Safe Learning Reference Governor: Theory and Application to Fuel Truck Rollover Avoidance
This work addresses safety-critical constraint enforcement for systems like fuel trucks where accurate models are lacking, though it appears incremental as it builds on existing reference governor methods by adding learning capabilities.
The paper tackled the problem of enforcing state and control constraints in systems with unavailable accurate models by proposing a learning reference governor (LRG) approach, which improved command tracking performance while ensuring safety during and after learning, and demonstrated its effectiveness in fuel truck rollover avoidance through simulations under various conditions.
This paper proposes a learning reference governor (LRG) approach to enforce state and control constraints in systems for which an accurate model is unavailable, and this approach enables the reference governor to gradually improve command tracking performance through learning while enforcing the constraints during learning and after learning is completed. The learning can be performed either on a black-box type model of the system or directly on the hardware. After introducing the LRG algorithm and outlining its theoretical properties, this paper investigates LRG application to fuel truck (tank truck) rollover avoidance. Through simulations based on a fuel truck model that accounts for liquid fuel sloshing effects, we show that the proposed LRG can effectively protect fuel trucks from rollover accidents under various operating conditions.