Spin measurement and estimation method

Abstract

An exemplary method for determining a spin rate of an object may include receiving a radar signal of a particular object in motion. The method may further include converting the radar signal into an input vector. The method may also include providing the input vector as an input to a neural network. The neural network may include accessing an initial data set generated based on a plurality of initial radar signals of a plurality of objects in motion at an initial time. The method may further include determining a spin rate of the particular object in motion based on an analysis performed by the neural network of the input vector including time-frequency information of the particular object in motion in view of the initial data set. The analysis may include comparing one or more elements of the input vector to one or more elements of the initial data set.

Description

[Technical Field]

[0001] The embodiments described in this disclosure relate to methods for measuring and estimating spin. [Background technology]

[0002] Objects in motion often have an associated spin rate, among other motion attributes. Determining this spin rate may be desirable because it may contribute to understanding the object's trajectory, the object's launch conditions, and / or other factors related to the object's motion. Determining an object's spin rate under certain conditions may be achieved using multiple, costly sensors. Additionally, costly sensors may be limited in availability, both of which may limit the options and / or opportunities for determining an object's spin rate.

[0003] The subject matter claimed in this disclosure is not limited to embodiments that solve any shortcomings, nor does it operate only in environments such as those described above. Rather, this background is only provided to describe one example technology area where some embodiments described in this disclosure may be practiced. Summary of the Invention

[0004] In one embodiment, a method for determining the spin rate of an object may include receiving a radar signal of a particular object in motion. The method may further include converting the radar signal into an input vector. The input vector may include time-frequency information of the particular object in motion. The method also includes providing the input vector as an input to a neural network. The neural network may include accessing an initial data set generated based on multiple initial radar signals of multiple initial moving objects. The method may further include determining the spin rate of the particular object in motion based on an analysis performed by the neural network of the input vector including the time-frequency information of the particular object in motion, taking into account the initial data set. The analysis may include comparing one or more elements of the input vector to one or more elements of the initial data set.

[0005] The object and advantages of the embodiments will be realized and achieved at least by the elements, features, and combinations particularly pointed out in the claims.

[0006] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed. [Brief explanation of the drawings]

[0007] Exemplary embodiments are described in detail using the specification and accompanying drawings.

[0008] [Figure 1] FIG. 1 is a block diagram of an example environment including spin measurement and estimation. [Figure 2] FIG. 2 is a block diagram illustrating an example system for determining an initial data set. [Figure 3] FIG. 3 shows a time-frequency image of an object in motion. [Figure 4] FIG. 4 is a flowchart illustrating an example of a method for measuring and estimating spin. [Figure 5] FIG. 5 is a flow chart illustrating an example of a method for obtaining an initial data set. [Figure 6] FIG. 6 illustrates an example of a system that can be used to measure and estimate spin. DETAILED DESCRIPTION OF THE INVENTION

[0009] Deep learning, such as using a neural network, can be employed to determine the spin rate of a moving object using the modified radar signal of the object. In some situations, the neural network can be trained using multiple data sets of known spin rates and radar signals. The neural network can use data from the training data set to analyze the radar signal from the moving object and determine its spin rate using data from the training data set.

[0010] In some circumstances, the neural network may be configured to identify portions of the image based on the radar signal and compare them to similar portions of the initial time-frequency image having known spin rates. The neural network may be configured to determine the spin rate of the moving object based on an analysis between the time-frequency images.

[0011] 1 shows a block diagram of an example environment 100 including spin measurement and estimation according to at least one embodiment described in this disclosure. The environment 100 may include a moving object 105, a radar device 110, an output radar signal 116, a reflected radar signal 118, a computing system 120, and data storage 130. The radar device 110 may include a processor 112 and a communication device 114.

[0012] In some embodiments, the radar device 110 can emit an output radar signal 116 that may be intended to be reflected from a moving object 105. In some embodiments, the output radar signal 116 may include a radio frequency output from the radar device 110. In some embodiments, the output radar signal 116 may be configured to be continuously emitted from the radar device 110 when the radar device 110 is powered on. Alternatively or additionally, the output radar signal 116 can be emitted from the radar device 110 upon receiving a trigger input. For example, the output radar signal 116 may begin emitting from the radar device 110 upon detection of motion by the radar device 110. Alternatively or additionally, the trigger input to the radar device 110 may include a user instruction to begin emitting the output radar signal 116.

[0013] In response to output radar signal 116 reflecting off moving object 105, radar device 110 can receive reflected radar signal 118 therefrom. In some embodiments, reflected radar signal 118 may include output radar signal 116 after contacting an object. For example, radar device 110 may emit a signal, such as output radar signal 116, which may reflect off an object, such as moving object 105, and return to radar device 110 as reflected radar signal 118.

[0014] In some embodiments, radar device 110 may include processor 112. In some embodiments, processor 112 may include any computing device, such as system 600 of FIG. 6. In some embodiments, processor 112 may be configured to convert reflected radar signal 118 into an input vector. For example, processor 112 may apply Fourier analysis to reflected radar signal 118 to generate the input vector from reflected radar signal 118. The conversion of reflected radar signal 118 into an input vector is further described with reference to FIG. 3.

