apparatus

Abstract

The present invention relates to a system for attenuating disturbances, the system comprising: an interface; a proof mass actuator, PMA, arrangement arranged on the interface, and arranged to oscillate a mass in response to control signals; at least one passive filtering element, having a predetermined mass structurally coupled to the interface; and, control circuitry arranged to provide control signals to the PMA arrangement, wherein the control signals are arranged to control oscillations of a mass.

Description

[0001] Apparatus

[0002] Technical Field

[0003] The present invention is concerned with vibrational control in highly sensitive systems. There is presented a disturbance attenuating arrangement for use in systems such as may be used in micro-gravity missions, optical missions requiring high stability platforms and scientific or Earth observations missions.

[0004] The present invention may be easily applied to space technologies as well as other ground- based systems that may benefit from mitigation or removal of disturbances from components within the systems.

[0005] The generation of vibrations in sensitive systems is sometimes unavoidable; the present system provides an effective and robust mechanism that may account for these vibrations. The present invention is more easily integrated and installed than solutions that use removal of elements creating the vibrations, as often such vibration-generating elements are important functionally.

[0006] Components such as wheels, control momentum gyros, and cryo-coolers may be vital, or highly useful, to various experimental systems used in micro-gravity and in high stability Earth observation or Science missions. A modular arrangement is provided that can be used to improve significantly the disturbance mitigation within such systems.

[0007] The present system provides a mechanical interface that can be installed into an arrangement comprising a sensitive payload and a disturbance source that can broadly isolate the payload from the vibrations generated by the disturbance source.

[0008] Summary of the Invention

[0009] Aspects of the invention are set out in the accompanying claims.

[0010] Viewed from first aspect there is provided a system for attenuating disturbances, the system comprising: an interface; a proof mass actuator, PMA, arrangement arranged on the interface, and arranged to oscillate a mass in response to control signals; at least one passive filtering element, having a predetermined mass structurally coupled to the interface; and, control circuitry arranged to provide control signals to the PMA arrangement, wherein the control signals are arranged to control oscillations of a mass.

[0011] Thus, according to an invention, vibrations and disturbances can be efficiently and effectively removed from the system without removal of the elements that create the vibrations and disturbances. In particular, a gain in disturbance mitigation in the regions of 10 Hz to 500 Hz is achievable by the present modular system.

[0012] In effect, a modular system is presented that can account for disturbances from torques generated by wheels, Solar Array Drive Mechanism (SADM) harmonics as well as cryosystem harmonics. The modular system may be applied to the measurement system and therefore significantly decrease the impact of such vibrations on the measuring equipment within the system. This in turn leads to a more robust and reliable system which provides, with greater accuracy, measurements and experimental results. The system is lightweight and therefore flight-efficient. When arranged in a micro-gravity experiment, for example, this increases the reliability of the measurements and therefore reduces the likelihood of a need for repeated experiments. This in turn has a beneficial environmental impact associated with fewer repeat missions.

[0013] In an example, the PMA arrangement is arranged to operate in the frequency range of from around 15 Hz to around 100 Hz. It has been found that this range of frequencies is particularly advantageous for damping in aerospace payload arrangements.

[0014] In an example, the at least one passive filtering element is arranged to provide passive filtering in the frequency range of from around 30 Hz to around above 1000 Hz. The passive filtering element may comprise several passive elastomer elements or metallic springs or the like. The combination of passive filtering and active PMA vibration mitigation provides a very robust and broadband damping in relevant frequencies for payload arrangements. Furthermore, this system reduces the negative impacts of each of the individual damping systems.

[0015] In an example, at least one passive filtering element is an elastomer-based or metallic-based passive filtering element and wherein the at least one passive filtering element is arranged, in use, between the interface and a payload. It has been found that these materials are highly effective for vibration mitigation in the relevant frequency ranges for aerospace application.

[0016] In an example, the PM A arrangement comprises two PMAs for mitigating disturbances along relevant degrees of freedom for the disturbance sources.

[0017] In an example, the PMA arrangement comprises at least three PMAs, wherein the first PMA is arranged to oscillate on a first axis; the second PMA is arranged to oscillate on a second axis; and, the third PMA is arranged to oscillate on a third axis. Greater number of PMAs increases the degrees of freedom that can be effectively mitigated against vibrations.

[0018] In an example, the PMA arrangement is arranged to operate in response to a control signal from the control circuitry, wherein the control circuitry is arranged to use the response of a sensor set at payload level to generate the control signals. By designing the system to provide control signals based on the response of the sensor set at the payload level, a highly effective and easily tuned system for vibration mitigation can be provided. Such self-tuning is particularly useful when a human user is unable to easily provide tuning. In aerospace applications in-person tuning is at times not feasible. Therefore, remote tuning is advantageous when the payload is not accessible.

[0019] In an example, the payload response is measured by sensors. The use of sensors further improves the self-reliability of the present system for self-tuning.

[0020] In an example, the control circuitry comprises a low power data processor and communication electronics. In an example, the control circuitry includes sensors acquisition, controller commands computation unit and PMA drive amplifiers.

[0021] In an example, the PMA arrangement comprises at least six PMAs. In this way, vibrations in six degrees of freedom may be mitigated. In an example, the system provides 10 to 40 dB damping. In an example, the system provides around 10 dB damping in broad-band performance and around 40 dB damping in narrowband performance. Each of these are relevant for aerospace use.

[0022] Viewed from another aspect there is provided an arrangement for use in a spacecraft comprising the above system.

[0023] In an example, the arrangement further comprises a payload and a disturbance source, wherein the system is arranged functionally between the payload and the disturbance source. This arrangement advantageously protects the payload from the disturbance source and uses the vibration mitigation system to do so.

[0024] In an example, the disturbance source is at least one of a reaction wheel and a cryo-cooler.

[0025] Viewed from yet another aspect there is provided a method of damping disturbances comprising: providing a noise signal to an arrangement comprising a payload and a disturbance source and the above system; detecting the noise signal by at least one sensor on the payload; receiving, by the control circuitry, a measured output signal from the at least one sensor; obtain a transfer function in a frequency domain relating to the input noise signal sent to the PMAs and the measured output by the sensors; configuring the proof mass actuator arrangement and the at least one passive filtering element to provide damping between the payload and the disturbance source.

[0026] Brief Description of the Drawings

[0027] One or more embodiments of the invention will now be described, by way of example only, and with reference to the following figures in which:

[0028] Figure 1a and 1b show a schematic of a system in which the present arrangement may be used;

[0029] Figure 2 shows a schematic of a system for attenuating disturbances according to examples;

[0030] Figure 3 shows a schematic of a system for attenuating disturbances according to examples;

[0031] Figure 4 shows a schematic of a system for attenuating disturbances according to examples;

[0032] Figure 5 shows a graph plot of an isolator transfer function with frequency in Hz against amplification for a series of Q values;

[0033] Figure 6 shows a schematic of an overall system interconnection according to examples;

[0034] Figure 7 shows a schematic of an overall system interconnection according to examples; and

[0035] Figure 8 shows a payload with its axes and an arrangement of two PMAs according to an example.

