Cyclic convolution fast positioning method and device for laser radar ranging

Abstract

The application provides a cyclic convolution fast positioning method and device for laser radar ranging, and the method comprises the following steps: applying a laser radar device to generate a laser signal, and obtaining a reference signal and an echo signal based on the laser signal; performing a cross-correlation calculation based on cyclic convolution according to the reference signal and the echo signal to obtain a cross-correlation sequence; constructing a cross-correlation time delay diagram based on the cross-correlation sequence and a time sequence, and obtaining the flight time of laser according to the cross-correlation time delay diagram to perform laser radar ranging, wherein the time sequence is generated by a preset cross-correlation point number and a sampling frequency. The application can efficiently improve the calculation speed, and improve the efficiency and accuracy of laser radar ranging.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of signal processing, and particularly relates to a cyclic convolution fast positioning method and device for laser radar ranging. BACKGROUND

[0002] In the field of radar ranging, the most commonly used method to measure the distance between the radar device and the target object is to calculate the time of flight of the signal in the medium. For analog ranging, only the state of the signal to the analog device needs to be detected to calculate the distance of the target to the transmitting end. However, with the development of science and technology and the times, the requirements for radar performance are constantly increasing, especially the resolution, speed and anti-interference, and the traditional analog radar ranging method cannot fully meet these requirements due to hardware limitations. Therefore, digital radar, which mainly stores data and processes digital signals, has become the mainstream in the field of radar;

[0003] The radar that can best meet the resolution, speed and anti-interference in digital radar is a wideband chaotic laser radar, and one of the most commonly used methods for processing such high-bandwidth signals is cross-correlation calculation. In fact, the essence of this signal processing method is still to calculate the propagation time of the signal in the medium. By calculating the correlation coefficient of the reference signal and the echo signal, the corresponding correlation coefficient is the highest, and the time point corresponding to the cross-correlation peak point is the propagation time of the signal in the medium.

[0004] However, to meet the requirements of high resolution and strong anti-interference, the amount of data collected will inevitably be large. The traditional cross-correlation calculation method calculates by the product difference method, and is based on the reference signal sequence and the echo signal sequence. The correlation between the two variables is reflected by the two mean-centered signals. This calculation method will slow down the calculation speed as the data volume continues to stack, and is not suitable for applications such as high-speed imaging that require high-speed ranging. SUMMARY

[0005] The present application provides a cyclic convolution fast positioning method and device for laser radar ranging to solve the problem that the existing cross-correlation calculation will slow down the calculation speed as the signal data volume continues to stack, and is not suitable for high-speed ranging applications.

[0006] To achieve the above purpose, the present application provides a cyclic convolution fast positioning method for laser radar ranging, which comprises: applying a laser radar device to generate a laser signal, and obtaining a reference signal and an echo signal based on the laser signal; performing cross-correlation calculation based on cyclic convolution according to the reference signal and the echo signal to obtain a cross-correlation sequence; constructing a cross-correlation time delay graph based on the cross-correlation sequence and a time sequence, and obtaining the flight time of the laser based on the cross-correlation time delay graph to perform laser radar ranging, wherein the time sequence is generated by a preset cross-correlation point number and a sampling frequency.

[0007] Optionally, the cyclic convolution-based cross-correlation calculation according to the reference signal and the echo signal to obtain a cross-correlation sequence comprises: sampling the reference signal and the echo signal respectively to obtain a reference signal sequence and an echo signal sequence; obtaining the echo mean normalization sequence according to the echo signal sequence; and performing cyclic convolution-based cross-correlation calculation according to the echo mean normalization sequence and the reference signal sequence to obtain the cross-correlation sequence.

[0008] Optionally, the sampling the reference signal and the echo signal respectively to obtain a reference signal sequence and an echo signal sequence comprises: sampling the reference signal and the echo signal by N+n points and N points respectively according to a preset sampling point number N, a cross-correlation point number and a cross-correlation starting point, to obtain the reference signal sequence and the echo signal sequence, wherein n is the cross-correlation point number.

[0009] Optionally, the obtaining the echo mean normalization sequence according to the reference signal sequence and the echo signal sequence comprises: performing mean centering processing on the echo signal sequence, and adding n zeros at the end to obtain the echo mean normalization sequence.

[0010] Optionally, the cyclic convolution-based cross-correlation calculation according to the echo mean normalization sequence and the reference signal sequence to obtain the cross-correlation sequence comprises: constructing a cyclic matrix according to the reference signal sequence; constructing a standard sequence ONE with N ones and n zeros, and constructing a standard cyclic matrix one according to the standard sequence ONE, wherein N is the sampling point number and n is the cross-correlation point number; and performing cyclic convolution-based cross-correlation calculation according to the echo mean normalization sequence Es, the cyclic matrix and the standard cyclic matrix one by using the following relationship:

[0011]

[0012] wherein s and rs are the sampled echo signal and the reference signal respectively, is a cyclic convolution operation with a length of n, and xcor is the cross-correlation sequence.

[0013] Optionally, the cyclic convolution-based cross-correlation calculation according to the echo mean normalization sequence and the reference signal sequence to obtain a cross-correlation sequence comprises: constructing a standard sequence with N 1s and n 0s, where N is the number of sampling points and n is the number of cross-correlation points; performing fast Fourier transform on the echo mean normalization sequence and the standard sequence to obtain a frequency domain echo mean normalization sequence and a frequency domain standard sequence; performing fast Fourier transform on the reference signal sequence and the square of the reference signal sequence to obtain a first reference frequency domain sequence and a second reference frequency domain sequence; and performing cyclic convolution-based cross-correlation calculation according to the frequency domain echo mean normalization sequence, the frequency domain standard sequence, the first reference frequency domain sequence and the second reference frequency domain sequence to obtain a cross-correlation sequence.