[0015] Alternatively or additionally, the communication device 114 of the radar device 110 may be configured to transmit the reflected radar signal 118 to a remote device, such as the computing system 120. The remote device may be configured to convert the reflected radar signal 118 into an input vector. In response to receiving the reflected radar signal 118, the remote device, such as the computing system 120, may be configured to convert the reflected radar signal 118 into an input vector using the same or similar process as the processor 112 of the radar device 110. The conversion of the reflected radar signal 118 into an input vector is further described in connection with FIG. 3 .

[0016] In some embodiments, the moving object may include a ball, such as a sports ball, including a golf ball, a baseball, and / or other sports-related ball, and / or other object that has a spin rate and may be capable of being tracked.

[0017] In some embodiments, the input vector may include time-frequency information related to a moving object. For example, the input vector may include an image including time and frequency components related to a moving object, such as time-frequency image 300 of FIG. 3 . In some embodiments, the input vector including the time-frequency image may be generated by applying Fourier analysis to reflected radar signal 118. Computational system 120 may then obtain an input vector that can be used in analysis to determine the spin rate of the moving object. In some embodiments, the input vector may include one or more components that may be related to the velocity of the moving object and / or the spin rate of the moving object. For example, the input vector may include a main lobe and one or more side lobes that may be related to the velocity and spin rate of the moving object, respectively. Additional details regarding the input vector as a time-frequency image including the main lobe and its side lobes are described in connection with FIG. 3 .

[0018] In some embodiments, data storage 130 may be configured to store an initial data set. The initial data set may include multiple radar signals, each of which may be associated with multiple moving objects. For example, for any given moving object recorded by the radar device, an associated initial radar signal and / or initial input vector may be stored in data storage 130. Alternatively or additionally, each initial data of the initial data set may include an initial input vector that may be generated from received radar signals associated with an initial moving object, which may be obtained using a process similar to that described above for the input vectors. For example, the initial data set may include one or more initial input vectors associated with one or more moving objects, each of which may include various spin rates and / or other motion components.

[0019] In these and other embodiments, each initial data in data storage 130 may include a spin rate associated with an initial input vector. For example, a first object in motion may include a first received radar signal that may be converted into a first input vector. Additionally, the first object in motion may include a first known spin rate, and this first known spin rate and first input vector may be stored in data storage 130. In these and other embodiments, data storage 130 may include multiple iterations of the initial input vector and initial known spin rate of the initial object in motion, which may be used to train computing system 120.

[0020] In some embodiments, elements stored in data storage 130 may be arranged and configured to be retrievable upon request. Alternatively or additionally, elements in data storage 130 may be provided as input to a computing system, such as computing system 120, and may be used to train neural networks and / or other machine learning systems.

[0021] In some embodiments, computing system 120 may be communicatively coupled to radar device 110 and / or data storage 130. For example, computing system 120 may be configured to receive radar signals from radar device 110 and / or to receive and / or store data in data storage 130.

[0022] In some embodiments, computing system 120 may include a neural network. For example, computing system 120 may include a convolutional neural network, a recurrent neural network, a long short-term memory, and / or other deep learning network system. In some embodiments, computing system 120 may be configured to access elements of data storage 130 to train computing system 120 to determine the spin rates of moving objects. For example, computing system 120 may be trained using elements of data storage 130 that include radar signals of moving objects associated with known spin rates of each moving object.

[0023] In some embodiments, computing system 120 may be configured to perform an analysis of an input vector of object 105 in motion to determine and / or estimate its spin rate. In some embodiments, the analysis performed by computing system 120 may include comparing a portion of the input vector to a portion of initial data in data storage 130 on which computing system 120 was trained. In some embodiments, computing system 120 may determine and / or estimate the spin rate of object 105 in motion based on the results of the comparison of the input vector to the initial data in data storage 130.

[0024] In some embodiments, computing system 120 may determine one or more aspects of the input vector and / or use initial data in data storage 130 to compare the input vector to the initial data, which may enable computing system 120 to determine and / or estimate the spin rate of moving object 105. For example, computing system 120 may compare the number of harmonics associated with side lobes around the main lobe, the amount of spacing between the harmonics of the side lobes, the rate at which the strength of the side lobes decays over time, and / or other identifiable aspects associated with the input vector and / or the initial data in data storage 130.

[0025] In these and other embodiments, computing system 120 may be configured to determine and / or estimate a spin rate associated with object 105 in motion based on a comparison of the input vector (e.g., as an image) with an initial input vector in the initial data in data storage 130. For example, in instances where computing system 120 identifies an initial input vector having the same or nearly the same image characteristics (e.g., number of harmonics, spacing between harmonics, etc.) as the input vector, computing system 120 may use the initial known spin rate associated with the initial input vector to determine the spin rate of object 105 in motion. Alternatively or additionally, if the initial input vector is not identical or nearly identical to the input vector, computing system 120 may extrapolate between two or more initial input vectors that are similar to the input vector. Computing system 120 may estimate the spin rate of object 105 in motion using the extrapolated spin rate between two or more initial known spin rates associated with the two or more initial input vectors.

