The Software-in-the-Loop (SiL) test offers a practicable approach for functional testingof embedded software. In this case, the real environment (e.g., space and satellite hard- ware) is simulated on one or several usual computers and coupled with the software un- der test.
Although at the time the controller software had already been tested in the daylight laboratory, some bugs mainly relevant to thermal comfort had gone unnoticed dur- ing the testing period and became more prominent during the hot summer months leading to issues related with the overheating prevention. These issues were identi- fied when the controller failed to provide comfortable conditions and the monitored system behaviour did not agree with the function principles of the controller. Concerning the tuning of the controller variables, the time allocated between the controller actions became an issue during the testing period. During the first weeks of the installation the user stated that in the early morning of sunny days the 20- minute time-step originally set between the controller’s actions was very long. That would prevent the controller from sufficiently blocking the direct radiation between the actions and as a result uncomfortable visual conditions occurred. A shorter pe- riod of 10 minutes was tested for two more weeks without completely solving the problem. After communication with the user, the controller reaction time was set to 5 minutes for a testing period of three weeks. At the end of this period, the user reported that although in the early morning the glare protection was indeed im- proved, the disturbance due to such frequent movements during the rest of the day was substantial. For lack of a better solution in the current time, a compromise was reached when the time step was set to 8 minutes.
is developed in cooperation with the German Space Operation Center (DLR/GSOC). It also supplies the standard real-time position and velocity needed by the attitudecontrol system for the nadir and target pointing mode. The GENIUS GPS system consists of three independent GPS receiver boards, each connected to a separate antenna and low noise amplifier (LNA) as shown in Fig. 2. The GPS Box is connected to the on-board computer (OBC), the power control and distribution unit (PCDU) and the ultra stable oscillator (USO). The used Phoenix boards are commercial 12-channel GPS L1 receivers with a DLR/GSOC developed firmware for space and high dynamics applications . Three GPS antennas are mounted on the middle solar panel in an L-shaped arrangement, creating two baselines with a length of 440 mm and 610 mm respectively (Fig. 1). The antennas are pointing in the opposite direction of the payload cameras and therefore have optimum visibility of the GPS constellation during Earth observation. The three GPS receivers are integrated in a single 100 × 80 × 67 mm box together with an interface board for RS-422 conversion, the USO signal distributor and a latch-up protection for each receiver (Fig. 3). To achieve a high level of redundancy, each receiver can be switched on/off independently varying the system input power from 0.9 W for 1 receiver to 2.6 W for all 3 receivers according to measurements at the testing model. Among other hardware modifications to prepare the receiver boards for space usage, the oscillators were removed to synchronize all receivers to the central 10 MHz USO. This external oven controlled crystal oscillator with a high accuracy of 10 −13 over a period of 1000 s drives the clocks of the receivers and also eliminates variations between individual receiver clocks.
In addition to the previously analyzed models, the results of a model called craigb18 with nine static interface constraint modes and nine fixed interface normal modes are presented. Here, the deviation at the beginning of the simulation is maximal and is also decreasing with the progressing excitation, like for the craigb36 model discussed before. Of course, the model is not as accurate as the others. However, the deviations are for the most cases acceptable, and the model size is quite small, so its computational cost is quite cheap. Furthermore, the results of a model named modalfix36 are depicted in Figure 4.31. It is based on Guyan’s Static Condensation according to Secion 2.3, where the first 36 fixed interface normal modes of the mirror are used for the projection basis. In comparison with the sophisticated methods of Krylov and Craig-Bampton, the deviation is significantly higher. It is proportional to the magnitude of the RMS, but maybe in some cases also acceptable. Nevertheless, the Guyan reduced model only behaves appropriately, if the interface nodes in the EMBS are suitably fixed to each other during the excitation. In contrast, the other methods are valid for more general excitations at the interfaces. Figure 4.32(a) depicts the transient results of the RMS of the WFA at an excitation of 300 Hz for various MOR methods. In particular, the krylov36 model shows for some time instants large deviations which results in a relative error of up to 18 %. The related exposed images are illustrated in Figure 4.32(b), which are the results of an exposure simulation by considering the irradiance over the time, see also Section 3.4.4. On the upper side, the perfect static case for the parabolic mirror is shown and on the lower side, the case of the dynamical excitation with 300 Hz is shown for a scale factor of 1000. Thereby, the models krylov108 and krylov36 are represented. The LOS is visualized with the line in the center of the image. The discrepancy between the two images is apparent, although the WFA differs only in short time ranges.
