Contract-based hierarchical control using predictive feasibility value functions
This work provides an incremental improvement for control system designers dealing with hierarchical control, particularly in applications like autonomous driving, by enhancing modularity and safety assessment.
This paper addresses the challenge of provably safe and modular design in hierarchical model predictive control by introducing a contract-based strategy. It allows a higher-level controller to assess the feasibility of reference trajectories without detailed knowledge of the lower-level controller's model, constraints, or cost, using optimal slack variables as a contract. The method is tested in an autonomous driving scenario involving a planner and a motion controller.
Today's control systems are often characterized by modularity and safety requirements to handle complexity, resulting in the use of hierarchical control structures. Although hierarchical model predictive control offers favorable properties, achieving a provably safe, yet modular design remains a challenge. This paper introduces a contract-based hierarchical control strategy to improve the performance of control systems facing challenges related to model inconsistency and independent controller design across hierarchies. We consider a setup where a higher-level controller generates references that affect the constraints of a lower-level controller, which is based on a soft-constrained MPC formulation. The optimal slack variables of the lower-level MPC serve as the basis for a contract that allows the higher-level controller to assess the feasibility of the reference trajectory without exact knowledge of the model, constraints, and cost of the lower-level controller. To ensure computational efficiency while maintaining model confidentiality, we propose using an explicit function approximation, such as a neural network, to represent the cost of optimal slack values. The approach is tested for a hierarchical control setup consisting of a planner and a motion controller as commonly found in autonomous driving.