White-Box Neural Ensemble for Vehicular Plasticity: Quantifying the Efficiency Cost of Symbolic Auditability in Adaptive NMPC
This addresses the need for auditability in adaptive control systems for vehicles, though it is incremental in quantifying the efficiency trade-off.
The paper tackled the problem of vehicular plasticity by developing a white-box adaptive NMPC architecture that uses a neural ensemble to adapt to varying operating regimes without retraining, achieving rapid adaptation (~7.3 ms) and near-ideal tracking fidelity where baselines fail. It quantified the transparency cost, showing that symbolic graph maintenance increases solver latency by 72-102X compared to compiled models.
We present a white-box adaptive NMPC architecture that resolves vehicular plasticity (adaptation to varying operating regimes without retraining) by arbitrating among frozen, regime-specific neural specialists using a Modular Sovereignty paradigm. The ensemble dynamics are maintained as a fully traversable symbolic graph in CasADi, enabling maximal runtime auditability. Synchronous simulation validates rapid adaptation (~7.3 ms) and near-ideal tracking fidelity under compound regime shifts (friction, mass, drag) where non-adaptive baselines fail. Empirical benchmarking quantifies the transparency cost: symbolic graph maintenance increases solver latency by 72-102X versus compiled parametric physics models, establishing the efficiency price of strict white-box implementation.