Abstract
The prescribed-time stabilization problem for a general class of nonlinear systems with unknown input gain and appended dynamics (with unmeasured state) is addressed. Unlike the asymptotic stabilization problem, the prescribed-time stabilization objective requires convergence of the state to the origin in a finite time that can be arbitrarily picked (i.e., prescribed) by the control system designer irrespective of the initial condition of the system. The class of systems considered is allowed to have general nonlinear uncertain terms throughout the system dynamics as well as uncertain appended dynamics (that effectively generate a time-varying non-vanishing disturbance signal input into the nominal system). The control design is based on a time scale transformation, dynamic high-gain scaling, and adaptation dynamics with temporal forcing terms.
Introduction
In classical control theory, the standard objective is asymptotic stability, where system states approach zero as time approaches infinity. While finite-time and fixed-time controllers have been developed to ensure faster convergence, their settling times often depend on initial conditions or specific system parameters.
This research addresses the more stringent requirement of prescribed-time stabilization. In this framework, the designer selects a specific deadline $T$ (e.g., 0.2 seconds), and the controller guarantees that the system reaches the origin exactly by that time, regardless of the starting state. This paper extends this capability to a highly complex class of nonlinear systems characterized by unknown input gains, unmeasured hidden dynamics, and non-vanishing disturbances.
Problem Formulation
The challenge lies in stabilizing a system where the relationship between the control input and the system response is not fully known. Specifically, the model accounts for:
- Unknown Input Gain: The controller lacks precise knowledge of the magnitude of the "push" provided by its input signal.
- Appended Dynamics: The system contains hidden, unmeasured internal states ($z$) that influence the primary dynamics.
- Non-Vanishing Disturbances: Unlike many idealized models where noise disappears as the system reaches equilibrium, this framework accounts for persistent disturbances that must be actively rejected.
Methodology
The authors propose a multi-layered control strategy to enforce convergence within the prescribed deadline $T$:
1. Time-Scale Transformation
The core of the approach involves mapping the finite time interval $[0, T)$ to an infinite horizon $[0, \infty)$ using a nonlinear temporal transformation. This mathematical shift allows the researchers to apply robust tools from asymptotic stability theory to solve what is inherently a finite-time problem.
2. Dynamic High-Gain Scaling
A dynamic scaling parameter ($r$) is employed to overpower system uncertainties. Unlike static gains, this parameter is adaptive; it grows automatically based on the system's real-time requirements to ensure that tracking errors are suppressed as the deadline approaches.
3. Adaptation with Temporal Forcing
Because the bounds on uncertainties and input gains are unknown, the controller uses adaptation laws to estimate parameters ($\hat{\theta}$ and $\hat{\theta}_1$) on the fly. To meet the strict deadline, "temporal forcing terms" are integrated into these laws, ensuring the estimates evolve rapidly enough to compensate for uncertainties before $t = T$.
4. Robust Non-Smooth Control
The control law $u$ is bifurcated into two components: a nominal scaling part and a robust, non-smooth component utilizing a signum function. This latter part is specifically designed to cancel out the non-vanishing disturbances generated by the appended dynamics.
Key Insights and Theoretical Results
A primary concern in prescribed-time control is the potential for the control signal ($u$) to "blow up" or reach infinity as the deadline approaches. The authors provide a rigorous Lyapunov-based proof (Theorem 1) demonstrating that while internal scaling parameters grow toward infinity as $t \to T$, the actual control input $u(t)$ remains uniformly bounded and safe for physical actuators.
Furthermore, this framework relaxes several common assumptions in the field. Previous literature often required prior knowledge of the lower bound of the input gain or the upper bound of disturbances. This design removes those requirements, making the controller applicable to a wider range of uncertain environments.
Simulation and Validation
The theoretical framework was validated using a numerical simulation of a fifth-order nonlinear system. The test system included complex trigonometric uncertainties, exponential terms, and hidden linear dynamics. In a scenario with a prescribed deadline of $T = 0.2$ seconds, the system states successfully converged to the origin by the deadline. The simulation confirmed that even with significant parametric uncertainty, the adaptation laws correctly compensated for the lack of prior knowledge without exceeding control limits.
Practical Considerations and Future Work
While theoretically sound, the authors note that numerical sensitivity is a practical hurdle. As the system clock approaches $T$, the effective gains become very large, which can lead to numerical chattering or overflow in digital implementations. To mitigate this, the researchers suggest:
- Implementing a "dead zone" or small tolerance around the origin.
- Saturating the control gains at a high but safe threshold.
- Setting the controller's internal deadline to a slightly later time than the physical requirement.
Future research will focus on extending this framework to non-triangular or feedforward system structures and addressing cases where the sign of the control gain is entirely unknown.
Summary updated