[0036] Any reference to prior art documents in this specification is not to be considered an admission that such prior art is widely known or forms part of the common general knowledge in the field. As used in this specification, the words “comprises”, “comprising”, and similar words, are not to be interpreted in an exclusive or exhaustive sense. In other words, they are intended to mean “including, but not limited to”. The invention is further described with reference to the following examples. It will be appreciated that the invention as claimed is not intended to be limited in any way by these examples. It will also be recognised that the invention covers not only individual embodiments but also combination of the embodiments described herein.

[0037] The various embodiments described herein are presented only to assist in understanding and teaching the claimed features. These embodiments are provided as a representative sample of embodiments only, and are not exhaustive and / or exclusive. It is to be understood that advantages, embodiments, examples, functions, features, structures, and / or other aspects described herein are not to be considered limitations on the scope of the invention as defined by the claims or limitations on equivalents to the claims, and that other embodiments may be utilised and modifications may be made without departing from the spirit and scope of the claimed invention. Various embodiments of the invention may suitably comprise, consist of, or consist essentially of, appropriate combinations of the disclosed elements, components, features, parts, steps, means, etc, other than those specifically described herein. In addition, this disclosure may include other inventions not presently claimed, but which may be claimed in future.

[0038] Detailed Description

[0039] An invention described herein relates to vibration mitigation for use with an arrangement comprising a disturbance source and a sensitive payload. A particular system for this invention may be Earth observation missions in spacecraft with reaction wheels or cryocoolers or the like that generate vibrations that are preferably mitigated.

[0040] Figure 1a shows a simple schematic overview of an arrangement 10a wherein a disturbance source 14 is separated from a sensitive platform or payload 16. The element 12 for reducing the impact of the disturbances from the source of disturbances 14 is located functionally between the disturbance source 14 and the spacecraft (S / C) 16. The S / C may include, refer to or represent the platform or payload, i.e. that which is to be isolated. “Functionally between” refers to the arrangement being such that disturbances from the disturbance source 14 pass the vibration mitigation system 12 prior to reaching the sensitive payload 16. There may be additional elements between the vibration mitigation system 12 and the disturbance source 14 and / or the sensitive payload 16, however at least the disturbance source 14 is separated from the sensitive payload 16 by the vibration mitigation system 12. Force cells may be set between system 12 and payload 16 to measure the total transmitted forces to payload 16. The total forces is the sum of the disturbance forces generated by disturbance source 14 and the forces generated by the system 12.

[0041] In the example of Figure 1a, the disturbance is isolated. The arrangement may be referred to as an “isolated disturbance” arrangement 10a. The disturbances, from source 14, are reduced and therefore impact the larger system less as the disturbance source 14 is located functionally-adjacent to the vibration mitigation system 12 presented herein. In practice, this may be an arrangement wherein the disturbance source 14 is located immediately above a mechanical interface of the vibration mitigation system 12. In the presently disclosed arrangement, when ideally arranged, the total forces sum to zero across the impact of the element 14 and element 12. The vibration mitigation system 12 perfectly mitigates the vibrations from disturbance source 14.

[0042] When implemented, the presently disclosed arrangement aims at reducing the total forces as much as possible. The reduction may be limited by several factors such as sensor noise, actuators maximum effort, and controller stability constraints. It may be that, in practice, a total null force is not often achieved. A reduction by a factor at least 10 can be readily achieved. In preferred embodiments, a reduction by a factor 100 can be achieved. Figure 1b shows a simple schematic overview of an arrangement 10b wherein a disturbance source 14 is separated from a sensitive payload 16. In the example of Figure 1b, the disturbance source 14 may comprise S / C disturbances (including any disturbances generated by the platform, i.e. the wheels and cryo-coolers and the like). In the example of Figure 1b, the vibration mitigation system 12 may be arranged towards the sensitive payload 16 and the disturbances from the source 14 on the payload 16 are reduced. The payload 16 may be described as isolated. The interface is stabilized and therefore the measurements of the payload (if the payload is a measurement device) are improved. This may be particularly useful in the event of use of accelerometers. For example, in Figure 1b sensors may be accelerometers or a similar type of inertial sensor such as high frequency gyroscopes or the like. In practice, this may be an arrangement wherein the sensitive payload 16 is located immediately above a mechanical interface of the vibration mitigation system 12.

[0043] The arrangement of Figure 1b may differ from that of Figure 1a as in one arrangement the disturbance source 14 may be isolated and in another arrangement the sensitive payload 16 may be isolated. In both instances, the vibrations from the disturbance source 14 are mitigated before the vibrations are able to impact the measurement equipment 16 (or other similarly sensitive payload). In this way, the vibration mitigation system 12 provides a modular system that can be added to an arrangement containing disturbance sources (or the like) 14 and measuring equipment (or the like) 16 to improve the functioning of the measuring equipment 16. The interface between the disturbance sources 14 and the measuring equipment 16 is partially, or fully, decoupled by the vibration mitigation system 12.

[0044] Referring now to Figure 2, there is shown an example of an arrangement 200 comprising a disturbance source, a vibration mitigation system and a sensitive platform or payload. The vibration mitigation system is arranged between the disturbance source and the sensitive platform or payload. In the example of Figure 2, the disturbance source may be a reaction wheel or a cryo-cooler (or the like). In the example of Figure 2, the sensitive payload may be for example an optical bench where different payload sensors, telescopes or the like are mounted. The disturbance source in this case may be for example a cryo-cooler which may be mounted on the same optical bench. The disturbance source need not be directly mounted on the payload sensor, rather the disturbance source is connected to the sensitive payload in such a way that disturbances from the disturbance source can transfer to the sensitive payload. In this way, such disturbances are advantageously removed (dampened etc) from impacting the sensitive payload. In the example of Figure 2, arrangement 200 comprises a disturbance source 210, a satellite body 220 and a vibration mitigation system 230. The disturbance source 210 may be a reaction wheel with, e.g., six degrees of freedom for movement. Such disturbance source 210 would generate vibrations and disturbances that may impact items connected to the body 220. The body 220 may be a spacecraft platform or a sensitive interface supporting a sensitive payload element such an optical bench or a telescope or other similar detecting, measuring or observing element. The part 220 may be able to move with six degrees of freedom. It may be preferable however for the body 220 to be completely stable during observation or other use. As such, it is advantageous for the vibrations in different directions from the disturbance source 210 to be mitigated in all directions of free movement. The system 230 is arranged to reject internal disturbance sources produced by high frequency mechanisms (such as wheels or cryo-coolers or the like).

[0045] Disturbance sources considered may be cryo-coolers and one or more reaction wheels. When considering degrees of freedom, a single wheel may have five degrees of freedom requiring damping; two radial forces; two radial torques; and, one axial force. A set of wheels (two or more) may have six degrees of freedom as the wheels are not (likely) aligned in order to generate torques in all directions. For a cryo-cooler alone, it may be that the number of degrees of freedom is theoretically only one (along the cryo-cooler piston axis). The cryo- cooler may comprise two pistons translating in opposite directions. The residual force (that to be accounted for by the single PMA) is the difference between the two opposing forces. The PMA may be arranged along that common piston axis and counteract the residual force.

[0046] In practice, disturbance sources generally generate disturbances of around five degrees of freedom.

[0047] In the example of Figure 2, any disturbance source is considered (i.e. considering one or more degrees of freedom).

