>»X<«

DESCRIPTION

The present invention refers to a method for detecting space debris and, more precisely, to a method for detecting the spatial position of space debris , which provides an algorithm for estimating the performance of a network of globally distributed optical sensors at several ground locations and equipped with standard components .

The subj ect of the present invention is also a system for implementing the method and the related operating algorithm .

State of the art

As it is known, a new space race has begun in recent years , driven by the incredible source of useful space resources and data available . The above lead into a rapid increase in the number of satellites put into orbit around the Earth every year, and in particular in the debris zone of low Earth orbit ( LEO) .

More satellites mean a greater probability of collisions and disasters for the environment in orbit ( as in the famous accident involving Cosmos and Iridium in 2009 , which generated thousands of new small orbiting fragments ) . A recent ESA report estimated the presence of 670 . 000 obj ects larger than 1 cm, and more than 1 . 000 . 000 obj ects larger than 1 mm .

For example, an orbital debris detection and tracking system using solar or lunar occlusion is known from US7105791B1 . According to this document , a system is described for detecting obj ects traveling through the Earth ' s atmosphere using the image of the sun . The system includes a receiver for collecting incident sunlight ( solar energy) and a light-sensitive device that produces a signal in response to light exposure . A signal processor is coupled to the light-sensitive device , the signal processor detects collected incident sunlight and is programmed to provide a corresponding output signal in order to provide a detection signal in response to a shadow moving across the light sensitive device .

However, one of the main problems is that most of these orbiting obj ects are neither catalogued nor observed . In addition, position measurements for a given satellite are usually not available in real time , since such position measurements depend on the visibil ity of the latter via the selected sensor as well as the accuracy of the sensor, both aforementioned factors influencing directly the precision of the satellites ' orbital predictions .

Therefore , the development of a distributed sensors network, shared between di f ferent countries around the world, would represent an important improvement in thi s regard, allowing for a signi ficantly improvement in monitoring of space obj ects orbiting the Earth . More precisely, an ideal network should be composed of multiple sensors , insensitive to any atmospheric or astronomical perturbation, such as the synthetic aperture of radars . Unfortunately, such radars are expensive , di f ficult to maintain, their installation requires several government permits , and operations must be manned by highly trained personnel .

On the other hand, optical technologies , although limited by weather and lighting conditions , are a cheaper solution to implement and maintain .

The state of the art on this topic is not very rich . The available literature mainly concerns the study of operational networks of existing telescopes , or the study of the performance of a single telescope in terms o f signal-to-noise ratio or other parameters . An example of this technology is provided by the document US2013211778A1 , which illustrates a method for creating a space detection system for the LEO areas of Earth orbit , which involves the positioning of a network stations of optical detection systems onto the surface of the globe and according to a grid designed to of fer an ef fective daily cycle of the system close to 24 hours and a revisitation period chosen for the observed LEO zone .

Thus , obj ect of the present invention is to provide a method and system for the performance estimation and optimi zation of a ground-based network of telescopes .

The present invention aims to achieve the aforementioned obj ect through the implementation of an algorithm for the providing of the best distribution at a global level of a given number of ground-based telescopes starting from a list of possible sites , and providing as a result the measurement of a performance of a such a global optical sensors network, with the aim of providing to the users of the relevant art the prediction of a possible collision between a spacecraft and any debris orbiting the Earth .

BRIEF DESCRIPTION OF THE INVENTION

Therefore , obj ect of the present invention is a method for detecting the spatial position of space debris , which provides an algorithm both for the estimating the performance of a network of ground optical sensors globally distributed at di f ferent Earth locations and equipped with standard components , and for the computing of the optimal positioning of a given number of optical sensors or telescopes for a global optical network, in terms of coverage of known obj ects in LEO ( Low Earth Orbit ) according to a list of possible sites and a given number of deployable optical sensors or telescopes with the same features , and wherein each optical sensor has an observation interval of 24 hours , and wherein at least three acquisitions of an obj ect are necessary to consider said acquisition acceptable , wherein the opening time of the sensor is determined, wherein all measurements are at least 10 degrees apart one from each other, which includes the estimating of the si ze of the obj ect to be detected, and which includes the veri fication of visibility conditions at the ground station, the method is characteri zed by the fact that it includes the following steps :

- Computing the geometrical and optical visibility of all space obj ects at all selected ground observation sites ;

- Identi fying the best overall configuration of ground observation sites using the data received from the preceding step ; and

Determining the best combination of ground observation sites available by identi fying the best possible locations considering all possible combinations .