[0026] In some embodiments, upon determining and / or estimating the spin rate of the object 105 in motion, the computing system 120 may be configured to include the input vector and associated spin rate in the initial data in the data storage 130.

[0027] Modifications, additions, or omissions may be made to environment 100 without departing from the scope of the present disclosure. For example, in some embodiments, computing system 120 may be configured to communicate the determined spin rate of the object in motion to a user device for display. The user device may include a mobile device, a desktop computer, a tablet computer, and / or other user device. In some embodiments, the user device may include a graphical user interface (GUI) that may be configured to display the determined spin rate received from computing system 120. In some embodiments, computing system 120 may include a wired connection with the user device, which may include PCI, PCI Express, Ethernet, etc. Alternatively or additionally, computing system 120 may be configured to wirelessly communicate with the user device via Bluetooth, Wi-Fi, WiMAX, cellular communications, etc.

[0028] In some embodiments, the determined spin rate displayed on the user device may include a numerical value, a graph of the determined spin rate over time, a chart of the determined spin rate over periodic time intervals, and / or other visual display suitable for illustrating the determined spin rate.

[0029] Alternatively or additionally, computing system 120 may be configured to generate a visual model of the trajectory of the moving object. In some embodiments, the visual model may include a predicted trajectory of the moving object based on the determined spin rate. Alternatively or additionally, the visual model may include the velocity of the moving object, the launch angle of the moving object, and / or other factors related to the moving object. In some embodiments, computing system 120 may be configured to transmit the visual model to the user device, such as via the same or similar methods as described above with respect to displaying the determined spin rate on the user device. For example, computing system 120 may determine the spin rate of the moving object, as well as the path, velocity, apex, endpoint, etc., and may combine some or all of these to generate the visual model.

[0030] Alternatively or additionally, computing system 120 may be configured to transmit elements of the object in motion, including the determined spin rate, velocity, etc., to a user device, which may be configured to generate a visual model of the object in motion. In these and other embodiments, the visual model may include interactive elements that allow a user to rotate, zoom, pause, speed up, slow down, and / or otherwise interact with the visual model displayed on the user device. Additionally, the interactive elements presented to the user may allow the user to restrict portions of the visual model from being displayed. For example, a user may wish to remove a velocity portion associated with a moving object and may make a selection in the GUI to remove the display of the velocity portion of the visual model.

[0031] In some embodiments, environment 100 may be implemented using different types of objects. For example, elements of data storage 130 may include objects of a first type, such as golf balls, and an input vector may be generated from objects of a second type, such as baseballs. In some embodiments, computing system 120 may be trained using elements of data storage 130 such that computing system 120 may be configured to determine the spin rate of an input vector regardless of whether the object type of the elements of data storage 130 is similar to the object type of the input vector. Alternatively or additionally, elements of data storage 130 may be categorized into object types such that when computing system 120 receives an input vector associated with a first object type, computing system 120 performs spin rate analysis thereon using elements of data storage 130 that match the first object type.

[0032] 2 illustrates a block diagram of an example system 200 for determining an initial data set, in accordance with at least one embodiment described in this disclosure. The system 200 may include an object 205, a radar device 210, a computing module 220, and an initial data set 230.

[0033] In some embodiments, radar device 210 may be the same as or similar to radar device 110 of FIG. 1 . In these and other embodiments, radar device 210 may be configured to output radar signals and receive corresponding reflected radar signals from a moving object, such as object 205. In some embodiments, multiple radar devices may be included in radar device 210. For example, two or more radar devices may be configured to output and receive radar signals associated with a moving object. Alternatively, or additionally, the multiple radar devices of radar device 210 may be configured to output and receive radar signals associated with multiple moving objects. For example, the multiple radar devices may be configured to track one moving object, a number of moving objects equal to the number of the multiple radar devices, and / or any combination of any number of radar devices configured to track any number of moving objects. In these and other embodiments, each received radar signal from radar device 210 may be converted into an initial input vector.

[0034] In some embodiments, radar device 210 may be configured to convert received radar signals (e.g., reflected from a moving object) into an initial input vector. The initial input vector may include time-frequency information related to the moving object. For example, the initial input vector may be obtained by Fourier analysis of the received radar signals, which may include a representation of the moving object as frequency over time. Alternatively or additionally, radar device 210 may be configured to transmit the received radar signals to another device, such as computing module 220, which may be configured to convert the received radar signals into an initial input vector.

[0035] In some embodiments, the initial input vector may comprise an image, for example, a time-frequency image that represents an object in motion and its associated properties as an image of a frequency that varies over time.