[0048] The vibration mitigation system 230 has a number of components in the example of Figure 2. Some of the components are not required for the system to function. The vibration mitigation system 230 may comprise a proof mass actuator (PMA) 240, a passive filtering element 234 and control circuitry. In the example of Figure 2, the system 230 comprises six PM As 240a, 240b, 240c, 240d (two not shown) and four passive isolators 234a, 234b (two not shown). The PMAs 240a, 240b, 240c, 240d and isolators 234a, 234b are arranged either side of a central section of the system 230. Two PMAs 240a, 240c are connected to element 242a and project in different directions (in the example shown vertical and horizontal). Two PMAs 240b, 240d are connected to element 242b and project in different directions (in the example shown vertical and horizontal). PMAs 240a, 240b project vertically in the example shown. PMAs 240c, 240d project horizontally in the example shown.

[0049] The isolator elements 234a, 234b are part of the isolator filtering system which may comprise 3, 4 or more isolator elements. The number of isolator elements may be based on the configuration of the arrangement and so can be chosen to provide a suitable filter response.

[0050] In examples, the present disclosure provides a system that includes at least two PMAs to mitigate disturbances along two degrees of freedom. The system in these examples include at least one passive damping element such as a passive filtering element to limit transmission of high frequency disturbances. Such a passive filtering element may comprise passive elastomer elements and / or similar resilient members such as metallic springs. In these examples, control electronics are used and may include sensor(s) electronics, controller commands computation unit and PMA drive amplifiers. These provide a system that is able to mitigate vibrations in such an arrangement of a sensitive platform or payload.

[0051] The system 230 may comprise a larger number of PMAs 240a, 240b and isolators 234a, 234b, and the number of PMAs 240a, 240b need not match the number of isolators 234a, 234b. The number of PMAs need not match the number of isolators as isolators normally provide isolation in six degrees of freedom, which is not the case for PMAs. The PMAs 240a, 240b and isolators 234a, 234b may be arranged in any arrangement within the system 230 although some level of symmetry (e.g. six-fold symmetry around a central longitudinal axis of the arrangement 200) has been found to be beneficial for multi-dimensional vibration mitigation.

[0052] The system 230 may further comprise a flexible support 232 and an active plate 233. In the specific example of Figure 2, the flexible support 232 is included and connects the disturbance source 210 to the active plate 233. The active plate 233 is able to move. In an example, the active plate 233 is able to move in six degrees of freedom. As such, the movement of the disturbance source 210 may be transmitted through the flexible support 232 to the active plate 233. The flexible support 232 mitigates partially against some of the disturbances of the disturbance source 210. In some examples, the plate may be able to move in three degrees of freedom and attenuate vibrations. In some examples of the arrangement, the number of PMAs might differ from that in Figure 2. In some examples, the number of PMAs may be reduced. For example, if a payload is mainly sensitive along a single translational axis (e.g. a cryo-cooler as discussed above), then it is feasible to use a single PMA along this axis to dampen the vibrations. The single arrangement PMA may however generate additional torque disturbances which themselves may preferably be mitigated.

[0053] Similarly, in another example, the payload may be sensitive to two different translational axes. In this example, it is feasible to arrange two PMAs along these two identified axes. The two PMAs provide damping along the relevant axes such that the vibrations can be dampened. Depending on the rest of the arrangement, the PMAs may generate additional torque disturbances which may need to be dampened. The same may be said for a payload that is sensitive three translational forces which requires only three PMAs.

[0054] The active plate 233 may be formed of reasonably light material and may be broadly in any shape. The plate 233 may be a disk that supports elements in the system 230. The payload to be isolated or the disturbance source to be mitigated may be placed on top of the disk. The plate 233 is an interface on which either the disturbance source 210 or the sensitive payload 220 may be arranged. The plate 233 assists vibration mitigation in conjunction with the PMAs 240 and the isolators 234.

[0055] In the example of Figure 2, the plate 233 may be formed from light materials. The disturbance sources are mitigated and the plate 233 has mechanical interfaces with the disturbance sources (e.g. a wheel(s) or cryo-cooler(s)), the PMAs, the passive isolator system and sensors.

[0056] The PMAs 240a, 240b are connected to the active plate 233 by elements 242a, 242b. In the example of Figure 2, PMA 240a is connected to the active plate 233 by element 242a. In the example of Figure 2, PMA 240b is connected to the active plate 233 by element 242b. The PMAs 240a, 240b are arranged to counteract movement of and vibrations from the disturbance source 210 in different axes of movement. For example, a first PMA 240a may isolate and account for movement of and vibrations from the disturbance source 210 in a first dimension (e.g. an X-axis [in Cartesian], or a polar axis [in cylindrical]). A second PMA 240b may isolate and account for movement of and vibrations from the disturbance source 210 in a second dimension (e.g. a Y-axis [in Cartesian], or a longitudinal axis [in cylindrical]). Further PMAs 240 may be included in arrangements to isolate account for such disturbances in further dimensions. Translational and rotational aspects are to be considered when designing the vibration mitigation system. Additional PMAs 240 may be provided to improve the isolation and mitigation of such disturbance, and therefore improve the fault tolerance of the overall arrangement 200.

[0057] In practice, one PMA may not fully mitigate disturbances along its direction. The PMAs instead are prepared as a whole and treated as a vibration mitigation system. The PMAs are therefore arranged and operated to provide a net vibrational mitigation along a number of axes. While a PMA can mitigate disturbances along a direction, one PMA will likely introduce a torque and therefore this requires additional mitigation. The arrangement can be provided to fully mitigate vibrations considering also the disturbances introduced by the active damping elements.

[0058] In a specific example, the number of PMAs 240 may correspond to the number of degrees of freedom to be isolated. The number of PMAs 240 may be related to the required redundancy of the system 230. The position and orientation of the PMAs 240 may be based on the accommodation constraints. For example, for a large payload 220 or disturbance source 210, the PMAs 240 may be set under the plate 233. In another example, for a small payload 220 or disturbance sources 210, the PMAs 240 may be set on top of the plate 233 to improve the overall compactness of the system 230.

[0059] The PMAs 240a, 240b provide active damping. The PMAs 240a, 240b may be considered as active elements of vibration mitigation. The PMAs 240a, 240b provide a damping to frequencies in a low to mid frequency range of the spectrum. In an example, the PMAs 240a, 240b are arranged to provide a damping in the region of around 15 Hz to around 100 Hz.

[0060] In an example, the PMAs may be referred to as “shakers”. The PMAs may be actuators for providing vibration mitigation.

[0061] The selection of the sensing technology and placement may be based on the isolation requirement for the system 230 accounting for the arrangement 200 as a whole. This selection may also take into account the specific mission for the arrangement 200 (e.g. Earth observation or micro-gravity experiments etc.). The system 230 is modular and may be connected to at least one of the disturbance source 210 and the platform or payload 220. The apparatus may comprise six PMAs to perform a six-degrees-of-freedom isolation of force and / or torque. In the event that only force (not torque) is to be isolated, three PMAs may be sufficient for highly accurate and effective isolation and mitigation. Selection of the number of PMAs may be based on the vibration mitigation required and the space constraints of the arrangement 200. The PMAs may be arranged to dampen the disturbance generated by the PMAs, i.e. the PMAs are used to dampen vibrations from the disturbance source as well as the vibrations from the other PMAs. As such, this is a fully dampening system.