Thus , the present invention provides a method for detecting the spatial position of space debris substantially according to the appended claims .

The invention also provides a system for implementing the method, and an operating algorithm for the system .

DETAILED DESCRIPTION OF THE INVENTION

A detailed description of a pre ferred embodiment of the method for detecting the spatial position of space debris of the present invention wi ll now be provided, given by way of a non limiting example , and making reference to the appended drawings , wherein :

Figure 1 is a schematic view of the flow chart of the algorithm that implements the method of the present invention;

Figure 2 is a schematic view of a part of the flow chart of the algorithm that implements the method of the present invention;

Figure 3 is a graph illustrating the percentage o f NORAD catalogue visibility for each ground station;

Figure 4 is a graph showing the percentage of detected obj ects grouped according to the s i ze thereof , and for both considered optical configurations ;

Figure 5 is a graph showing the percentage of coverage of a space obj ect as a function of the number of ground stations ; Figure 6 is a graph showing detection probability as a function of orbit altitude and debris si ze for a given scenario ;

Figure 7 is a graph showing the percentage of detected obj ects as a function of the si ze for di f ferent optical configurations of ground stations ;

Figure 8 is a graph showing the coverage achieved using 5 ground stations simultaneously and compared to the coverage achieved using all ground stations considered for a given scenario ; and

Figure 9 is a graph showing detection probability as a function of orbit altitude and debris si ze for a given scenario .

With reference now to figure 1 , the flow chart of the algorithm that implements the method of the present invention is illustrated .

According to the present invention, the method involves the following initial steps :

- Acquiring of a database relating to the catalogue of space traj ectories of obj ects orbiting the Earth ( satellites ) . The database is obtained from an already existing database called NORAD ( SATCAT ) . The catalogue contains a list of orbiting obj ects having a si ze equal to or greater than 10 cm .

- Acquiring of the database of the list of available ground observation sites ; and

- Optical setting of each observation site .

Once the data acquisition steps above indicated have been completed, the method involves the following two calculation steps .

The first calculation step relates to the calculation of the so-called " geometric visibility" .

It is important to highlight here that the condition of " geometric visibility" to be usable by the algorithm i s obtained i f the following conditions are satis fied :

1 ) The satellite and the ground station are in a " line of sight" ( LOS ) condition; 2) The elevation angle of the ground station to see the satellite is greater than a determined value (in the case of Italy, the angle must be 30 degrees) ;

3) The satellite is in LOS with the Sun, and the ground station is in a dim condition;

4) The trajectory of the satellite must have an arc longer than a determined value (in the case of Italy, it is three acquisitions of optical sensors separated by 10 degrees each) .

All objects fulfilling these features are considered geometrically visible. Optical visibility is calculated only for objects that are geometrically visible and it is expressed by the "target detection probability", which is based on the signal-to-noise ratio (SNR) .

The detection probability is the key parameter to evaluate the performance of the entire optical configuration, and the computing thereof is influenced by the object passing over an established location multiple times, and by the object as seen from multiple positions.

This step of the algorithm provides as input data: a) The entire satellite catalogue and its propagations (database of catalogue trajectories) ; b) All positions of the ground stations in WGS84 (latitude, longitude, altitude) and the magnitude thereof at the zenith; c) Sensor parameters: pixel pitch, pixel resolution, quantum efficiency, readout of sensor noise, disturbance noise, binning mode; d) Telescope parameters: f-number, aperture diameter, optical transmission, obscuration percentage.