[0036] In some embodiments, object 205 may include a ball, such as a sports ball. For example, object 205 may include a golf ball, a baseball, a volleyball, and / or various other sports balls. Alternatively or additionally, object 205 may include any non-ball object, so long as the object can be tracked by radar device 210 and may include a spin rate.

[0037] In some embodiments, computing module 220 may be the same as or similar to computing system 120 of Figure 1. Alternatively or additionally, computing module 220 may include any computing device, such as system 600 of Figure 6. In some embodiments, computing module 220 may be communicatively coupled to radar device 210 and / or initial data set 230.

[0038] In some embodiments, computing module 220 may obtain an initial input vector from radar device 210. Alternatively or additionally, computing module 220 may be configured to obtain a received radar signal, and computing module 220 may be configured to generate an initial input vector from the received radar signal. In these and other embodiments, computing module 220 may be configured to determine a spin rate from the initial input vector, and this spin rate may be stored as one of known spin rate data. The one of known spin rate data may include the determined spin rate associated with a time-frequency image as obtained from the initial input vector. In some embodiments, the determined spin rate may be obtained from computing module 220 processing the initial input vector, which may include converting time-series data from the radar signal into a frequency-domain signal using a Fourier transform and combining the time-series data and the frequency-domain signal into a time-frequency image. In some embodiments, computing module 220 may be configured to transfer the one of known spin rate data to initial data set 230.

[0039] In some embodiments, initial dataset 230 may be configured to store output from computing module 220. In some embodiments, initial dataset 230 may be the same as or similar to data storage 130 of Figure 1. For example, initial dataset 230 may include a plurality of known spin rate data entries that may be used to train a neural network, such as computing system 120 of Figure 1.

[0040] 3 illustrates a time-frequency image 300 (hereinafter referred to as "TFI 300") of an object in motion, according to at least one embodiment described in this disclosure. The TFI 300 may include a frequency axis 310, a time axis 320, a main lobe 330, and side lobes 340.

[0041] In some embodiments, the TFI 300 may be generated from a radar signal of a moving object, such as the moving object 105 of Figure 1, such as the reflected radar signal 118 of Figure 1, using a Fourier analysis performed thereon. For example, the radar signal of the moving object may be obtained from a radar device, such as the radar device 110 of Figure 1, and the radar signal may be converted to the TFI 300 using Fourier analysis.

[0042] In some embodiments, the TFI 300 may be generated iteratively in discrete elements from the radar signal. For example, a first portion of the radar signal may be converted to a first time and frequency element by Fourier analysis. After the first portion, a second portion of the radar signal may be converted to a second time and frequency element by Fourier analysis, etc. The first portion may be illustrated as a frequency element of an object moving during a first time, the second portion may be illustrated as a frequency element of an object moving during a second time, etc. The first portion, second portion, and all subsequent portions may be assembled into the TFI 300, which may be an input vector that may be input to a computing device, such as computing system 120 of FIG. 1 . Alternatively or additionally, the TFI 300 may be generated from continuously converting the radar signal into time and frequency elements and assembling the frequency elements into an image showing the frequency elements over time.

[0043] Alternatively or additionally, the TFI 300 may be generated by any radar processing algorithm that may be configured to convert time series data based on radar signals into a time-frequency image.

[0044] In some embodiments, the TFI 300 can illustrate a moving object as frequency over time. For example, a moving object may be represented as a linear frequency over a frequency axis 310 and a time axis 320. The linear frequency of the moving object may include a main lobe 330 and / or side lobes 340.

[0045] In some embodiments, the main lobe 330 may include more energy than the side lobes 340, which may include less energy than the main lobe 330. For example, the main lobe 330 may be illustrated as including a wider frequency band than any one of the side lobes 340, which may exhibit more energy than any one of the side lobes 340. Alternatively or additionally, the intensity of the main lobe 330 in the TFI 300 may be illustrated as being greater than the intensity of any one of the side lobes 340.

[0046] In some embodiments, the side lobes 340 may be symmetrically positioned around the main lobe 330. For example, given a main lobe 330 in the TFI 300, the side lobes 340 on either side of the main lobe 330 may be equal in number and spacing relative to the main lobe 330. Alternatively or additionally, the side lobes 340 may include asymmetric elements relative to the main lobe 330, such as more side lobes 340 on one side of the main lobe 330 than on the other side. In these and other embodiments, the side lobes 340 may include harmonic frequencies centered about the main lobe 330.

[0047] In some embodiments, the main lobe 330 of the TFI 300 can represent the velocity of a moving object, converted into frequency and time. Alternatively or additionally, the side lobes 340 of the TFI 300 can represent the spin rate of a moving object, converted into frequency and time. In some embodiments, the main lobe 330 can be substantially horizontal with respect to the time axis 320, which can indicate a substantially similar frequency over time. Alternatively or additionally, the main lobe 330 and / or the side lobes 340 can include time-varying frequencies and / or intensities that can be related to changes in object velocity and / or spin rate over time. For example, as illustrated in the TFI 300, the main lobe 330 includes a frequency that decreases and an intensity that decreases over time, which can be related to a decrease in the velocity of the object of the TFI 300 over time.

