Patent Application
20200292699
The field of the invention is the field of visualization and use of data from sonar signals scattered from sparse objects immersed in a fluid.
The following US Patents and US patent applications are related to the present application: U.S. Pat. No. 6,438,071 issued to Hansen, et al. on August 20; U.S. Pat. No. 7,466,628 issued to Hansen on Dec. 16, 2008; U.S. Pat. No. 7,489,592 issued Feb. 10, 2009 to Hansen; U.S. Pat. No. 8,059,486 issued to Sloss on Nov. 15, 2011; U.S. Pat. No. 7,898,902 issued to Sloss on Mar. 1, 2011; U.S. Pat. No. 8,854,920 issued to Sloss on Oct. 7, 2014; and U.S. Pat. No. 9,019,795 issued to Sloss on Apr. 28, 2015; U.S. patent application Ser. Nos. 14/927,748 and 14/927,730 filed on Oct. 30, 2015, Ser. No. 15/978,386 filed on May 14, 2018, Ser. No. 15/908,395 filed on Feb. 28, 2018, and Ser. No. 15/953,423 filed on Apr. 14, 2018 by Sloss are also related to the present application. The above identified patents and patent applications are assigned to the assignee of the present invention and are incorporated herein by reference in their entirety including incorporated material.
It is an object of the invention to improve the tracking of unknown possible objects using sonar imaging. It is an object of the invention to measure and record the positions and orientations of possible objects with unknown shapes. It is an object of the invention measure the increase or decrease in the probability that a possible object is a real object as the possible object is tracked. It is the object of the invention to refine the measurements of the shapes of unknown objects over time while the objects are tracked.
One or more large arrays of sonar detectors are used to produce three dimensional sonar images possible unknown objects. A series of sonar pings are sent into an insonified volume of water and the reflected or scattered sonar pings are analyzed to produce a 3 dimensional map of points which have scattered the sonar ping. The points are segregated by one or more techniques, and one of a defined set of shapes is assigned to each segregated set of points. Each of the defined set of shapes has an associated defined unique 3 dimensional position and possibly a defined spacial orientation. For each ping, the defined position is recorded and tracked from ping to ping to provide a record of the track of the defined shape.
It has long been known that data presented in visual form is much better understood by humans than data presented in the form of tables, charts, text, etc. However, even data presented visually as bar graphs, line graphs, maps, or topographic maps requires experience and training to interpret them. Humans can, however, immediately recognize and understand patterns in visual images which would be difficult for even the best and fastest computers to pick out. Much effort has thus been spent in turning data into images.
In particular, images which are generated from data which are not related to light are often difficult to produce and often require skill to interpret. One such type of data is sonar data, wherein a sonar signal pulse is sent out from a generator into a volume of fluid, and reflected sound energy from objects in the insonified volume is recorded by one or more detector elements. The term “insonified volume” is known to one of skill in the art and is defined herein as being a volume of fluid through which sound waves are directed. In the present invention, a sonar signal pulse of sound waves called a ping is sent out from one or more sonar ping generators, each of which insonifies a roughly conical volume of fluid.
The field of underwater sonar imaging is different from the fields of medical ultrasonic imaging and imaging of underground rock formations because there are far fewer surfaces in the underwater insonified volume.
The sonar generators 15 may be attached to or in a known location in proximity to one or more of the vessels 10, or to one or more mobile underwater probes 19. Sound waves 18 are shown reflected from possible objects 11, 12 and real objects 17 back towards the one or more multielement sonar detectors. Objects and possible objects may be suspended in the water, lying on the seabed 12, or be buried in the sea bed 12. Objects 17 may be both real and known. A known object is an object with a known shape which has been previously identified and imaged, and can be recognized by the sonar imaging system or a skilled operator. Objects 11 are not immediately recognized as real, and may be spurious collections of reflections and re-reflections. Objects 12 may be one object or two. None of these possible objects can be classed as real immediately. They are affectionately known as “blobs”.
However, all segmented possible objects must be treated as if they are real for the purposes of tracking.
The method of the invention starts out be choosing a defined known shape from a set of defined shapes to assign to each of the blobs. In the preferred method of the invention, the simplest mathematical shape which can be used is a sphere of radius r. We assign a position to each of the defined shapes in order to locate the defined shape.
For the sphere, we choose the center of the sphere as the assigned position for the spherical shape. In general, we calculate the center of mass for each shape as if the assigned shape were filled uniformly with material of constant density. We then choose the position of the center of mass as the position of the shape for the purposes of tracking.
