Eric Bornschlegl

Gain-Scheduled FDI for a Re-entry Vehicle

This paper presents the design of a gain scheduled fault detection and isolation (FDI) filter for the Hopper reusable launch vehicle (RLV). The fault scenario is that of faults in the vehicles rudder actuator and sideslip sensor during a focused 90 second period of the re- entry. Both of the considered faults strongly affect the lateral response of the vehicle, making simultaneous FDI difficult. A dynamically stable model of the Hopper RLV is considered and FDI filter design is performed on linearised models of the vehicle trimmed about the re- entry trajectory. H-infinity theory is employed for the FDI filter synthesis, with a set of LTI FDI filters designed at the trim points and then scheduled to form the gain scheduled FDI filter. The effectiveness of the LTI point-design filters and the gain-scheduled filter are determined by simulation using a tightly gain-scheduled model of the linearised vehicles open-loop response that captures the strongly parameter varying vehicle behaviour as it tracks the re-entry trajectory. The advantages of using gain-scheduled FDI filters for FDI on RLVs are highlighted via the simulations.

Position and Attitude Model-Based Thruster Fault Diagnosis: A Comparison Study

This paper deals with performance and reliability evaluation of a fault diagnosis scheme based on two distinct models to detect and isolate a single thruster fault affecting a chasing spacecraft during rendezvous with a passive target in a circular orbit. The analysis is conducted in the frame of a terminal rendezvous sequence of the Mars Sample Return mission. A complete description of a robust residual generation design approach based on eigenstructure assignment is presented. Unknown time-varying delays, induced by the thruster drive electronics and uncertainties on thruster rise times, are considered as unknown inputs. Particular novelty of the work is a new method for estimating the unknown input directions used to enhance the robustness properties of the diagnosis scheme. Monte Carlo results from a high-fidelity industrial simulator and carefully selected performance and reliability indices allows us to evaluate the effectiveness of both schemes. The obtained results reveal that the proposed fault diagnosis scheme based on a position model is a justified competitor to the conventionally used attitude model-based scheme.

A class of nonlinear unknown input observer for fault diagnosis: Application to fault tolerant control of an autonomous spacecraft

In this paper, the problem of Nonlinear Unknown Input Observer (NUIO) based Fault Detection and Isolation (FDI) scheme design for a class of nonlinear Lipschitz systems is studied. The proposed FDI method is applied to detect, isolate and accommodate thruster faults of an autonomous spacecraft involved in the rendezvous phase of the Mars Sample Return (MSR) mission. Considered fault scenarios represent fully closed thruster and thruster efficiency loss. The FDI scheme consists of a bank of NUIOs with adjustable error dynamics, a robust fault detector that is based on judiciously chosen frame and an isolation logic. The bank of observers is in charge of confining the fault to a subset of possible faults and the isolation logic makes the final decision about the faulty thruster index. Finally, a thruster fault is accommodated by re-allocating the desired forces and torques among the remaining healthy thrusters and closing the associated thruster valve. Monte Carlo results from “high-fidelity” MSR industrial simulator demonstrate that the proposed fault tolerant strategy is able to accommodate thruster faults that may have effect on the final rendezvous criteria.

Robust Fault Diagnosis strategies for Spacecraft Application to LISA Pathfinder experiment

Abstract This paper presents research activities conjointly led by EADS Astrium Satellites and the European Space Agency on innovative and robust health monitoring system for the next generation of spacecraft. Two robust FDI schemes are presented to detect and isolate faults affecting the micro-Newton colloidal thrust system of the LISA Pathfinder spacecraft. The first FDI strategy is based on a bank of eight H ∞ / H − residual generators designed according to the Generalized Observer Strategy whereas the second strategy consists of Kalman-based projected observers. The efficiency of the proposed FDI techniques is assessed through non linear simulations performed under realistic conditions (physical parameter uncertainties, disturbances, measurement noises, measurement delays, thruster jet misalignment,…). The results are quite encouraging, illustrate the effectiveness of the proposed techniques and suggest that the solutions could be practical viable candidates.