[0048] 4 shows a flowchart of an example method 400 of spin measurement and estimation in accordance with at least one embodiment described in this disclosure. Method 400 may be implemented in accordance with at least one embodiment described in this disclosure.

[0049] The methods may be performed by processing logic that may include hardware (circuitry, dedicated logic, etc.), software (such as may be executed on a general-purpose computer system or a dedicated machine), or a combination of both, and may be included in computing system 120 of Figure 1, computing module 220 of Figure 2, or another computer system or device. However, another system or combination of systems may also be used to perform the methods.

[0050] For ease of explanation, the methods described herein are depicted and described as a series of acts. However, acts in accordance with the present disclosure may occur in various orders and / or simultaneously, and with other acts not shown and described herein. Moreover, not all illustrated acts may be used to implement a method in accordance with the disclosed subject matter. In addition, those skilled in the art will understand and appreciate that the method may alternatively be represented as a series of interrelated states via a state diagram or events. Furthermore, the methods disclosed herein may be stored on an article of manufacture, such as a non-transitory computer-readable medium, to facilitate transport and transfer of such methodologies to a computing device. The term article of manufacture, as used herein, is intended to encompass a computer program accessible from any computer-readable device or storage medium. While depicted as individual blocks, various blocks may be divided into additional blocks, combined into fewer blocks, or eliminated, depending on the desired implementation.

[0051] The method 400 may begin at block 402 where processing logic receives a radar signal of a particular object in motion.

[0052] At block 404, processing logic converts the radar signal into an input vector. The input vector may include time-frequency information of a particular object in motion. In some embodiments, the time-frequency information of the input vector may include a time-frequency image. In some embodiments, the time-frequency image may be obtained by performing a Fourier analysis on the radar signal.

[0053] In some embodiments, the time-frequency image may include a main lobe and one or more side lobes modulated symmetrically about the main lobe. In some embodiments, the main lobe may include a high-energy portion of the input vector relative to a low-energy portion of the input vector associated with one or more side lobes. In some embodiments, the main lobe may be illustrated in the time-frequency image as a substantially linear frequency that varies with time. In some embodiments, the main lobe of the time-frequency image may indicate the velocity of a particular object in motion, and one or more side lobes of the time-frequency image may indicate the spin rate of a particular object in motion.

[0054] At block 406, processing logic may provide an input vector as an input to the neural network. The neural network may include access to an initial data set generated based on a plurality of initial radar signals of a plurality of initial moving objects. In some embodiments, a first piece of data in the initial data set may include a first spin rate of the initial moving object associated with first time-frequency information of the initial moving object. In some embodiments, the particular moving object may be a first type of object, and the initial plurality of moving objects may include a second type of object.

[0055] At block 408, processing logic may determine the spin rate of the particular object in motion based on an analysis performed by the neural network of an input vector containing time-frequency information of the particular object in motion, taking into account the initial data set. The analysis may include comparing one or more elements of the input vector to one or more elements of the initial data set. In some embodiments, the analysis performed by the neural network may include comparing a first set of harmonics of the input vector to a second set of harmonics of the time-frequency image in the initial data set.

[0056] It is understood that for this and other processes, operations, and methods disclosed herein, the functions and / or operations performed may be performed in differing order. Furthermore, the outlined functions and operations are provided only as examples, and some of the functions and operations may be optional, combined into fewer functions and operations, or expanded into additional functions and operations without detracting from the essence of the disclosed embodiments.

[0057] For example, in some embodiments, method 400 may further include displaying a visualization of the spin rate. Alternatively or additionally, method 400 may further include displaying a simulated visual model of the trajectory of the particular object.

[0058] 5 shows a flowchart of an example method 500 for obtaining an initial data set, in accordance with at least one embodiment described in the present disclosure. The method 500 may be prepared in accordance with at least one embodiment described in the present disclosure.

[0059] Method 500 may be performed by processing logic that may include hardware (circuitry, dedicated logic, etc.), software (such as is executed on a general-purpose computer system or a dedicated machine), or a combination of both, and this processing logic may be included in computing system 120 of Figure 1, computing module 220 of Figure 2, or other computer system or device. However, other systems or combinations of systems may be used to perform method 500.

[0060] For ease of explanation, the methods described herein are depicted and described as a series of acts. However, acts in accordance with the present disclosure may occur in various orders and / or simultaneously, and with other acts not shown and described herein. Moreover, not all illustrated acts may be used to implement a method in accordance with the disclosed subject matter. In addition, those skilled in the art will understand and appreciate that the method may alternatively be represented as a series of interrelated states via a state diagram or events. Furthermore, the methods disclosed herein can be stored on an article of manufacture, such as a non-transitory computer-readable medium, to facilitate transport and transfer of such methodologies to a computing device. The term article of manufacture, as used herein, is intended to encompass a computer program accessible from any computer-readable device or storage medium. While depicted as individual blocks, various blocks may be divided into additional blocks, combined into fewer blocks, or eliminated, depending on the desired implementation.

