust control (IARC) to estimate friction-related parameters, combining model compensation with robust feedback to maintain both transient and steady-state tracking performance.

To prevent poor learning, an estimation monitor detects model mismatch, and a smooth parameter reset reduces the estimated tire capacity, limiting abrupt tightening of constraints. The updated parameters are fed back into the planner's model and constraints, enabling trajectory replanning as friction conditions change.

Simulation results indicate that the method achieves accurate tracking and online friction adaptation under large uncertainty while preserving time-optimal motion. The authors have made simulation demos available alongside the paper.