Malak Samaan

Hybrid Navigation System for Spaceplanes, Launch and Re-Entry Vehicles

In 2010 the German Aerospace Center (DLR) will conduct the second experiment of the DLR hypersonic SHarp Edge Flight EXperiment Program SHEFEX-2. The purpose is to investigate possible new shapes for future launcher or re-entry vehicles with faceted surfaces and sharp edges and to demonstrate key technologies for re-entry like hypersonic ight control using steerable canard ns. Accurate control of the vehicle using the canards requires a highly accurate knowledge of the angle of attack and the side slip angle. Both angles can only be derived from the ight path and an attitude measurement. The rst can be achieved using GPS measurements. The second can be provided only by the most accurate Inertial Navigation Systems (INS) because drifts due to launch vibrations exceed the accuracy requirements. Therefore, a star tracker will be used to update the attitude information shortly before entry. The SHEFEX-2 mission describes an entry scenario which is applicable to other entry missions. There is a general need to develop a high accuracy integrated navigation system which can be used for multiple missions. This navigation system should combine the measurements from an inertial measurement unit (IMU), GPS receiver and star tracker with the option to include additional sensors. This paper will describe the concept of the integrated navigation system with a focus on integrating the star tracker into the SHEFEX-2 experiment.

Attitude Determination for the SHEFEX 2 Mission Using a Low Cost Star Tracker

After the successfully launched hypersonic SHarp Edge Flight EXperiment (SHEFEX) in 2005, DLR has scheduled a second re-entry experiment SHEFEX 2 for 2010. The novel flight control system based on steerable canard fins requires a precise knowledge of the inertial spacecraft attitude. For this purpose, a star tracker is to be conceived and integrated in an inertial navigation system, updating the attitude information before re-entry. The developed star tracker is considered to be a low cost and low accuracy sensor that is suitable for the proposed mission and the attitude accuracy requirements. This sensor is based on an off-the-shelf camera and a PC104 computer communicating with the navigation computer. The attitude determination software consists of a processing chain including the camera control, the image processing, the star identification and the attitude estimation. The emphasis of the work is on developing a simple but robust system. Therefore, different algorithm concepts for each element of the chain have been implemented, tested and compared. Furthermore, star image simulation software has been developed, allowing performance and speed tests of the single blocks and the processing chain as a unit. The paper presents the star tracker system, details some newly developed parts and discusses the test results.

SHEFEX-3 Optimal Feedback Entry Guidance

SHEFEX is a DLR-led series of missions for scienti c experiments and reentry technology development. SHEFEX-2 was successfully launched from Norway (Andoya Rocket Range) in June 2012. To go on with the e ort to increase the technological level for real space missions, a new challenge in the next years with the development of SHEFEX-3 arises. SHEFEX-3, foreseen to be launched in 2016, will be more complex than SHEFEX-2 in virtue of the presence of a real guided re-entry phase, while for SHEFEX-2 an autonomous Guidance and Control phase was only partially foreseen. As a consequence, the mission will be ambitious, especially in the development of the GNC subsystem. DLR GNC Systems Department will be responsible for the development of Guidance and Navigation modules, while Control will be developed by Airbus Defense and Space, in cooperation with DLR. In this work the development of the nominal entry guidance, based on the use of PseudoSpectral Methods, is discussed. This feedforward control is then coupled with a Gain-Scheduled LQR tracking controller to reduce the error on the terminal points of the mission. Results show that the proposed approach meets the requirements on the physical constraints and the terminal states, satisfying at the same time the strong limitations coming from the need to have a highly-constrained angle of attack pro le.

Flight Results from the SHEFEX2 Hybrid Navigation System Experiment

The Hybrid Navigation System was flown on the second SHarp Edge Flight EXperiment sounding rocket mission on June 22, 2012 from Andya Rocket Range in Norway by the German Aerospace Center (Deutsches-Zentrum f�ur Luft- und Raumfahrt, DLR). The on-board navigation algorithm fuses measurements from an IMU, GPS receiver and star tracker with a delayed extended Kalman filter to estimate a navigation solution over time. The in-flight navigation performance is calculated by comparing the navigation solution returned via telemetry to an accurate reconstruction of the trajectory. Trajectory reconstruction combines all available data sent via telemetry, more accurate state transition and measurement models and additional off-line information using the unscented Kalman filter and unscented Rauch-Tung-Striebel backward smoother. The reconstructed trajectory is computed off-line and is much more accurate than the in-flight navigation solution. Comparing the reconstructed trajectory to the telemetry data shows that the system behaved as expected, although it did not meet its performance requirements. The Hybrid Navigation System software is now at TRL 7.

Reconfigurable Hardware-in-the-Loop Test Bench for the SHEFEX2 Hybrid Navigation System Experiment

The Hybrid Navigation System is being developed to achieve the required attitude accu- racy and calculate a navigation solution for the SHEFEX2 mission, which will be launched in February 2012 from Andya Rocket Range in Norway by the German Aerospace Center (Deutsches-Zentrum fur Luft- und Raumfahrt, DLR). This system fuses data from an IMU, GPS receiver, and star tracker using the extended Kalman filter. For testing it needs a re- alistic software- and hardware-in-the-loop test bench that can simulate the expected flight conditions and test the system. The hardware test bench requirements include: real-time simulation progression, accurate synchronization of all components, and providing stimu- lation inputs for the GPS receiver, star tracker and gyro instruments (inside the IMU) representative of the expected flight profile. Another key requirement is that the test bench must be easily reconfigurable by swapping real and modeled instruments. This al- lows the system to be tested with variable levels of complexity, which decreases debugging time by allowing the user to separate or remove the problem points by simply swapping cables. The developed software-in-the-loop simulation consists of the navigation flight code wrapped in a Simulink s-function and fed by several high-fidelity Matlab/Simulink models which simulate the rocket dynamics and system sensors. The hardware-in-the-loop test bench runs the models from the software simulation on a dSPACE real-time simulator, which feeds the modeled sensor outputs to the hybrid navigation systems navigation com- puter. With the help of a Spirent GSS7700 GPS signal generator, a Jenoptik Optical Sky field Simulator, and an Acutronic 3-axis rotation table all synchronized to the dSPACE simulation, the modeled sensors can be replaced with real sensors fed with stimulated in- put. Navigation results from the hardware test bench are compared with results from the software-in-the-loop simulation to ensure that the hardware-in-the-loop simulation is working correctly.