[0062] Elements 242a, 242b may be formed of metal or a similarly relatively rigid material. In examples, the elements 242a, 242b are formed of stiff aluminium or the like. In examples ,the elements 242a, 242b are small aluminium structures that provide an interface for the PMAs 240a, 240b, 240c, 240d and the plate 233. The shape of the elements 242a, 242b may depend on the angles between the main axes of the PMAs 240a, 240b, 240c, 240d and the active plate normal axis.

[0063] The active plate 233 is connected to isolators 234a, 234b. The isolators 234a, 234b may be described as being structurally coupled to the active plate 233. This may be in the form of physical connection or chemical connection (such as by soldering, welding or adhesion or the like). It is important that the isolators 234 are able to receive and filter vibrations generated by the disturbance source 210 on one side of the plate 233 to reduce the vibrations through the plate 233 to the payload 220. In the example of Figure 2, an upper side of the plate 233 is connected to elements 242a, 242b, the elements 242a, 242b being connected respectively to PMAs 240a, 240b, and the flexible support 232. In the example of Figure 2, a lower side of the plate 233 is connected to the isolators 234a, 234b. In examples therefore, the PMAs 240a, 240b may be on opposite sides of the plate 233 to the isolators 234a, 234b.

[0064] The plate 233 need not be in the form of a “plate”, in the sense of being flat. The plate 233 may be a 3D element that allows movement in the same number of dimensions as the disturbance source (in the example of Figure 2 the reaction wheel 210). The plate 233 therefore may be a block or the like. The plate 233 may take any shape that performs well in the space provided within the arrangement 200. The isolators 234a, 234b may be formed of resilient materials such as elastomers or the like and are arranged to provide a filtering of frequencies above 100 Hz. Elastomer-based isolators are present available and can be used in the present system as shown in Figure 2. The isolators 234a, 234b provide a passive filtering. The isolators 234a, 234b provide a filtering to frequencies in high frequency range of the spectrum. In an example, the isolators 234a, 234b are arranged to provide a filtering in the region of around typically above 100 Hz (i.e. an entire rejection above 100 Hz). In an example, the isolators 234a, 234b are arranged to provide a filtering in the region of around 100 Hz and above. The isolators 234a, 234b may be formed of an elastomer based and / or metallic based element. The isolators 234a, 234b may be arranged below the plate 233. The isolators 234a, 234b may be arranged around a central longitudinal axis of the arrangement 200. The isolators 234a, 234b may be arranged with some level of symmetry around at least one of the reaction wheel 210 and the sensitive payload 220. Some passive amplification occurs at around typically 30 Hz. As such between about 20 Hz and 100 Hz active damping compensates for disturbances and for this passive amplification.

[0065] As used herein, damping and filtering both refer to reducing of vibrations. More specifically though, (low pass) filtering relates to reduction of the impact of high frequency disturbances: the higher the disturbances frequency, the better the rejection. Damping relates to the reduction of the amplification at the resonance peak: the higher the damping, the lower the peak at the resonance. Filtering may be used in relation to high frequencies and damping applies to a more limited frequency zone around a resonance peak. In some examples, the higher the damping, the lower the filtering.

[0066] The isolators 234a, 234b connect to sensors 236a, 236b. The sensors 236a, 236b may be force sensors. The sensors 236a, 236b may be piezo-electric force cells and are arranged to measure an amount of force transmitted through a sensor in response to an external stimulus, such as shaking or the like from the disturbance source. The sensors 236a, 236b may be three axis force sensors. In an example, there are four such sensors providing a total of twelve axis sensors. This can, in practice, be treated as a six axis sensors to account for three degrees of freedom in force and three degrees of freedom in torque. The satellite body 220 (or platform, or bench or the like) may move in six degrees of freedom and therefore the sensor arrangement preferably operates in six degrees of freedom to account for all motions experienced by the platform 220. The sensors 236a, 236b are connected to the platform 220. They consists of three-axis force cells (sensors) as described above. The photoreceptor 222 is set on the platform 220 and can move around six degrees of freedom. The main degrees of freedom however are the two rotational axes around its transverse axis. This is not to say the damping system only mitigates vibrations in such axes but rather that in minimum a system could be provided to dampen in such degrees of freedom.

[0067] The flexible attachment 250 is arranged to connect to the platform or payload 220. The flexible attachment 250 is not required in the arrangement of Figure 2. The present invention does not require such an arrangement, for example where the wheels are directly set on the active plate. The flexible attachment 250 may correspond to, e.g., a solar panel attached to the body of a spacecraft. The microvibration control system may be used to isolate the reaction wheel vibrations and can be tuned based on the structural dynamics of the satellite underneath it.

[0068] The arrangement 200 may also comprise various electronic equipment for providing electrical communication between the relevant components of the arrangement 200. In an example, the arrangement 200 further comprises control circuitry in communication with the PMAs 240. The control circuitry is arranged to generate active isolation commands and send these to the PMAs 240. The commands may be signals that are generated via advanced robust control to provide a generic solution to the issue of multi-dimensional vibrations.

[0069] The controller in the arrangement is designed and tuned using robust control techniques. These techniques aim at optimizing the controller to meet stability and performances objectives in the presence of uncertainties, using a model of the complete system from sensors to actuators.

[0070] The controller operates fully autonomously. The coefficients of the controller are computed using robust control techniques or using data driven techniques. The controller can be scripted to assist adapting the controller to new configurations.

[0071] The isolators 234 provide a passive filtering and the PMAs 240 provide active damping. The elements are used together in the arrangement 200 to provide damping between around 10 Hz to around 500 Hz. This accounts for the vast majority of likely relevant vibrational frequencies from elements such as wheels and cryo-coolers. As such, the present system provides a highly resilient vibration mitigation for sensitive payloads from disturbance sources. In an example presented herein there is provided a mechanical interface that provides bi-directional adaptive narrow-band or broad-band isolation between a sensitive payload and a disturbance source.

[0072] In an example, the system may be used to isolate the harmonic source (i.e. located at the disturbance side of the arrangement) and in another example, the system may be used to stabilise the active plate on which the sensor is located (i.e. located at the sensor side of the arrangement).

[0073] Referring now to Figure 3, there is shown a further example of a system according to examples herein. Figure 3 shows an arrangement that is similar to the arrangement of Figure 2. Similar features (i.e. features providing the same or a similar function) are shown with their numerals increased by 100. E.g. the satellite body or platform 220 of Figure 2 is shown as satellite body or platform 320 in Figure 3.

[0074] The arrangement 300 of Figure 3 has a payload 322 connected to the platform 320 by passive isolators 334a, 334b. Connected to the payload 322 are inertial sensors 336a, 336b, 336c, 336d. Also shown are PMAs 340a, 340b, 340c, 340d. The four PMAs are connected to element 342a (for PMAs 340a, 340c) and element 342b (for PMA 340b, 340d). The PMA 340a and PMA 340c project in different directions (in the example shown vertical and horizontal). The PMA 340b and PMA 340d project in different directions (in the example shown vertical and horizontal).