of a structure into the equations of motion is explained and implemented for a particular case. This method is based in the use of Lagrange’s equations combined with the approach known as assumed modes method, which is based in decomposing the flexible motion into a series of flexible modes. The implementation of this methodology over a particular structure gives a step-by-step example on how this approach can be used to ob- tain a meaningful model of a flexible structure. This modeling approach leads to a definition of the equations of motion of the object containing both the flexible and the rigid motion of the structure. These equations, in combination with the simulation environment, give an insight in the behavior of LFTSs in space. This way, for example, the duty cycle of the magnetorquer and the interaction with the solar pressure before and after going through an eclipse are identified as sources of vibrations for the satellite. Even though the process to derive these equations is mathematically and conceptually complex, the final plant obtained is relatively simple and of low order. This allows to use the model to derive attitude controllers targeting not only the rigid motion of the spacecraft but also the effect of the flexibility. The most relevant challenge or limitation of this modeling approach is the availability of a suitable combination of shape functions and generalized coordinates. This combination should be able to accurately describe the continuous flexible structure using a discrete number of coordinates, related to flexible modes. In the structure studied, the only flexible elements are booms, which lowers the difficulty of finding this combination but also lowers the final accuracy. In is also to be taken into account that the inherent mathematical complexity of the derivations and assumptions, require that the final equations of motion obtained are verified.
In the last step of the quality control algorithm, time differences of pseudorange and carrier-phase double-difference observations (triple-differences) are formed for each individual satellite. The absolute value of each triple-difference is then compared to a test thresh- old, and the observation is rejected if the threshold is exceeded. This editing procedure is possible due to the high data rate of the receivers and the low angular veloc- ity of the CASSIOPE satellite, which only causes a small effect in the triple-difference due to the geometric change in the baseline. In this editing step, pseudorange jumps and carrier-phase cycle-slips are detected. If cycle-slips are detected, the corresponding ambiguities are newly initialized as float values based on code-carrier differences. The next step is the measurement update. The Kalman- filter processes both pseudorange and carrier-phase observations. It can be configured to use single-frequency measurements from any number of signals. In the case of CASSIOPE, it can process only L1 C/A, only L2 P(Y), or both signals together. No combination is formed when processing more than one frequency. Instead, mea- surements from the different frequencies are treated as independent and complementary observations. The ionospheric delay does not need to be estimated since it cancels out in the differencing over the short baselines. Using both L1 C/A and L2 P(Y) is a particularly interest- ing option, since observations on a second frequency with
Based on the presented research and its results, this question was positively answered. In a ﬁrst concept description, the new method was presented as an intermediate step of the procedure proposed by ISO 7870-8. Moreover, basic assumptions were listed and explained. Since the method considers control chart performances expressed by ARLs resulting from grouped processes with non-identically distributed individual val- ues, a calculation formula was developed based on the previously introduced Markov chain approach. Calculation examples were presented and analyzed for Shewhart and CUSUM control charts. For the moving mean and moving range chart, a discrete state space required for the Markov chain approach was derived. Furthermore, an ap- propriate transformation of individual values drawn from Rayleigh distributions was proposed. Findings from the analysis of calculation examples were considered dur- ing the development of the statistical hypothesis test and of related preparing and applying substeps within the procedure which were explained in detail. These sub- steps comprise the creation of predeﬁned and randomly generated process sequences, the calculation of ARL estimators as test statistics as well as the deﬁnition of rele- vant process shifts and of conditions for suﬃcient control chart performances. The fulﬁllment of these conditions is considered as null hypothesis. The determination of critical values is based on a Monte Carlo simulation. Instructions about how to pro- ceed after an acceptance or a rejection of the hypothesis were further presented. The application of the testing method can be performed with the implemented software demonstrator.