Toward an Autonomous Lunar Landing Based on Low-Speed Optic Flow Sensors

For the last few decades, growing interest has returned to the quite challenging task of the autonomous lunar landing. Soft landing of payloads on the lunar surface requires the development of new means of ensuring safe descent with strong final conditions and aerospace-related constraints in terms of mass, cost and computational resources. In this paper, a two-phase approach is presented: first a biomimetic method inspired from the neuronal and sensory system of flying insects is presented as a solution to perform safe lunar landing. In order to design an autopilot relying only on optic flow (OF) and inertial measurements, an estimation method based on a two-sensor setup is introduced: these sensors allow us to accurately estimate the orientation of the velocity vector which is mandatory to control the lander’s pitch in a quasi-optimal way with respect to the fuel consumption. Secondly a new low-speed Visual Motion Sensor (VMS) inspired by insects’ visual systems performing local angular 1-D speed measurements ranging from 1.5°/s to 25°/s and weighing only 2.8 g is presented. It was tested under free-flying outdoor conditions over various fields onboard an 80 kg unmanned helicopter. These preliminary results show that the optic flow measured despite the complex disturbances encountered closely matched the ground-truth optic flow.

Backup state observer based on Optic Flow applied to lunar landing

The observer presented in this paper, which was based on the use of three minimalistic bio-inspired Visual Motion Sensors (VMS) detecting Optic Flow (OF) cues, states was intended as a backup solution in the case of Inertial Measurement Unit (IMU) failure. Contrary to most previous Guidance Navigation and Control (GNC) solutions for planetary landing, which have involved a sensor suite including an IMU, an innovative strategy is presented here for estimating states without any need for inertial measurements, based solely on information about the relative velocity of the images of the surrounding environment. A Linear Parameter Varying (LPV) observer designed on a LPV system linearized around a reference trajectory, estimates: the ventral OF, the expansion OF and the local pitch angle. A previously developed observer was applied here to a larger class of nonlinear systems by making an ingenious change of variable. Simulations performed on a lunar landing scenario yielded satisfactory performance and showed the robustness of the OF based observer to initial uncertainties and measurement noise.

Suboptimal lunar landing GNC using nongimbaled optic-flow sensors

Autonomous planetary landing is a critical phase in every exploratory space mission. Autopilots have to be safe, reliable, energy saving, and as light as possible. The 2-D guidance, navigation, and control strategy presented here makes use of biologically inspired landing processes. Based solely on the relative visual motion known as optic flow (OF), assessed with minimalistic 6-pixel 1-D OF sensors and inertial measurement unit measurements, an optimal reference trajectory in terms of the mass was defined for the approach phase. Linear and nonlinear control laws were then implemented to track the optimal trajectory. To deal with the demanding weight constraints, a new method of OF estimation was applied, based on a nongimbaled OF sensor configuration and a linear least-squares algorithm. The promising results obtained with software-in-the-loop simulations showed that the present full guidance, navigation, and control solution combined with our OF bio-inspired sensors is compatible with soft, fuel-efficient lunar spacecraft landing and might also be used as a backup solution in case of conventional-sensor failure.

Verification & Validation of a Satellite Fault Detection and Isolation Scheme Based on Sliding-Mode Observers

Abstract In this article, the verification and validation (V&V) of a fault detection and isolation scheme based on sliding mode observer residual evaluators and threshold-based residual analysis of gyro and thruster faults for the Mars EXpress (MEX) satellite is presented. The results were part of a European Space Agency project with the goal of examining the potential applicability of modern model-based FDI techniques for on-board satellite deployment. The V&V campaign, consisting of firstly a set of specified fault simulations and secondly a Monte Carlo campaign, has been performed using an industrial-level functional engineering simulator developed around a high-fidelity model of the MEX satellite operating during the Sun Acquisition Mode (SAM), which includes up to 6 different controller mode changes. The results show the good performance and robustness of the FDI scheme throughout the SAM phase.

Guidance and Control Design for the Ascent Phase of the Hopper RLV

In this article the design of a guidance and control system for the automated ascent of the Hopper reusable launch vehicle is presented. The considered ascent starts at the pull-up maneuver performed immediately after horizontal take off and ends near main-engine-cutoff. The guidance (trajectory control) law is based on the coupled inversion of flight-path-toangle-of-attack and heading-to-bank-angle dynamics. The control (attitude) law uses also nonlinear dynamic inversion to obtain the required aerodynamic surfaces and engine gimbal deflections for robust tracking of the attitude angles from the guidance law. The used NDI attitude law is inherited from a previous design for the Hopper re-entry phase (with only minor modifications apart from the inclusion of thrust vector control) showcasing the reusability of NDI designs for quite different types of configurations. The resulting design has been validated using a Monte Carlo campaign with realistic aerodynamic mismatch, corrupted measurements, parametric uncertainty and high fidelity atmospheric and 6DoF vehicle dynamics models.