[0075] The arrangement of Figure 3 is a more general overview of the present invention. The system provides the advantages as discussed above in terms of highly effective vibrational reduction for sensitive payloads.

[0076] In the arrangement of Figure 3, there is no active plate as the PMAs, isolators and sensors can be directly accommodated on the payload. There may also be no active plate to limit mass and to interface with the sensitive payload, the PMAs, the isolator system and the sensors.

[0077] Referring now to Figure 4, there is shown a further example of a system according to examples herein. Figure 4 shows an arrangement that is similar to the arrangement of Figure 3. Similar features (i.e. features providing the same or a similar function) are shown with their numerals increased by 100. E.g. the satellite body or platform 320 of Figure 3 is shown as satellite body or platform 420 in Figure 4.

[0078] In the arrangement 400 of Figure 4, the payload 422a is arranged to one side of the platform 420 while the isolators 434a, 434b and force sensors 436a, 436b are arranged on another. The payload 422a is arranged between the platform 420 and a photoreceptor (or the like) 422b.

[0079] There is an active plate 433 in the arrangement of Figure 4 (in contrast to Figure 3) arranged between the platform 420 and the PMAs 440a, 440b, 440c, 440d. The PMAs are arranged as per Figure 3 to elements 442a, 442b.

[0080] Also shown in Figure 4 is a disturbance source 410. The disturbance source may be a single cryo-cooler. The disturbance source 410 may be one or more wheels, such as reaction wheels or the like. The vibrations generated during the use of the source are those that are to be mitigated by the disclosed vibration mitigation system 400.

[0081] Referring now to Figure 5, there is shown a graph plot 500 of an isolator transfer function with frequency in Hz against amplification for a series of Q values. The graph plot 500 shows an X-axis of frequency in Hz 520 set against a Y-axis of amplification 530. Plots for a number of Q values are shown. The plot for Q=2 is shown as line 502. The plot for Q=5 is shown as line 505. The plot for Q=10 is shown as line 510. The active control domain and passive isolation domain are shown as regions A and B respectively.

[0082] The graph plot 500 shows the relationship between high Q factor, the filtering performance of the passive isolator and the amplification at the resonance frequency. Specifically, graph plot 500 shows that the higher Q factor, the better the high frequency filtering performance of the passive isolator. However, the higher the Q factor the higher the amplification at the resonance frequency (can be seen by the relative heights of the peaks of the different Q factor lines, Q=2 is lower 502 than Q=5 505 which is lower than Q=10 510.

[0083] As shown in Figure 5, the passive filtering provides benefits and drawbacks. The active damping of the present invention therefore aims to mitigate the impact of the passive amplification alongside mitigation of disturbances. The arrangement therefore self dampens alongside dampens the vibrations of the disturbance sources. The PMAs enable the use of high Q-factor isolators, as the drawbacks associated with such isolators can be mitigated by the PMAs. The system provided herein therefore provides high frequency passive filtering and low frequency disturbance mitigation.

[0084] The system presented herein attenuates micro-vibrations generated by disturbance sources on sensitive payloads. In an example, the disturbance sources may be spacecraft disturbance sources such as wheels and cryo-coolers or the like. In an example, the payloads may be spacecraft payloads such a micro-gravity experiments, telescopes, cameras or other optics, or the like. The present system is able to attenuate these microvibrations up to a factor of 100.

[0085] The system herein therefore provides vast improvements over the presently available systems. In particular, the present system enables missions that require extremely high stabilization for frequencies above around 15 Hz without necessarily using cold gas. This in turn provides a low complexity system that also has benefits in terms of cost, mass and fuel consumption when used in an aerospace context. Furthermore, the present system can be attached to a “plate” or the like, and does not require a high-performance platform or the like.

[0086] Examples of missions that may be highly benefitted by this system include low-cost Earth Observation missions. Such missions are very sensitive to high frequency disturbances for the combination of the following criteria:

[0087] Lower inertia

[0088] Longer focal length

[0089] Small mirror or lens aperture

[0090] Higher disturbances in case of low-cost reaction wheels.

[0091] Each of these criteria leads to great advantages being provided by the present system in such missions.

[0092] The system may use a tuning and optimisation stage to help provide optimal performance. This involves the use of control circuitry that is able to provide control signals to the active damping provided, in an example, by a series of PMAs.

[0093] The system herein may be used in a mission to observe astronomical targets. Relevant structures in this instance include the disturbance sources (or sources) through to the pointing performance of the sensitive payload; “pointing performance” may be viewed as a general measure of the stability of a satellite (or the like) and may include relative pointing error, absolute pointing drift and absolute pointing error.

[0094] The tuning and optimization approach may therefore include developing or devising a model of the arrangement from the disturbance sources (or sources) through to the pointing performance of the sensitive payload.

[0095] Referring now to Figure 6 there is shown a diagram of the considerations from the system. Figure 6 shows an end to end system. The diagram of Figure 6 shows a schematic illustrating a model 600 of a system arranged on a satellite with its attitude control system, in closed loop with additional blocks used to tune the controller. The model shows that the design and tuning of the controller takes into account the full system and the environment in which the system is arranged in a fully coupled manner.

[0096] The model has a grey zone, pink boxes and blue boxes. The grey zone may be referred to as the integrated structural dynamics. The grey zone shows an example of the mechanical structure (wheel, the wheel’s connection to an active plate, the passive isolator, the satellite flexible structure and the attitude control system). The pink boxes represent the models of the PMAs, the force sensors and of the controller. Input blue boxes represent transfers from normalized power spectrum densities to sensor noise spectrum, and disturbances spectrum. Output blue boxes represent transfers from output signals of the system to be controlled to their normalized spectrum (so called weight).

[0097] One weight is used to limit PMA efforts in order to avoid their saturation. The other weight is used to encourage the required performance (typically attitude pointing should be below a specified value).

[0098] The diagram shown in Figure 6 does not include “uncertainties” or the like.

[0099] Referring now to Figure 7, there is shown a further diagram of the considerations from the system. The model 700 of Figure 7 is similar to the model 600 of Figure 6, however the model 700 of Figure 7 also shows “uncertainties”. The model 700 includes a series of A blocks from each of the reaction wheel, active plate, passive isolator and platform blocks. Each A block is an uncertainty block. The A blocks adds disturbance signal to the input of each (respectively) of the reaction wheel, active plate, passive isolator and platform block only based on the block uncertainties.

[0100] For example, a multiplicative uncertainty is such that the block transfer GO becomes G=G0*(1+A), or an additive uncertainty is such that GO becomes G=G0+A. For example, multiplicative uncertainty is uncertainty on the inertia by 10% (so the uncertainty factor is comprised between 0.9 and 1.1).

[0101] Figure 7 (as with Fig. 6) shows structural model dynamics (shown in grey box), sensor and actuator models (pink boxes) and a controller (pink) alongside the weighing transfer functions (blue) used to tune a controller.

[0102] The system herein uses active control to provide extremely high pointing stability, as would be advantageous for many space missions. The method of use involves producing, via modelling, an analytic model of the relevant components of the system (from disturbance source to sensitive payload). From this model, a simulation providing a “digital twin” of the real life system can be formed. The analytical model (or models) and digital twin simulation support the design of the high-performing control aspect used to design and optimise the passive filtering elements and the active damping elements. The designed passive filtering elements and active damping elements are then introduced into the real life system and tuned for optimal performance in stabilising the payload from the disturbances of the disturbance source.

