TL;DR
This paper presents three novel control strategies for spacecraft attitude tracking that effectively combine the high-authority, rapid response of thrusters with the fine-tuning capabilities of momentum wheels. These controllers are rigorously proven to achieve globally asymptotically stable tracking, offering robust solutions for complex spacecraft maneuvers.
What Problem Does This Paper Address?
Controlling a spacecraft's orientation (attitude) is crucial for many missions, from pointing telescopes to precise orbital maneuvers. Traditional methods often rely on either thrusters for quick, large attitude changes or momentum wheels for gradual, precise adjustments. However, achieving both rapid maneuvers and high-precision pointing simultaneously is challenging. Thrusters, while powerful, often lack the fine control needed for delicate adjustments and consume propellant. Momentum wheels offer precision but have limited torque authority, making them less effective for large, fast reorientations. This paper addresses the problem of developing integrated control laws that leverage the strengths of both actuator types to achieve robust and precise attitude tracking for large-angle maneuvers.
Methods
The authors develop three distinct control laws based on the nonlinear equations of motion for a rigid spacecraft equipped with both thrusters and momentum wheels. They utilize Modified Rodrigues Parameters (MRPs) to describe the spacecraft's attitude kinematics, a common and efficient representation in spacecraft dynamics. The core of their approach involves designing feedback control laws that minimize tracking errors in both angular velocity and attitude.
- Controller I: Employs a "bang-bang" control strategy for the thrusters, providing the primary, rapid maneuver torque. Momentum wheels are then used for fine-tuning and correcting any tracking errors.
- Controller II: Similar to Controller I, but designed to minimize momentum wheel usage. The wheels are primarily engaged only to correct initial condition errors, with thrusters handling the bulk of the control effort.
- Controller III: Utilizes a linear feedback control law for the momentum wheels and a nonlinear feedback for the thrusters, offering a different balance between the two actuator types.
A key element across all controllers is the use of a candidate Lyapunov function to rigorously prove global asymptotic stability of the closed-loop system. This mathematical framework ensures that the tracking errors will converge to zero over time, regardless of the initial conditions.
Datasets
Not discussed in the paper. The paper focuses on theoretical development and simulation-based validation.
Key Insights and Ideas
- Synergistic Actuation: The paper demonstrates that a well-designed combination of thrusters and momentum wheels can outperform systems relying on a single actuator type, offering both maneuverability and precision.
- Robustness through Stability Proofs: The rigorous mathematical proofs of global asymptotic stability provide a strong theoretical foundation for the proposed controllers, assuring their reliability in real-world applications.
- Flexible Control Architectures: The three distinct controllers offer flexibility in how thruster and momentum wheel resources are allocated, allowing for adaptation to specific mission requirements and actuator capabilities (e.g., on-off thrusters vs. continuously variable thrust).
- MRPs for Kinematics: The use of Modified Rodrigues Parameters (MRPs) is highlighted as an effective method for handling spacecraft attitude kinematics, particularly in the context of large-angle maneuvers, while also acknowledging potential singularity issues that can be managed.
Key Results
The paper presents numerical simulations that illustrate the effectiveness of the three proposed controllers.
- Error Convergence: Figure 1 shows that for all three controllers, the angular velocity errors ($\delta\omega$) and attitude errors ($\delta\sigma$) are driven asymptotically to zero, confirming the stability proofs.
- Controller Performance:
- Controller I: Thrusters provide ideal time-optimal bang-bang control, while momentum wheels deliver piecewise smooth torques to eliminate tracking errors.
- Controller II & III: These controllers show similar performance to Controller I, with thruster torque profiles being close to, but not identical to, the ideal bang-bang control due to attitude error feedback. The momentum wheel controls for Controllers II and III are smooth, unlike Controller I, but require variable thrust levels from the thrusters.
- Quantitative Gains: Specific gains ($k_1 = 54$, $k_2 = 47$) were selected to achieve comparable performance across all three controllers, demonstrating their effectiveness in a simulated minimum-time rest-to-rest maneuver.
Limitations and Future Work
- MRPs Singularities: The paper acknowledges that MRPs have singularities (e.g., at a 360° eigenaxis rotation). While the attitude error singularity doesn't pose a problem for stability, large-angle tracking might require using dual or shadow sets of MRPs, or alternative non-singular representations like Euler parameters for attitude description.
- Real-time Reference Profile Generation: The paper mentions that the nominal bang-bang control profiles were computed offline. In practice, a real-time approach for generating these reference thrust profiles would be necessary.
- Actuator Constraints: While the controllers are designed for thrusters and momentum wheels, the paper doesn't delve deeply into specific actuator saturation or rate limits, which would be critical in a real implementation.
- Future Work: The authors suggest that the control laws can be adapted for different attitude representations and that further investigation into real-time generation of reference trajectories and handling of actuator constraints would be beneficial.
Summary updated