Astrodynamics analysis

Since the optics of optical sensors arranged in ground stations are implemented, satellite visibility from a ground station requires certain conditions to be met, such as :

1) line of sight between satellite and ground sensor,

2) satellite illuminated by the Sun; and 3 ) ground station in darkness .

To achieve what has j ust been indicated, the calculations of the visibility windows are performed by three main software modules as illustrated in figure 2 , and described below :

Module 1 - Generate Filtered Catalogue

This module computes the traj ectories of a selected batch of satellites . Satellites are filtered based on orbit altitude and physical si ze . An approximate batch si ze i s approximately 10 thousand satellites .

Module 2 - Compute Batch Schedule MultiThread

This module calculates the visibility windows using the ground station parameters ( latitude , longitude , altitude , type ) and the related window limitations ( elevation threshold, arc length threshold) for each ground station available on the database ;

Module 3 - Convert To Study DataFrame

This module converts the processed data into a format suitable for optical performance algorithms . This is a data reduction process that keeps only the metrics relevant to the use thereof .

The results are then stored in an observation window DataFrame containing the following data :

- NORAD ID;

- Passing ID;

- Elevation angle ;

- Distance ;

- Phase angle ;

- Obj ect diameter ;

This data serve as input for the evaluation of optical performance .

Optical analysis

To evaluate the probability of detection it is necessary to obtain the signal -to-noise ratio of the system . To obtain a good evaluation of the signal-to-noise ratio , in this case it is necessary to consider the integration time as the time spent by the target in a single pixel .

The algorithm also considers whether a satellite makes multiple passes over the same or di f ferent ground station sites during the selected time window, and obviously this af fects the probability of detection of the single satellite by the network . This is done by considering that all events are independent and thereby composing the probabilities , and such that the probability that the obj ect has been observed ' at least once ' by the entire network is obtained .

Atmospheric transmittance

To accurately evaluate the influence of the atmosphere on optical performance , the present algorithm performs altitude-dependent atmospheric transmittance modelling .

To find the equation that simulates the behaviour of the zenithal transmittance as a function of the observer which depends on the altitude , a speci fic wavelengthdependent atmospheric transmittance ( obtained from the MODTRAN software ) has been considered, and the established equation was derived varying the altitude . Curve simulation is also implemented using MODTRAN ( a consolidated solution for this type of analysis ) .

Selected Configuration Features

To achieve the best performance in terms of minimum satellite si ze , it is crucial to choose the right optical asset to ensure the detecting of the maj ority of catalogued obj ects , and last but not least , such asset must be physically realistic .

To detect small obj ects (< 15 cm) it is necessary to use large apertures telescopes ( > 400 mm) . These types of telescopes have a very high fixed focal length value ( > 1000 mm) . The aperture and focal length are related parameters , i . e . a high aperture means a long focal length .

Acquiring measurements with larger aperture optics means that the received signal is stronger but with a very narrow instantaneous field of view ( I FOV) , which will result in a very short elapsed time of the target in the pixel , and so on in an exponential decrease in the signal and in the probability detection .

To evaluate the performance of each configuration, detection probabilities are calculated for the following conditions : a ) Elevation angle between 30 and 90 degrees ; b ) Target altitude between 200 and 2000 km; c ) Phase angle of 30 degrees

Starting from the above conditions , di f ferent configurations can be chosen .

Test examples

Following the settings above identi fied, the present inventors have performed some proof tests of the system o f the present invention .

Two configurations and their test results are provided below . A summary of the chosen parameters is listed in Table 1 below :

Table 1 On the other hand, in Table 2 (here below) the visibility altitude limit as a function of the size of the debris is shown and with reference to two chosen configurations. Accordingly, the visibility condition is considered satisfying if the probability of detection is greater than or equal to 20%.

Table 2 - Optical performances

Simulation and results

To define the network performances, two scenario simulations were performed.

For the first scenario, it was assumed to have ten different sites to position the observation assets, while for the second scenario it was assumed to have twenty-one observation sites available.