[0103] The steps for use may therefore include qualification of the PMAs, isolators and the sensors in the system and arrangement. Design of the hardware architecture including the positioning and orientation of sensors and actuators within the arrangement. The location of the disturbance sources are also to be considered. Mechanical, electrical and thermal interfaces within the arrangement may be accounted for during the modelling and analysis described above. As a last stage, controller tuning by the control circuitry occurs to optimise the mitigation performance provided by the system disclosed herein.

[0104] These steps may provide an extremely well designed and installed mitigation system that provides damping over a vast range of frequencies. Linear parameter-varying control (LPV control) achieves uniformingly robust 40 dB adaptive vibration attenuation with a safety margin of 2. The harmonic disturbances generated by a wheel (or other disturbance source) are proportional to the wheel rate. LPV control allows the controller to adapt to the varying wheel rate, hence to the harmonic disturbance frequencies. The present system can therefore achieve excellent disturbance mitigation as the actuator’s energy is concentrated in the vicinity of the disturbance frequency, and counteracts its effects in the most efficient way. The goal of the Linear Parameter Varying techniques is to exploit to a great extent the physics I actuator authority and signal-to-noise ratio in the sensors. With improved sensing and / or improved actuation, greater than 40 dB attenuation is possible with the present arrangement.

[0105] The full process from system identification to controller tuning can be performed in an automatic fashion thus limiting operator workload, knowledge and competency requirements in control design. System identification and controller tuning can be performed using existing MATLAB toolboxes. Scripts making use of these toolboxes can lead to automation of the full process.

[0106] System identification comprises generating input signals to the PMAs and measuring the associated responses on the sensors. The transfer functions from PMAs to sensors can then be deduced from these input and output signals. From these transfer functions it is then possible to compute the model from actuators to sensors in the Laplace domain.

[0107] From this identified model, it is possible to add additional blocks to take into account the uncertainties in the system, stability and performance objectives. This design phase can be performed once by a skilled controller engineer, who can also create additional functions or graphical interfaces to allow e.g. a less skilled engineer to make minor modifications to some parameters.

[0108] Once the augmented plant is created or modified, another script can be used to find the best controller meeting the all the objectives using optimization techniques such as H-infinity or mu-synthesis.

[0109] In an example of a method of use of the system, the system is designed using the “digital twin” model as outlined above. The disturbance source, e.g. a reaction wheel may be arranged on a platform (plate 233 form Figure 2) and vibrations controlled using a set of PMAs linked to control circuitry for providing a control signal to the PMAs. High frequencies can be attenuated by the isolators (passive filtering elements). In use, sensors may measure forces transmitted to the plate. The control circuitry may identify and adapt to gyroscopic events and may provide a robust “worst-case” pointing performance for the arrangement. In the present arrangement, the LPV controller adapts to the disturbance frequency but also to possible dynamic changes to the wheel gyroscopic effect. Indeed, this gyroscopic effect changes the frequency of some modes with respect to the wheel rate.

[0110] The method of use of the system involves a detailed identification of the arrangement and tuning of the active damping based on the identification of the arrangement. The tuning process is based on advanced system identification and uses robust control design techniques. The process involves active control design, tuning and implementation to provide extremely effective micro-vibration mitigation on-board the arrangement (which may be e.g. a spacecraft or the like).

[0111] The digital control system (control circuitry) may be implemented as part of a central processing unit (CPU). The process involves matrix multiplications at each sampling time. The matrices may be calculated online or offline during the tuning process, this may be based on the type of controller used. The control circuitry may be deployed to a variety of microcontrollers, CPUs or field-programmable gate arrays (FPGAs). The digital output of the control circuitry may be converted to analogue actuator signals using a digital-to-analogue circuit in series with a low-pass filtering stage. Analog sensor measurements needed by the digital controllers may be filtered then sampled using analogue-to-digital converters.

[0112] The tuning process has at least two steps. The first involves obtaining a model of the system. The second involves robust control tuning and optimization.

[0113] The system identification may comprise the following steps:

[0114] Send excitation signals such as band limited white noise or multi-sines (calibration signals) to the actuators (e.g. PMAs) and record the associated sensor measurements

[0115] Compute the frequency response functions (FRFs) from the actuator input commands to the output sensors measurements

[0116] Fit a state-space model to the identified frequency response functions

[0117] - Add an uncertainty structure to the state-space model

[0118] The control tuning and optimization may comprise the following steps:

[0119] Convert the performance objectives, sensor noise and actuator limitations into corresponding weighting filters - Augment the plant model with said weighing filters

[0120] Find a controller that can optimize a given performance objective while meeting relevant constraints and providing a certain level of robust performance and stability. The tuning may be performed using a combination of convex and nonconvex optimization techniques.

[0121] Discretize the controller at a specific sample time

[0122] Verify the worst-case closed loop performances and stability using formal robust analysis tools across additional uncertainty models.

[0123] The “plant model” is a model of the system from the actuators to the sensors in the Laplace domain. It is a mathematical description of the system that includes the way actuator signals are mapped to sensor outputs, a description of the uncertainty in the dynamics and a description of the predicted noise environment. This mathematical description is then used to optimize the controller. With these criteria, a calculation to optimise the controller can be performed as the system can be understood on a vibrational level. The “plant model” is therefore the representation of the linear dynamics in the Laplace domain between actuators commands (in Volts) and sensors measurements (in Volts). This includes all elements in the system i.e. PMA dynamics, electro-dynamics, amplifiers, delays, structure dynamics with flexibilities. System identification process captures all the elements from commanded voltages to measured voltages.

[0124] The above describes an approach for using calibration signals to design control signals. Specifically, the white noise or multi-sines signals provided to the actuators are calibration signals and provide an understanding of the characteristics of the arrangement (based on sensor responses to the calibration signals). The results from these calibration signals (as detected by the sensors) are then used in the design of control signals i.e. the signals provided to the actuators to provide active damping in use.

[0125] The provision of narrowband or broadband isolation can be made based on the performance objective and accommodation of disturbance sources to be mitigated. As noted above, the controller can be based on a LPV structure that is adapted based on the desired central isolation frequency, the frequency at which maximum attenuation is to occur. This is the case for narrowband isolation. In an example, the arrangement has a reaction wheel microvibration for isolation. The central isolation frequency corresponds to the main harmonic disturbance at the same frequency as the wheel rate. This may also include other frequencies, such as another wheel harmonic. LPV control is used in the present system, in an example, to mitigate the disturbance generated in the vicinity of a single harmonic: this may be the case, for example, to mitigate the harmonic H1 of a reaction wheel due to its unbalance, or to mitigate harmonics from a cryo-cooler. In this case, the controller is a variable notch around the harmonic to be mitigated.

[0126] It is also (or alternatively) possible, in the present system, to use LPV control to mitigate several harmonics at the same time, all of them occurring at frequencies multiple of the wheel rate. In this case, the controller consists in a series of different notches around the harmonics to be mitigated.

[0127] It is also (or alternatively) possible, in the present system, to use a broadband controller which mitigates disturbances generated by many different disturbances at different frequencies. In this case the controller is not LPV but Linear Time Invariant (LTI). When using the same PMAs, expected performances in this case may be reduced compared to LPV case, as the energy is spread in a larger frequency domain.