The following diploma thesis is related to the AsteroidFinder/SSB project of the German Aerospace Center (DLR). AsteroidFinder/SSB is a compact satellite that will be used for detecting aster- oids with an optical payload (telescope). Especially the objects that are completely inside the Earth orbit are of interest. The task of detecting weak light sources like asteroids places chal- lenging requirements to the AttitudeControl System (ACS). Therefore the AsteroidFinder/SSB needs to be stabilised and controlled by an active there axis ACS. A preliminary design was established during Phase 0/A of the project. Based on these results the following diploma the- sis focus on the development of the attitude determination and control algorithm. Therefore a simulation environment is programmed and two different KALMAN filters are investigated as attitude determination algorithms. These are the extended KALMAN filter and the unscented KALMAN filter. Afterwards a guidance strategy is derived to reach the main mission goals. It is followed by the development of an attitudecontrol strategy which is based on linear quadric GAUSSIAN control. At the end the algorithm functionality is validated through simulation.
The programs of the acceptance and qualification tests are elaborated according to ESA requirements . The heat pipe test programs elaborated by authors for VEGA Project (1984-86), Phobos Project (1986-87), and Mars – 96 have been applied as well. The following main tests have been foreseen: (a) inspection and physical measurements; (b) proof pressure test - leak test; (c) performance testing; (d) burst test; (e) random vibration; (f) storage simulation test; (g) thermal cycles/shock test; (h) aging test (long life test).
mercially available, 12.7 mm diameter CCR mounted into a specifically designed titanium holder that recesses the CCR’s entrance face. In order to select a CCR with a high optical and manufacturing quality, FFDPs of eight retroreflectors were measured and the best retroreflector was selected as a flight module. We believe that the recessed CCR assembly, which is a lightweight and small design optimized for use on a CubeSat, should allow for a coarse verification of the attitudecontrol to within ± 2°.
In 2018 JAXA’s Hayabusa-II spacecraft will arrive at the C-type asteroid 1999 JU3 to perform in-depth analysis of the NEO and collect samples for the return to Earth. One of its payloads is the Mobile Asteroid Surface Scout (MASCOT), developed at the DLR Institute of Space Systems. After being released from Hayabusa-II, MASCOT will autonomously perform in-situ analyses on the asteroid surface. To align its instruments, MASCOT must determine its attitude and reposition itself if necessary. One of the attitude determination concepts makes use of temperature sensors. This study focuses on the development, testing, and simulationof these Orientation Tem- perature Sensors (OTS), based on an existing design with three distinct sensor-types. The new OTS hardware features improved performance, and removes problems expe- rienced with previous iterations. Data gathered during thermal-vacuum tests is used for evaluation, and to validate a newly developed thermal model of these sensors. This model allows simulating sensor behavior under diﬀerent conditions and during the mission, using a number of baseline cases as reference. The collected data is analyzed to see if one of the sensor types – or a combination of them – can be used to determine the orientation of MASCOT on the asteroid surface.