[0061] The method 500 may begin at block 502 where processing logic receives an initial radar signal of an object in motion at an initial time.

[0062] At block 504, processing logic may convert the initial radar signal into an initial input vector that includes initial time-frequency information.

[0063] At block 506, processing logic may process the initial input vector to determine an initial spin rate.

[0064] At block 508, processing logic may save the initial spin rate as one of the known spin rate data.

[0065] At block 510, processing logic may add one of the known spin rate data to the initial data set.

[0066] 6 is a block diagram illustrating an example system 600 that may be used for spin measurement and estimation, in accordance with at least one embodiment of the present disclosure. System 600 includes a processor 610, a memory 612, and a communication unit 616, all of which may be communicatively coupled. In some embodiments, system 600 may be part of any of the systems or devices described in this disclosure.

[0067] For example, system 600 may be part of computing system 120 or processor 112 of radar device 110 of Figure 1 and may be configured to perform one or more of the tasks described above with respect to computing system 120 or processor 112. As another example, system 600 may be part of computing module 220 of Figure 2 and may be configured to perform one or more of the tasks described above.

[0068] In general, processor 610 may include any computing entity or processing device, including various computer hardware or software modules, and may be configured to execute instructions stored on any available computer-readable storage medium. For example, processor 610 may include a microprocessor, a microcontroller, a parallel processor such as a graphics processing unit (GPU) or a tensor processing unit (TPU), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or any other digital or analog circuit configured to interpret and / or execute program instructions and / or process data.

[0069] 6 as a single processor, it is understood that processor 610 may include any number of processors distributed across any number of networks or physical locations configured to individually or collectively perform any number of operations described herein. In some embodiments, processor 610 may interpret and / or execute program instructions and / or data stored in memory 612. In some embodiments, processor 610 may execute program instructions stored in memory 612.

[0070] For example, in some embodiments, processor 610 may execute program instructions stored in memory 612 related to path encoding with delay constraints such that system 600 may be performed or directed to perform operations related thereto as directed by the instructions. In these and other embodiments, the instructions may be used to perform method 400 of FIG. 4 or method 500 of FIG. 5.

[0071] Memory 612 may include one or more computer-readable storage media for storing or executing computer-executable instructions or data structures stored thereon. Such computer-readable storage media may be any available media that can be accessed by a general-purpose or special-purpose computer, such as processor 610.

[0072] By way of example, and not limitation, such computer-readable storage media include random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), compact disk read-only memory (CD-ROM) or other optical disk storage, magnetic disk storage or other magnetic storage, flash memory devices (e.g., solid-state memory devices), or any other storage medium that can be used to execute or store specific program code in the form of computer-executable instructions or data structures and that can be accessed by a general-purpose or special-purpose computer. Combinations of the above can also be included within the scope of computer-readable storage media.

[0073] Computer-executable instructions may include, for example, instructions and data configured to cause processor 610 to perform a particular operation or group of operations, as described in this disclosure. In these and other embodiments, the term "non-transitory" as used in this disclosure should be construed to exclude only those types of transitory media held to be outside the scope of patentable subject matter in In re Nuijten, 500 F.3d 1346 (Fed. Cir. 2007). Combinations of the above may also be included within the scope of computer-readable media.

[0074] The communications unit 616 may include any component, device, system, or combination thereof configured to transmit and receive information over a network. In some embodiments, the communications unit 616 can communicate with other devices at other locations, the same location, or other components within the same system. For example, the communications unit 616 may include a modem, a network card (wireless or wired), an infrared communications device, a wireless communications device (such as an antenna), and / or a chipset (such as a Bluetooth® device, an 802.6 device (e.g., a metropolitan area network (MAN)), a Wi-Fi® device, a WiMax® device, a cellular communications facility, etc.). The communications unit 616 may enable data exchange with a network and / or any other device or system described in this disclosure. For example, if the system 600 is included in the computing system 120 of FIG. 1 , the communications unit 616 may enable the computing system 120 to communicate with the radar device 110 and / or the data storage 130.

[0075] Modifications, additions, or omissions may be made to system 600 without departing from the scope of the present disclosure. For example, in some embodiments, system 600 may include any number of other components that may not be explicitly shown or described. Furthermore, depending on the particular implementation, system 600 may not include one or more of the components shown and described.

[0076] As indicated above, the embodiments described herein may involve the use of a computing system (e.g., processor 610 of FIG. 6) including various computer hardware or software modules, as described in more detail below. Additionally, as indicated above, the embodiments described herein may be implemented using a computer-readable medium (e.g., memory 612 of FIG. 6) for executing or storing computer-executable instructions or data structures or having stored thereon.