[0128] In the present system, it is therefore possible to tune the control system to perform narrowband isolation at multiple harmonic frequencies. This is highly advantageous in the case of reaction wheel payloads. Such disturbance sources can produce multiple harmonic disturbances. In such arrangements, therefore, the present system provides a significant improvement over other systems.

[0129] At each sample time k, the digital controller (control circuitry) multiplies the sensor measurement vector y(k) and past controller state x(k) with the tuned controller matrix M in order to obtain the new controller states at for the next sample time x(k+1) and the actuator command u(k). Depending on the application case, the matrix M can be either fixed or recomputed when the scheduling parameters change.

[0130] As mentioned above, either of the payload or disturbance source may be arranged on the “top” of the disk, plate etc. The user performs system identification to get the nominal plant model from the actuators to the sensors. Subsequently, an uncertainty structure is added around the nominal model to account for all possible plant variations that might be experienced at runtime. The controller is then tuned on a standard computer for the specific plant model, uncertainty structure, requirements and constraints. Optionally, the robust performance and robust stability of the closed-loop can be verified with respect to other uncertainty models not directly considered during the tuning phase. This tuning phase allows optimisation the control signals provided to the actuators based on the physical and vibrational properties of the arrangement.

[0131] The controller matrix M including its possible parametric dependency is then uploaded to the system processing board (CPU or FPGA) to be used during use. Ideally, the system disclosed herein is accommodated inside the arrangement (e.g. spacecraft) for system identification, as the plant model between the actuator and sensors (needed for controller tuning) may depend on other mechanical elements in the arrangement that cannot be accurately predicted during testing (ground conditions may not perfectly mirror in flight conditions). In any case, once in-use in motion, if arrangement capacity allows, it is possible to perform system identification on-board by exciting the actuators and recording the associated sensor measurements. The arrangement then sends the raw identification data to the users for further processing. The user may compute and upload the new controller matrices.

[0132] The computation requirements are relatively limited. As such, the present methods of tuning may be performed on a raspberry pi or similar machine. In particular, use of low power processors is very useful in terms of electrical efficiency. Low power in this instance may refer to around 3 Watts or so.

[0133] In an example, the above may be performed in-orbit and the calculations performed on the ground. In this way, the ground users may perform calculations and improve the performance of the system during flight. This allows for maintenance of an extremely high performance throughout the lifetime of the system. This also means that if there is a structural change during the use of the system, the system can adapt to this change and provide an extremely high performance.

[0134] The system described above can operate in a broadband manner or a narrowband manner. In the narrowband case, the controller coefficients are computed in real-time in the continuous domain with respect to the disturbance frequency (which in examples may be a varying parameter). This uses the process of LPV control. Discretization introduces a frequency shift degrading the rejection performances, but the present system implements a solution to suppress this problem via real-time pre-warping. Dynamics uncertainties increase with the frequency, thus limiting the bandwidth as these uncertainties are taken into account during the controller tuning.

[0135] Pre-warping comprises compensating the frequency error that occurs when converting the controller coefficients from the continuous domain to the discrete domain. The LPV controller coefficients are computed in real-time in the continuous domain (Laplace domain) using the central notch frequency as an input parameter. This controller then needs to be converted into a discretized controller.

[0136] This conversion is usually performed by a bilinear transformation: during this transformation, the Laplace variable s is replaced by s=2 / T.(1-zA-1 / 1+zA-1), T being the controller sampling frequency, and z the discrete operator. This is known as the Tustin transformation.

[0137] We note that other transformations exists: others are simplified approximation of the real transform which is the inverse of the following transformation: z=eA(sT), T being the sampling frequency.

[0138] When performing this bilinear transform on a notch filter, the central frequency of the notch is changed: the closer this notch frequency it is to the Nyquist frequency (1 / 2T), the bigger the frequency shift.

[0139] Without performing any correction, this frequency shift would make the notch filter nonefficient. The correction to perform comprises applying a pre-warping computed at the notch frequency wO. So the bilinear transformation formulae given above is replaced by the following one : s=wO / tan(wO*T / 2).(1-zA-1) / (1+zA-1).

[0140] In a mathematical example, consider the affine case: K(w) = KO+w.KI , K are the controller coefficients, KO and K1 are fixed matrices uploaded once, so K depends on the wheel rate w.

[0141] In the broadband case, the bandwidth is limited by the sensor noise but high sensitivity structure accelerometers introduce very low noise. The bandwidth may be limited by dynamics uncertainties. PMAs (or other active damping elements) are not typically a limiting factor as they can generate very high frequency actuations (above 1kHz) to account for very high frequency disturbances. The system is able to self-adapt to the disturbances from the disturbance sources. For the variable narrowband (LPV), the disturbance frequency may be a signal stemming from the commandable source (e.g. the measured wheel rate) and used to compute the control signal in real-time. The disturbance frequency can also be estimated by an on-board real-time estimator (e.g. a running fast Fourier transform), and used to compute the controller in realtime.

[0142] For the broadband case, the self-adaptation is performed on the dynamics and not on the disturbance, and not in real-time but during the system identification process. For the broadband case, a LPV controller can be scheduled on the maximum level of measured disturbances, in order to reduce the overall gain in case the disturbances are much stronger than anticipated thus avoiding saturating or breaking the actuators.

[0143] During use, if the controller detects a higher disturbance at the payload than anticipated during control and tuning of the active damping, the controller is able to stop the operation of the system. A new system identification can occur to identify the faulty sensor or actuator (PMA). In an arrangement whereby there is a controller arranged for each degraded configuration, the controller associated with the degraded configuration is autonomously used. This provides very high fault tolerance.

[0144] This in-use self-adaptation of the vibration mitigation system may also account for changing disturbances from the disturbance sources. For example, if a cryogen in a cryo-cooler is vented, the mass of the cryo-cooler will vary over time. This may change the main frequency of the disturbance generated by the cryo-cooler. As such, the adaptation of the vibration mitigation system to a changing disturbance is important to allow high level performance over the course of a mission.

[0145] In another example, the system may implement a bank of filters, each filter being sensitive to one sensor or one actuator fault. This may increase the system availability time.

[0146] In more detail, when a fault on one sensor or one actuator is detected and isolated, it is possible to re-configure the system autonomously to the remaining healthy sensors / actuators using a controller previously tuned on ground for the new configuration (without the faulty sensor / actuator). This increases the system availability as it does not require any ground intervention to identify the fault, reconfigure the system and possibly re tune a new controller based on the new configuration. Therefore, the present system is highly adaptable and robust. This leads to increased performance and efficiency.

[0147] The above system operates effectively at both narrowband and broadband frequency damping. The control circuitry may be programmed for on-board real-time pre-warping adapted to the disturbance frequency in the case of narrowband disturbance mitigation. Narrowband mitigation can be applied to multiple simultaneous harmonics to provide a robust damping across narrow, but importantly relevant frequencies.

