Abstract:Input delays affect systems such as teleoperation and wirelessly autonomous connected vehicles, and may lead to safety violations. One promising way to ensure safety in the presence of delay is to employ control barrier functions (CBFs), and extensions thereof that account for uncertainty: delay adaptive CBFs (DaCBFs). This paper proposes an online adaptive safety control framework for reducing the conservatism of DaCBFs. The main idea is to reduce the maximum delay estimation error bound so that the state prediction error bound is monotonically non-increasing. To this end, we first leverage the estimation error bound of a disturbance observer to bound the state prediction error. Second, we design two nonlinear programs to update the maximum delay estimation error bound satisfying the prediction error bound, and subsequently update the maximum state prediction error bound used in DaCBFs. The proposed method ensures the maximum state prediction error bound is monotonically non-increasing, yielding less conservatism in DaCBFs. We verify the proposed method in an automated connected truck application, showing that the proposed method reduces the conservatism of DaCBFs.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is: how to reduce conservatism in safety - critical control systems in the presence of input delays. Specifically, the author focuses on the safety violation problems that input delays may cause in systems such as tele - operation and wireless autonomous connected vehicles. To solve this problem, the paper proposes an online adaptive safety control framework, aiming to reduce the conservatism of Delay - adaptive Control Barrier Functions (DaCBFs) by improving them. ### Main Problem Description 1. **The Impact of Input Delays on the System** - Input delays affect the safety of systems such as tele - operation systems and wireless autonomous connected vehicles, and may lead to safety violations. 2. **Limitations of Existing Methods** - Existing methods such as Control Barrier Functions (CBFs) and their extensions (such as DaCBFs considering uncertainty), although they can ensure safety to a certain extent, these methods are often too conservative and limit the performance of the system. 3. **Specific Problems to be Solved** - By reducing the maximum delay estimation error bound, make the state prediction error bound monotonically non - increasing, thereby reducing the conservatism of DaCBFs. ### Core Contributions of the Paper - Propose an online adaptive algorithm to make the maximum delay estimation error bound monotonically non - increasing, and then reduce the conservatism of the state prediction error bound. - Design two nonlinear programming procedures to update the maximum delay estimation error bound, and finally update the maximum state prediction error bound in DaCBFs. - Verify the proposed method in the application of autonomous connected trucks, proving that this method can effectively reduce the conservatism of DaCBFs. ### Formula Summary - **State Prediction Error Bound** \[ \|e(t)\| \leq \epsilon_{\text{max}} \int_{t_1}^{t_2} e^{a(t - \tau)} \|g(y(\tau))\| d\tau \triangleq e_{\text{max}}(t) \] where \( a = L_f+L_g (u_{\text{max}}+\epsilon_{\text{max}}) \) - **Maximum Delay Estimation Error Bound** \[ \tilde{D}_{\text{max}}^{t_j}=D^j_t - D^j_t \] - **Error Bound in Safety Conditions** \[ d_e(t)=(L_f h + L_\alpha\circ h) e_{p,\text{max}}+L_g h e_{p,\text{max}} \|u(t)\| \] where \( e_{p,\text{max}} = e_{\text{max}}(t+\hat{D}+\tilde{D}_{\text{max}}^{t_j})+\Delta y_{\text{max}} \) Through the above methods, the paper successfully reduces the conservatism of DaCBFs in systems with unknown input delays, and improves the safety and performance of the system.

Minimizing Conservatism in Safety-Critical Control for Input-Delayed Systems via Adaptive Delay Estimation

Adaptive Control Barrier Functions with Vanishing Conservativeness Under Persistency of Excitation

Constrained control of multi‐input systems with distinct input delays

Control Barrier Functions for Linear Continuous-Time Input-Delay Systems with Limited-Horizon Previewable Disturbances

Delay-aware Robust Control for Safe Autonomous Driving and Racing

Learning to Refine Input Constrained Control Barrier Functions via Uncertainty-Aware Online Parameter Adaptation

Consolidated Control Barrier Functions: Synthesis and Online Verification via Adaptation under Input Constraints

Learning-Based Control Barrier Function with Provably Safe Guarantees: Reducing Conservatism with Heading-Aware Safety Margin

Robust Control Barrier Functions with Uncertainty Estimation

Control Barrier Functionals: Safety-critical Control for Time Delay Systems

Adaptive Deep Neural Network-Based Control Barrier Functions

Safe and Stable Adaptive Control for a Class of Dynamic Systems

Robust Safety for Mixed-Autonomy Traffic with Delays and Disturbances

Advances in the Theory of Control Barrier Functions: Addressing practical challenges in safe control synthesis for autonomous and robotic systems

Adaptive input delays compensation under event-based control signal implementation

ESO-Based Safety-Critical Control for Robotic Systems With Unmeasured Velocity and Input Delay

Control Safety Function for Explicit Safety-Critical Control of Autonomous Vehicles

Safe robust adaptive control under both parametric and nonparametric uncertainty

Safe Adaptive Control for Uncertain Systems with Complex Input Constraints

Learning a Better Control Barrier Function Under Uncertain Dynamics