[0077] In some embodiments, the different components, modules, engines, and services described herein may be implemented as objects or processes (e.g., as separate threads) running on a computing system. Although some of the systems and methods described herein are generally described as being implemented in software (stored on and / or executed by general-purpose hardware), specific hardware implementations, or combinations of software and specific hardware implementations, are also possible and contemplated.

[0078] In accordance with common practice, the various features illustrated in the drawings may not be drawn to scale. The illustrations presented in this disclosure are not intended to be actual diagrams of any particular apparatus (e.g., device, system, etc.) or method, but merely idealized representations employed to describe various embodiments of the present disclosure. Accordingly, dimensions of various features may be arbitrarily increased or decreased for clarity. Furthermore, some of the drawings may be simplified for clarity. Thus, the drawings may not depict all of the components of a given apparatus (e.g., device) or all operations of a particular method.

[0079] The terms used in this specification and particularly in the appended claims (e.g., the body of the appended claims) are generally intended as "open" terms (e.g., the term "including" should be interpreted as "including, but not limited to," the term "having" should be interpreted as "having at least," the term "including" should be interpreted as "including, but not limited to," etc.).

[0080] Furthermore, when a specific number is intended when reciting a derived claim, such intention will be expressly stated in the claim; otherwise, no such intention exists. For example, as an aid to understanding, the following appended claims may include the use of the introductory phrases "at least one" and "one or more" to introduce the claim. However, even if the same claim includes the derived phrase "one or more" or "at least one" and an indefinite article such as "a" or "an" (e.g., "a" and / or "an" should be interpreted to mean "at least one" or "one or more"), the use of such phrases should not be interpreted as limiting the referenced claim subject matter derived from the indefinite article "a" or "an" to refer to only one subject matter contained in the specification. The same applies to definite articles used in referenced claim subject matter.

[0081] Furthermore, even if a particular number in an introduced claim is explicitly recited, it is understood that such recitation should be interpreted to mean at least the recited number (e.g., the mere recitation of "two recitations" without other modifiers means at least two recitations, or more than two recitations). Furthermore, when a convention analogous to "at least one of A, B, and C, etc." or "one or more of A, B, and C, etc." is used, such construction is generally intended to include A alone, B alone, C alone, both A and B, both A and C, both B and C, or both A, B, and C, etc. For example, use of the term "and / or" is intended to be interpreted in this manner.

[0082] Furthermore, whether in the specification, claims, or drawings, disjunctive phrases presenting two or more alternative terms should be understood to contemplate the possibility of including one term, either term, or both terms. For example, the phrase "A or B" should be understood to include the possibilities of "A" or "B," or "A and B."

[0083] Furthermore, the use of terms such as “first,” “second,” and “third” is not used herein to necessarily imply a particular order or number of elements. Generally, terms such as “first,” “second,” and “third” are used as general identifiers to distinguish between different elements. Unless there is evidence indicating that terms such as “first,” “second,” and “third” imply a particular order, these terms should not be understood to imply a particular order. Furthermore, unless there is evidence indicating that terms such as “first,” “second,” and “third” imply a particular number of elements, these terms should not be understood to imply a particular number of elements. For example, a first widget may be described as having a first side, and a second widget may be described as having a second side. The use of the term “second side” with respect to a second widget is intended to distinguish such side of the second widget from the “first side” of the first widget, and does not imply that the second widget has two sides.

[0084] All examples and conditional statements incorporated herein are intended for educational purposes to aid the reader in understanding the concepts contributed by the inventors to further advance the present invention and the art, and are not to be construed as being limited to such specifically incorporated examples and conditions. Although embodiments of the present disclosure have been described in detail, it should be understood that various changes, substitutions, and alterations can be made herein without departing from the spirit and scope of the present disclosure.

Claims

1. receiving reflected radar signals from a moving object; converting the reflected radar signal into an input vector containing time-frequency information of the moving object; providing the input vector as an input to a neural network having access to an initial data set including a plurality of initial input vectors generated based on a plurality of initial radar signals of a plurality of moving objects at an initial time, each of the initial input vectors being associated with a known spin rate; determining a spin rate of the object in motion based on an analysis performed by the neural network of the input vector comprising the time-frequency information of the object in motion given the initial data set, wherein the neural network: comparing a first portion including a first set of harmonics in the time-frequency images within the input vector with a second portion including a second set of harmonics in the time-frequency images within each initial input vector of the initial data set; determining initial input vectors from the initial data set that are similar to the input vector based on the results of the comparison; determining a spin rate of the moving object based on known spin rates associated with the similar initial input vectors; and performing the analysis, comparing the first portion with the second portion includes comparing the number of harmonics associated with side lobes around a main lobe, the amount of spacing between the harmonics of the side lobes, and the rate of decay of the intensity of the side lobes over time between the first portion and the second portion; A method comprising:

2. The method of claim 1 , wherein the time-frequency image is obtained by performing a Fourier analysis on the reflected radar signal.

3. 2. The method of claim 1, wherein the time-frequency image includes a main lobe and one or more side lobes modulated symmetrically about the main lobe, the main lobe including high-energy portions of the input vector relative to low-energy portions of the input vector associated with the one or more side lobes.