[0148] Although the invention described herein relates to disturbance mitigation for sensitive payloads within an aerospace framework, in particular in the example of a wheel or cryocooler creating vibrations which can disrupt micro-gravity measurements aircraft, it may also be applied to any applications where undesirable vibrations can disrupt the operation or performance of sensitive payloads. The system is flexible and is able to adapt to different kinds and sizes of sensitive payloads and disturbance sources, the system can use different mixes and arrangements of isolators and PMAs.

[0149] These applications may include automotive, space, domestic or commercial sensitive payload performance and so forth. In particular, micro-gravity missions, X-ray, gamma-ray or infrared missions and Earth observation optical missions are sensitive to high frequency disturbances and therefore will benefit significantly from the presently disclosed system. The present system may also be used on ground during the assembly, integration and test phases of sensitive payloads. Tests may be performed on the efficiency of isolation provided by the system for the sensitive payload.

[0150] The system herein is simple to manufacture, modular, adaptable to a large spectral range (i.e. many disturbances), generically applicable via use of tuning, bi-directional, scalable, non-intrusive and highly effective. The system herein has a reduced cost and reduced mass compared to modern vibration mitigation systems. The system herein an improvement over micro-propulsion systems (MPS). Such systems generate very low disturbances and therefore are seen as suitable systems however the drawbacks associated with those MPS systems are not present in the arrangement disclosed herein. The system herein has a reduced size and is fault-tolerant. The system herein is an improvement over present state of the art systems. The system herein is robust to uncertainties and self-adaptive to changing conditions. The isolation provided by the system may be multivariable, such as across six degrees of freedom with 3 positions and 3 angles. Present systems may use passive isolators alone for vibration mitigation. Such systems may involve implementing passive isolators at the base of the disturbance source or set under the sensitive payload. These systems may limit disturbances above around 100 Hz, however disturbances around the isolator resonance frequency (often around 20 Hz) are typically amplified. The system disclosed herein how uses a control system to mitigate disturbances belowbetween 15Hz and 100Hz, and image-based line-of-sight stabilisation techniques mitigate disturbances between 0.01 Hz and 20 Hz. As such, the present system may operate well in tandem with line of sight stabilization (LOSS) systems.

[0151] The two systems are complementary and most LOSS systems have a frequency range limited at the upper end to about 20 Hz. The present system has a far greater range of frequencies over which it is effective. Together, highly effective damping over an extremely large range of frequencies can be provided. The vibration-mitigation system herein can perform either broadband isolation or narrowband isolation depending on the needs of the arrangement. In particular, mission characteristics and the noise from the disturbance source can and should be taken into account for which is the most advantageous vibrationmitigation system configuration for any scenario. If narrowband isolation is required, the control system of the system can be adapted at runtime to follow one or more time-varying harmonic disturbances.

[0152] The system herein uses both of these by combining active control and passive isolators. The active control provides vibrational mitigation between 15 Hz and 100 Hz and the passive isolators provide vibrational mitigation above 100 Hz. The active control therefore accounts for the passive isolator amplification of frequencies around and below 20 Hz.

[0153] A method of use of the present system may include a design and a tuning stage. The vibration-mitigation system disclosed above is installed into an arrangement with a disturbance source and a sensitive payload. The vibration-mitigation system may be designed with the number, location and orientation of the sensors and vibration sources in mind so that the number, location and orientation of the isolators and PMAs may be designed. The number, location, type and orientation of the isolators and PMAs may be chosen to optimise the mass, volume, accommodation and cost constraints of any given use scenario (such as the intended mission). Once the vibration-mitigation system is designed and installed, based on the constraints and details of the arrangement into which the vibration-mitigation system will be used, the vibration-mitigation system may be tuned. The control tuning process is highly flexible. The control tuning process is flexible enough to adapt to different configurations and powerful enough to push the system to act at an extremely high performance level, balancing the various performance requirements for a specific disturbance profile (in use the disturbance from the disturbance source).

[0154] The active damping elements may contain magnetic material. This should be taken into account when mounting in the system.

[0155] Referring now to Figure 8, there is shown a payload 800 that is primarily sensitive around two rotational axes. In this example, the payload 800 may be a telescope 802 (for example) and has three axes, an X axis, a Y axis and a Z axis as shown. In this arrangement, use of two PMAs might be feasible provided that the induced translations are acceptable.

[0156] Specifically, a first PMA 810 is arranged to provide a translational effect in the Z axis and generates torque 812 around the X axis. A second PMA 820 is arranged to provide translation effect in the Z axis and generates torque 822 around the Y axis. Both PMAs 810, 820 generate forces along the Z-axis. This arrangement may be feasible provided the two rotations around the telescope axes are major contributors to the pointing performance of the telescope.

[0157] The rotational torques may equal each other out and no longer impact the payload or perhaps between them account for vibrations or movement in the telescope itself.

Claims

CLAIMS1. A system for attenuating disturbances, the system comprising: an interface; a proof mass actuator, PMA, arrangement arranged on the interface, and arranged to oscillate a mass in response to control signals; at least one passive filtering element, having a predetermined mass structurally coupled to the interface; and, control circuitry arranged to provide control signals to the PMA arrangement, wherein the control signals are arranged to control oscillations of a mass.

2. The system of claim 1, wherein the PMA arrangement is arranged to operate in the frequency range of from around 15 Hz to around 100 Hz.

3. The system of claim 1 or 2, wherein the at least one passive filtering element is arranged to provide passive filtering in the frequency range of from around 30 Hz to around above 1000 Hz.

4. The system of any of claims 1 to 3, wherein the at least one passive filtering element is an elastomer-based or metallic-based passive filtering element and wherein the at least one passive filtering element is arranged, in use, between the interface and a payload.

5. The system of any of claims 1 to 4, wherein the PMA arrangement comprises at least three PMAs, wherein the first PMA is arranged to oscillate on a first axis; the second PMA is arranged to oscillate on a second axis; and, the third PMA is arranged to oscillate on a third axis.

6. The system of any of claims 1 to 5, wherein the PMA arrangement is arranged to operate in response to a control signal from the control circuitry, wherein the control circuitry is arranged to use a payload response to the control signal to design the control signals.

7. The system of claim 6, wherein the payload response is measured by sensors.

8. The system of any of claims 1 to 7, wherein the control circuitry comprises a low power data processor and communication electronics.

9. The system of any of claims 1 to 8, wherein the PMA arrangement comprises at least six PMAs.

10. The system of any of claims 1 to 9, wherein the system provides 10 to 40 dB damping.

11. The system of claim 10, wherein the system provides around 10 dB damping in broad-band performance and around 40 dB damping in narrowband performance.

12. An arrangement for use in a spacecraft comprising the system of any of claims 1 to 11.

13. The arrangement of claim 12, further comprising a payload and a disturbance source, wherein the system is arranged functionally between the payload and the disturbance source.

14. The arrangement of claim 13, wherein the disturbance source is at least one of a reaction wheel and a cryo-cooler.

15. A method of damping disturbances comprising: providing a noise signal to an arrangement comprising a payload and a disturbance source and a system according to any of claims 1 to 11 ; detecting the noise signal by at least one sensor on the payload; receiving, by the control circuitry, a measured output signal from the at least one sensor; obtain a transfer function in a frequency domain relating to the input noise signal sent to the PMAs and the measured output by the sensors; configuring the proof mass actuator arrangement and the at least one passive filtering element to provide damping between the payload and the disturbance source.

Images