The paper studies safety guarantees for systems with time-varying control bounds. It has been shown that optimizing quadratic costs subject to state and control constraints can be reduced to a sequence of Quadratic Programs (QPs) using Control Barrier Functions (CBFs). One of the main challenges in this method is that the CBF-based QP could easily become infeasible under tight control bounds, especially when the control bounds are time-varying.
To address this issue, the authors propose a new type of adaptive CBFs called Auxiliary-Variable Adaptive CBFs (AVCBFs). The key idea is to introduce an auxiliary variable that multiplies each CBF itself, and define dynamics for the auxiliary variable to adapt it in constructing the corresponding CBF constraint. This approach can improve the feasibility of the CBF-based QP while avoiding extensive parameter tuning and non-overshooting control near the boundaries of safe sets.
The authors demonstrate the advantages of using AVCBFs and compare them with existing techniques, such as Penalty-based Adaptive CBFs (PACBFs), on an Adaptive Cruise Control (ACC) problem with time-varying control bounds. The results show that the proposed AVCBF approach can generate smoother and more adaptive control compared to existing methods, without requiring design of excessive additional constraints and complicated parameter-tuning procedures.
In eine andere Sprache
aus dem Quellinhalt
arxiv.org
Wichtige Erkenntnisse aus
by Shuo Liu,Wei... um arxiv.org 04-22-2024
https://arxiv.org/pdf/2304.00372.pdfTiefere Fragen