4. The method of claim 3 , wherein the main lobe is represented in the time-frequency image as a substantially linear frequency that varies with time.

5. The method of claim 3 , wherein the main lobe of the time frequency image indicates a velocity of the particular object in motion, and the one or more side lobes of the time frequency image indicate a spin rate of the particular object in motion.

6. The method further includes obtaining the initial data set, wherein obtaining one of the initial data from the initial data set includes: receiving an initial radar signal of the moving object; converting the initial radar signal into the initial input vector including initial time-frequency information; processing the initial input vector to determine an initial spin rate; storing the initial spin rate as one of known spin rate data; and adding one of the known spin rate data to the initial data set.

7. displaying a visualization of said spin rate; and The method of claim 1 , further comprising: displaying a simulated visualization model of the object's trajectory.

8. The method of claim 1 , wherein the moving object is a first object and the initial plurality of moving objects includes a second object different from the first object.

9. receiving a first reflected radar signal of the first object; converting the first reflected radar signal into a first input vector containing time-frequency information of the first object; providing the first input vector as an input to the neural network, the neural network having access to a second data set generated based on a plurality of second reflected radar signals of a plurality of second objects; 9. The method of claim 8, further comprising: determining a spin rate of the first object based on an analysis performed by the neural network of the first input vector containing the time-frequency information of the first object, the analysis comprising comparing one or more elements of the first input vector with one or more elements of the second data set, taking into account the second data set.

10. Memory and a processor operatively coupled to the memory; The processor, when executed, receiving reflected radar signals from a moving object; converting the reflected radar signal into an input vector containing time-frequency information of the moving object; providing the input vector as an input to a neural network having access to an initial data set including a plurality of initial input vectors generated based on a plurality of initial radar signals of a plurality of moving objects at an initial time, each of the initial input vectors being associated with a known spin rate; determining a spin rate of the object in motion based on an analysis performed by the neural network of the input vector comprising the time-frequency information of the object in motion given the initial data set, wherein the neural network: comparing a first portion including a first set of harmonics in the time-frequency images within the input vector with a second portion including a second set of harmonics in the time-frequency images within each initial input vector of the initial data set; determining initial input vectors from the initial data set that are similar to the input vector based on the results of the comparison; determining a spin rate of the moving object based on known spin rates associated with the similar initial input vectors; and performing the analysis, comparing the first portion with the second portion includes comparing the number of harmonics associated with side lobes around a main lobe, the amount of spacing between the harmonics of the side lobes, and the rate of decay of the intensity of the side lobes over time between the first portion and the second portion; The system is configured to cause the processor to perform operations including:

11. The system of claim 10 , wherein the time-frequency information of the input vector comprises a time-frequency image obtained by performing a Fourier analysis on the reflected radar signal.

12. 12. The system of claim 11, wherein the time-frequency image includes a main lobe and one or more side lobes modulated symmetrically about the main lobe, the main lobe including high-energy portions of the input vector relative to low-energy portions of the input vector associated with the one or more side lobes.

13. The system of claim 12 , wherein the main lobe of the time frequency image indicates a velocity of the moving object, and the one or more side lobes of the time frequency image indicate a spin rate of the moving object.

14. displaying a visualization of said spin rate; and The system of claim 10 , further comprising: displaying a simulated visualization model of the object's trajectory.

15. receiving reflected radar signals from a moving object; converting the reflected radar signal into an input vector containing time-frequency information of the moving object; One of the initial data in the initial dataset is receiving an initial radar signal of a moving object; converting the initial radar signal into an initial input vector containing initial time-frequency information; processing the initial input vector to determine an initial spin rate; storing the initial spin rate as one of known spin rate data; obtaining the initial data set, the initial data set including adding one of the known spin rate data to the initial data set; providing said input vector as an input to a neural network that has access to said initial data set; determining a spin rate of the object in motion based on an analysis performed by the neural network of the input vector containing time-frequency information of the object in motion taking into account the initial data set, the neural network comprising: comparing a first portion including a first set of harmonics in the time-frequency images within the input vector with a second portion including a second set of harmonics in the time-frequency images within each initial input vector of the initial data set; determining initial input vectors from the initial data set that are similar to the input vector based on the results of the comparison; determining a spin rate of the moving object based on the initial spin rate associated with the similar initial input vector; and performing the analysis, comparing the first portion with the second portion includes comparing the number of harmonics associated with side lobes around a main lobe, the amount of spacing between the harmonics of the side lobes, and the rate of decay of the intensity of the side lobes over time between the first portion and the second portion; A non-transitory computer-readable medium having processor-executable programming code for performing operations including:

16. 16. The non-transitory computer-readable medium of claim 15, wherein the object in motion initially is a first object and the object in motion is a second object different from the first object.

17. The method of claim 8 , wherein the first object is a ball of a first type and the second object is a ball of a second type.

Images