It is important to point out that each selected location is an existing land. The locations are chosen among the best magnitude levels of the zenith sky. The simulations were performed considering a filtered NORAD catalogue, where objects are retained only if they satisfy the following limitations when propagating for a 24-hour observation window: a) The altitude of the objects must be less than 2000 km; b) At least three acquisitions of the object can be carried out, and by considering the opening time of the optical sensor; c) Acquisitions are considered feasible and completed if the measurements have an angular separation of at least 10 degrees one from each other; d) Only objects of known dimensions are considered. The SATCAT catalogue (Celestrack) was used to obtain this data.

The assumptions made for the algorithm were as follows :

1) The object will always be centred in the field of view (FOV) and assumed to be spherical;

2) Visibility with clear skies (> 50 km) and very dark background (> 20 May) ; and

3) Telescope height > 30 degrees.

To obtain a more accurate representation of the network's performance, a week time average of the algorithm results was performed, considering that each day has its own results.

The sites considered possible were taken from Table 3 here below.

For each scenario, the number of available telescopes to be positioned ranges from 1 to the maximum number of available sites.

For each site, the same telescope configuration was assumed .

Table 3 The performance of each site is expressed in Figure 3 , which indicates the percentage of visibility of the NORAD catalogue for each ground station .

It may be noted that some locations are much more ef ficient than others . This depends on the coupling between the latitude of the site and the time of year considered for the simulation, which af fects the elevation angle of the observations and so on the atmospheric transmittance ( some latitudes may give better angles than others for LEO ob j ects ) .

It is also worth to be noted that even i f a ground station detects a lower percentage of the catalogue , it may detect obj ects that are completely di f ferent from those detected by a ground station with a higher percentage , i . e . in a configuration where both s tations could compensate both to each other , and the algorithm consider this condition .

Last but not least , it should be noted that one ground station is not giving results at all and has 0% coverage ; the ground station is in Norway . This is because during the simulation period this ground station is constantly under sunlight due to the particular conditions of these high latitudes . This is another aspect that the algorithm takes into consideration when evaluating the performance of the entire network .

Scenario 1

Figure 4 shows the percentage of detected obj ect clustered according to the si ze thereof , and for both optical configurations considered .

As can be seen, the highest percentage of observable obj ects are those larger than 10 cm, all being equal to or greater than 60% , while for smaller obj ects the value drops signi ficantly .

Figure 5 shows the percentage of space obj ect coverage as a function of the number of ground stations . As the graph shows , the increase stops after that five ground stations are operated simultaneously, with a constant value around 54%.

Figure 6 shows the detection probability as a function of orbit altitude and debris size for scenario 1. As can be seen, there is a massive concentration of small objects in the 600-800 km altitude layer having a very low detection.

The huge difference between this layer and the higher layers above is due to the very high speed of the objects of around 7.5 km/ s .

Scenario 2

In this scenario, new ground stations have been added compared to the preceding scenario, distributing them over a wider band in terms of latitude. As Figure 3 illustrates, the ground stations with the highest visibility (above 20%) are located at approximately the same latitude. One thing to note is the percentage of coverage of grst-16-NOR in Norway for this test time window, which has zero visibility and therefore does not play a role in this specific simulation. This is due to the fact that the time windows considered for this simulation are in spring/ summer , and in countries with higher latitudes the period of light is so long that optical visibility is not guaranteed.

Figure 7 shows the percentage of objects detected as a function of size for both optical configurations. When compared to the preceding results, an overall increase in detection probability can be seen for all object sizes.

A relevant result is seen for the smallest objects: with configuration 1 it is possible to detect objects smaller than 5 cm. This opens a new window for the cataloguing a new set of space micro-debris that turn out to be the most dangerous in LEO orbit.

As shown in figure 8, the coverage achieved by using 5 ground stations simultaneously is higher than that obtained by using all the ground stations, considered for scenario 1. An almost constant coverage value of around 74% (20% more than the performance of scenario 1) is achieved when at least 10 ground stations are managed simultaneously.

Figure 9 shows the probability of detection as a function of orbit altitude and debris size for scenario 2. Comparing this result with that of the preceding scenario, a significant improvement can be seen.

Despite a considerable number of objects with low probability values, the average probability value of being detected increases by approximately 70%.