2.3 ADR Scenario Definitions and Assumptions
This paper focuses on studying the attitude motion of the chaser-tether-target system for the critical parts of a tethered ADR mission, which are during and following a deorbit burn applied by an active chaser. A feedback control is developed to stabilize the tether and the target separation distance in a towing ADR mission. This improves safety substantially by reducing the danger of chaser-debris collision and thus fragmentation. The paper assumes that the challenging rendezvous and target capture has been already performed. The space debris is a passive, uncooperative satellite in low Earth orbit. The chaser is equipped with a large main engine (2000 N) for performing the deorbit burn and reaction control system (RCS) for applying force and torque corrections. The main engine is of an on-off type and the RCS can deliver variable thrust. Following the deorbit burn, the chaser activates two closed-loop controllers. First, the relative distance between the end bodies is controlled to maintain a small tension in the tether, and second, the chaser’s orientation is controlled to ensure the correct attitudeof the chaser. The end bodies are connected by a discretized viscous-elastic tether. Based on the attitude motion analysis [18,19] of the inactive Envisat satellite, which is the focus of the e.Deorbit mission , a representative ADR mission scenario is considered which accounts for small, residual initial angular rates of the target prior to the deorbit burn. Furthermore, no input shaping of the deorbit burn is used. The main thrust is modeled as a step function, which accounts for simplified and worst-case approach. However, the reader should note that a combination of discrete deorbit burn shaping and closed-loop control following the deorbit burn may improve the performance. In the proposed analysis, the emphasis is placed on avoiding the tether tangling around the target which can result in tether rupture or debris collision and thus debris fragmentation.
The performance of the onboard system was good over 2 years until the gyroscope assembly began delivering permanently misleading data to the onboard attitudecontrol system. As a consequence the orders sent to the reaction wheels were based on erroneous attitude estimation and were not related to the actual situation. The reaction wheels ran then to their limit rotation rate and stayed in this state. This happened during a non coverage period, leading to the complete depletion of the batteries… The board software entered a self-sustained cycle: the empty batteries triggered the board to boot itself, which consumed the remaining power available. A spin motion prevented a long exposure of the solar panels to the sun, rendering communication with the satellite very difficult. Finally, when the gyroscope assembly could be switched off, there was only one operational wheel left. Fortunately a complete check of the BIRD onboard systems and the payload showed good performances.
A 21 days simulation cycle shows an increase in the number of access as soon as all the ground stations are active. The number of access increases by one extra pass on an average per day. Most important factor is the redundancy in receiving data. Most of the smallsatellite missions especially CubeSats work on low bit rates. The probability of receiving bad telemetry is high therefore the communication protocols do need to be well scripted to attain the telemetry data in a best possible way. The Peruvian Satellite Network cleans up all the constraints by having redundant stations for the reception of the data. The other aspect for bad telemetry is the attitudecontrol being used for the pico satellites missions. Most of the CubeSats use the permanent magnets to control its attitude in space. The latest development in CubeSats is the use of reaction wheels or magneto- torques. Still the usage of such attitudesystems is yet to be verified for full effectiveness. The attitude system plays an important role in supporting the satellite to communicate with the ground station. Even though the permanent magnets are used to drag the tumbling in 2 axes, it still tumbles in all 3 axes, which leads to loss of contact and bad telemetry.
The consequence of this behavior is that the c. m. has to be moved towards the pivot point, and, if possible, this should be done automatically every time there is a change in the setup of the experimental satellite. To automate this procedure, the fine adjustment mechanisms are used. Before these small displaceable weights of 100 g can be used for the adjustment, the c. m. has to be tuned coarsely by moving the large weight masses on the assembly’s beams manually. In the beginning this is a trial-and-error based task to find a position where the table is stable and does not tip over or hit the stops at the maximal deflection angles. After such a rough position is found, very small change in the weights’ locations are necessary to locate the c. m. somewhere below the pivot point. At this point the system should be very stable, and the weights can now be symmetrically moved up- wards until they reach the point where the table is still stable, but a fine adjustment in the z or gravitational direction suﬃces to render the table unstable.
For the validation of the formation control strategy, the Test Environment for Applications of Multiple Spacecraft (TEAMS), a test facility for satellite formations and swarms based on air cushion vehicles, at the Institute of Space Systemsof the DLR in Bremen, Germany, is used. Two 5-DoF vehicles are floating on a granite table with a total experiment area of 5m x 4m. Each vehicle has a thruster system, a reaction wheel system and its own onboard computer running the control algorithms. A laser, attached to the first vehicle, is pointed at a fixed target on the other vehicle to demonstrate the telescope application. The Clohessy-Wiltshire equations are used for the guidance and feedforward controlof the satellites position in orbit. A second controller is used to calculate the correct pointing angles for the laser and the target. The control law is based on the relative distance and velocity of the two satellites. To achieve the optimal controlof the linear system, LQR controllers are used together with feedforward for optimal tracking and to cover the nonlinearities. The characteristics of the facility and its devices are introduced, the control algorithms for both satellites are explained and simulation and HIL test results are presented.
