Abstract
In this paper, the attitude control problem of rigid body is addressed with considering inertia uncertainty, bounded time-varying disturbances, angular velocity-free measurement, and unknown non-symmetric saturation input. Using a mathematical transformation, the effects of bounded time-varying disturbances, uncertain inertia, and saturation input are combined as total disturbances. A novel finite-time observer is designed to estimate the unknown angular velocity and the total disturbances. For attitude control, an observer-based sliding-mode control protocol is proposed to force the system state convergence to the desired sliding-mode surface; the finite-time stability is guaranteed via Lyapunov theory analysis. Finally, a numerical simulation is presented to illustrate the effective performance of the proposed sliding-mode control protocol.
Introduction
Maintaining the precise orientation (attitude) of a rigid body, such as a satellite or an unmanned aerial vehicle, is a fundamental challenge in aerospace engineering. While textbook control theory often assumes perfect sensor data and unlimited actuator power, real-world systems face significant hurdles:
- Velocity-Free Measurement: High-precision gyroscopes are often unavailable due to weight, cost, or reliability constraints, necessitating the estimation of angular velocity from orientation data alone.
- Inertia Uncertainty: A vehicle's mass distribution can change dynamically due to fuel consumption or component deployment.
- Actuator Saturation: Physical limits on thrusters or motors can lead to instability if the controller demands more torque than the hardware can provide.
- External Disturbances: Environmental factors like solar radiation pressure or atmospheric drag introduce persistent noise.
This paper proposes a robust control strategy that addresses these non-linearities and uncertainties simultaneously, ensuring high-precision pointing even under adverse conditions.
Methodology
The proposed solution relies on a sophisticated observer-controller architecture designed to handle "lumped" uncertainties.
The "Lumped" Transformation
Rather than modeling inertia errors, external noise, and saturation limits as separate entities, the authors mathematically combine them into a single "total disturbance" term. This simplification allows a unified approach to compensation.
Finite-Time Observer (FTO)
The core innovation is a novel state observer that estimates both the unmeasured angular velocity and the total disturbance. Unlike traditional asymptotic observers that converge as time approaches infinity, this FTO is mathematically proven to converge within a specific, finite duration.
Observer-Based Sliding Mode Control (SMC)
The controller uses the observer's estimates to force the system onto a predefined "sliding surface." To mitigate the "chattering" effect—high-frequency oscillations common in standard SMC—the authors employ hyperbolic tangent functions, resulting in a smooth and continuous control signal.
Numerical Simulation and Results
The researchers validated their approach through numerical simulations of a rigid body with a $3 \times 3$ nominal inertia matrix, subjected to $10\%$ inertia uncertainty, time-varying external torques, and non-symmetric saturation limits.
The results demonstrated exceptional robustness and precision:
- Rapid Convergence: The system's orientation (represented by unit quaternions) converged to the target state $[1, 0, 0, 0]^T$ in approximately 5 seconds, despite the injected noise and uncertainties.
- Estimation Accuracy: The FTO successfully tracked the unmeasured angular velocity and the complex total disturbance with minimal error shortly after initialization.
- Saturation Handling: The controller maintained stability even when the required control effort hit the simulated actuator limits, proving the effectiveness of treating saturation as a disturbance.
Conclusion
This work provides a robust framework for rigid body attitude control that bridges the gap between theoretical models and practical hardware limitations. By integrating a finite-time observer with a chattering-free sliding mode controller, the authors achieve rapid, high-precision orientation without the need for angular velocity sensors. Future research may focus on automating parameter tuning and evaluating the computational efficiency of these algorithms on low-power embedded systems.
Summary updated