Energy-Based Adaptive Cruise Control With Structurally Induced Self-Excited Driving for Car-Following

This paper proposes a novel energy-based adaptive cruise control (ACC) algorithm for car-following that fundamentally redefines the control target. Conventional ACC, formulated as a spacing-error regulation problem, structurally leads to frequent actuator switching under realistic driving conditions, which can increase control effort and limit opportunities for fuel-efficient behavior. To address these limitations, the proposed controller regulates a desired energy level within the car-following system instead of minimizing tracking errors with respect to the preceding vehicle. This change induces structurallyself-excited driving, forming a loop in the relative distance–velocity domain and enabling Pulse-and-Glide-like behavior without relying on rule-based switching or gain tuning. To ensure safety duringself-excited driving, a tube-based control structure is introduced, within which the system remains bounded. Additionally, a parameter update mechanism is incorporated to further reduce the influence of the preceding vehicle. Together, these features enable smoother control inputs, improved road load utilization, and reduced control effort from the lower-level controller. Simulation results demonstrate that the proposed algorithm outperforms conventional ACC and production-level Smart Cruise Control (SCC) systems in terms of control smoothness and energy efficiency, while maintaining a safe following distance. The algorithm also satisfies string stability, validating its applicability in multi-vehicle scenarios. Overall, the proposed approach offers a structurally interpretable and driver-personalizable framework for efficient and robust car-following.