We call the center of mass point a centroid. The next simplest shape is an ellipsoid. having a centroid position determined as the point halfway along the major axis. The orientation of the ellipsoid, in contrast to a sphere, can be recorded as the angular coordinates of the major axis. More complicated shapes may need two angular coordinates as well as three spatial coordinates.
Any number of defined shapes can be used, such as cylinders with square ends, cylinders with half spherical end caps, partial cones, etc. Each defined shape has its unique point defined. We call the assigned point the centroid of the shape.
The method of the invention uses a series of pings to assign a point for each blob for each ping.
The points are recorded, and can be plotted in 3 dimensional space as a function of time for tracking purposes. As more data is gathered, the shapes may become better defined, and a new shape chosen to better follow and track the object. Some of the objects may even coalesce into one shape, or one shape will break into two or more shapes depending on such things as the sonar reflectivity over the surface of the shape. As the shapes become better known, centroids can be calculated for them also without using the defined set of shapes.
Points may be associated with each shape other than the calculated center of mass. The most preferred method of the invention is to use the centroid of the defined shape.
The ultrasonic multielement sonar detectors measure the phase, intensity, and arrival time of the reflected sonar pings 18. The phase, intensity, and arrival time data are processed to provide three dimensional location data measuring sonar reflecting surface locations. The seabed 12 surface and the object 17 surface may be similarly measured to give three dimensional location data of the reflecting surfaces.
A series of outgoing ping pulses may be sent out with an outgoing ping frequency Pf. A sonar ping generally has a constant sound frequencyf. (The frequencyf is sometimes changed in the prior art during the ping in a method called a chirped pulse ping, where the pulse frequency either increases or decreases monotonically throughout the pulse.) A master oscillator (not shown) produces a square wave voltage output at frequency f, and the ping generator uses the master oscillator to produce outgoing sinusoidal sound waves in phase with the master oscillator.
The reflected sound waves 18 are received by each detector element one or more of the large multielement sonar detector arrays associated with each ping generator 15. The detector arrays measure the pressure vs time of the reflected ping sound waves at each element and return an analog electrical voltage signal representing the amplitude versus time of the sound wave impinging on the element. The electrical voltage signals are digitally sampled at precisely known times with respect to the phase of the sent out sound waves of each ping. A large array multielement detector is preferably constructed with 24 by 24 or more sonar detector elements arranged orthogonally as a square grid. A two dimensional sonar detector array which has m by n elements, where m and n are different integers will have different angular resolutions in two orthogonal angles.
The attenuation of the sent out and received sonar signals is dependent on the sent out frequency Pf. As the frequency Pf increases, the sonar resolution increases and the detection range decreases. The frequency Pf may be changed from ping to ping to either see further at the expense of resolution, or to see more detail of the closer sound reflecting objects. A skilled operator is needed for manual control of the ping frequency Pf, or an computer programmed to change frequency Pf according to a criterion, such as the need for higher resolution or greater range
The amount of raw digital data generated by large array sonar detectors is often too great either to transmit to the surface vessel from the array detector or to store for later analysis. This is especially true for independently operated probes without high speed data connections to the vessel 10. In these cases, the raw data must be analyzed close to the detector, so that signals representing the various tracks may be sent to the control vessel may be sent by low bandwidth means such as sound waves.
Independently operated ROV (remotely operated vehicles) are especially suited for using the method of the invention. They may be small enough to be less easily detected than manned vehicles. Such remotely operated vehicles must be programmed to make many decisions according to various criterion.
For example, the decision to switch from a spherical shape to an ellipsoidal shape could use the criterion that the raw data points could be contained in a ellipsoid with major to minor axis ratios greater than 1.5. Other such criterion are noted in Appendix I.
Flow charts of the invention are shown in Appendix I for manually and automatically carrying out the method of the invention. A skilled operator is usually carried by the cable laying vessel 10. The algorithms automatically distinguishes/segments signals from the sea-bed. Using sonar data describing More flowcharts describing the invention are described in Appendix I
Obviously, many modifications and variations of the present invention are possible in light of the above teachings. It is therefore to be understood that, within the scope of the appended claims, the invention may be practiced otherwise than as specifically described.
| Number | Date | Country | |
|---|---|---|---|
| 62818682 | Mar 2019 | US |