The magnetic coil system is an attractive method of desaturation for low Earth orbit (LEO) satellites due to the relatively high magnetic field intensity at lower elevations. This system also has high reliability since it includes only simple static devices (a magnetometer, a signal processor, and three coils). Other advantages are that it does not depend on a fuel supply and is much lighter than the simplest low specific impulse thruster system. Disadvantages of this system are that it may require significant amounts of power at higher altitudes. Also, coil commands may last over a large fraction of the orbit (or over several orbits) to reach desaturation. Magnetic systems may also interfere with the operation of certain payloads. The other well-known technique used for desaturation are mass expulsion torqueing. Thrusters are usually used to desaturate the momentum storage systems. In operation, a jet is fired to produce a torque opposite to the direction of the accumulated angular momentum while the satellite is commanded to maintain its attitude. This results in a wheel acceleration that counteracts the applied torque. For a desaturation system using body-fixed offset roll/yaw thrusters, the efficiency of the system can be defined as the ratio of the daily secular momentum increase to the angular impulse provided by the thruster. A reasonable design value for the efficiency is about 80 percent .
Looking at different sensors and actuator concepts it turned out that an active magnetic attitudecontrol scheme can ful- fill the requirements. The major advantage of such a sys- tem is its simple and cost efficient design. Nevertheless it is still a challenge to design, build and operate it. Three orthogonal arranged magnetic torquers are used as actua- tors. They can spin up the satellite and provide sufficient torque to compensate external disturbances. The key issue during the selection of the magnetic torquers is the step response time which has to be short enough to allow the maximum rotation rate. In addition to the magnetic tor- quers a passive nutation damper is build into the satellite to provide a sufficient damping of unknown nutation. Two cold redundant magnetometers are used as sensors which can measure the geomagnetic field. They are based on the AMR effect which results in a much more cost effective design than a flux gate technology. Additionally several sun sensors are used which are able to provide two axis information of the Sun direction. Inertial angular rates are measured by a set of four fiber optic rate gyros which un- dergo a delta qualification to meet DLR internal quality as- surance requirements. GPS-Receivers form DLR are used as main navigation sensor. A summery of all sensors and actuators with their key figures can be seen in table 1. V. ATTITUDE DETERMINATION
The dynamics models of the two docking satellites can be conveniently expressed in the body frames for the Chaser satellite D and Target satellite T defined at its center of masses (CoM). Fig. 3 shows the frames and the concept of HIL docking simulator. The satellites can be assumed purely as a free-floating object, i.e., the effect of earth gravity is neglected. When the two satellites are in prox- imity distance for docking along the V-bar direction, they are in the same orbit. Thus, the celestial mechanics effect is negligible in comparison with the contact forces dur- ing docking operation. Therefore, the orbital frame can be assumed to be an inertial frame G because the orbital dynamics has been ignored .
This process repeated until the end of the simulation period (one day for our experiment here) and the results were communicated back to the Control Optimisation algorithm for further processing. The coupling between Matlab and EnergyPlus was achieved using the Building Controls Virtual Test Bed (BCVTB) co-simulation software (v8.6, Lawrence Berkeley National Laboratory, Cyclotron Rd, Berkeley, CA, USA) [ 97 ]. For all the experiments, we used a MacBook Pro laptop, equipped with a 2.8 GHz Intel Core i7 processor and 16 GB of RAM and we did not use any form of parallel computation. For the GP_SS implementation, we used the GPy library [ 98 ] for the Gaussian Process definition and the Sequential Least Squares Programming (SLSQP) algorithm [ 99 ] of the pyOpt Python library [ 100 ] for the constrained optimisation setup.