[0014] Optionally, the cyclic convolution-based cross-correlation calculation according to the frequency domain echo mean normalization sequence, the frequency domain standard sequence, the first reference frequency domain sequence and the second reference frequency domain sequence to obtain a cross-correlation sequence comprises: applying the following relationship formula to perform cyclic convolution-based cross-correlation calculation according to the frequency domain echo mean normalization sequence FES, the frequency domain standard sequence Fone, the first reference frequency domain sequence FSr and the second reference frequency domain sequence FSr2:

[0015]

[0016] wherein xcor is the cross-correlation sequence and IFFT represents inverse Fourier transform.

[0017] Based on the same inventive concept, the embodiments of the present application further provide a cyclic convolution-based fast positioning device for laser radar ranging, comprising: a signal acquisition unit configured to generate a laser signal by applying a laser radar device, and acquire a reference signal and an echo signal based on the laser signal; a cross-correlation calculation unit configured to perform cyclic convolution-based cross-correlation calculation according to the reference signal and the echo signal to obtain a cross-correlation sequence; and a time of flight acquisition unit configured to construct a cross-correlation time delay map based on the cross-correlation sequence and a time sequence, and acquire the time of flight of the laser based on the cross-correlation time delay map to perform laser radar ranging, wherein the time sequence is generated by a preset number of cross-correlation points and a sampling frequency.

[0018] Based on the same inventive concept, the embodiments of the present application further provide an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the foregoing method when executing the program.

[0019] Based on the same inventive concept, the embodiments of the present application further provide a computer storage medium, wherein the computer storage medium stores at least one executable instruction, and the executable instruction causes the processor to execute the foregoing method.

[0020] The beneficial effects of the present invention are as follows: As can be seen from the above description, the embodiments of the present invention provide a fast positioning method and apparatus for lidar ranging using circular convolution. The method includes: generating a laser signal using a lidar device, and acquiring a reference signal and an echo signal based on the laser signal; performing cross-correlation calculation based on circular convolution according to the reference signal and the echo signal to obtain a cross-correlation sequence; constructing a cross-correlation delay map based on the cross-correlation sequence and the time series, and obtaining the flight time of the laser according to the cross-correlation delay map for lidar ranging. The time series is generated by a preset number of cross-correlation points and a sampling frequency, which can efficiently improve the calculation speed and improve the efficiency and accuracy of lidar ranging. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a flowchart illustrating the fast localization method using circular convolution for lidar ranging in an embodiment of the present invention.

[0023] Figure 2 This is a schematic diagram of the structure of the lidar device in an embodiment of the present invention;

[0024] Figure 3 This is a schematic diagram illustrating the complete process of cross-correlation calculation using matrix operations in an embodiment of the present invention;

[0025] Figure 4 This is a schematic diagram illustrating the complete process of cross-correlation calculation using the method of Fast Fourier Transform in an embodiment of the present invention;

[0026] Figure 5 For based on Figure 3 The cross-correlation delay plot is constructed from the cross-correlation sequences obtained by the method shown.

[0027] Figure 6 For based on Figure 4 The cross-correlation delay plot is constructed from the cross-correlation sequences obtained by the method shown.

[0028] Figure 7 For based on Figure 4 The graph showing the relationship between the number of relevant points and time for the method shown;

[0029] Figure 8 For based on Figure 4 The diagram illustrates the signal-to-noise ratio variation under different sample sizes for the method shown.

[0030] Figure 9 Based on Figure 4 A schematic diagram illustrating the signal-to-noise ratio variation under different numbers of cross-correlation points for the method shown.

[0031] Figure 10 This is a schematic diagram of the structure of the circular convolution fast positioning device for lidar ranging in an embodiment of the present invention;

[0032] Figure 11 This is a schematic diagram of an electronic device in an embodiment of the present invention. Detailed Implementation

[0033] To make the objectives, technical solutions, and advantages of this disclosure clearer, the following detailed description is provided in conjunction with specific embodiments and the accompanying drawings.

[0034] It should be noted that, unless otherwise defined, the technical or scientific terms used in the embodiments of this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this disclosure pertains. The terms "first," "second," and similar terms used in the embodiments of this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are only used to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0035] This invention provides a fast localization method using recurrent convolution for lidar ranging. This fast localization method using recurrent convolution for lidar ranging is applied to a server. (See attached...) Figure 1 As shown, the fast localization method using recurrent convolution for lidar ranging includes:

[0036] Step S11: Use a lidar device to generate a laser signal, and acquire a reference signal and an echo signal based on the laser signal.

[0037] In this embodiment of the invention, a chaotic lidar device was constructed to generate laser signals. Reference signals and echo signals were acquired based on the generated lidar signals and used as input to a signal processing system. The laser signals generated by this lidar device possess ultra-high bandwidth and high data characteristics, effectively representing the verification object of this invention. Figure 2As shown, the laser radar device includes a distributed feedback (DFB) laser, a first polarization controller, a circulator, a 50 / 50 coupler, an attenuator, a second polarization controller, an optical isolator, and a 20 / 80 coupler. The above devices are used to generate a chaotic laser signal with ultra-high bandwidth; the 80 end of the 20 / 80 coupler is input into a transceiver device, and the received chaotic signal is collected as a return signal and transmitted to an upper computer; the 20 end of the 20 / 80 coupler is transmitted to the upper computer as a reference signal. Specifically, the low-power output of the 20 / 80 coupler in the laser radar device is collected through a real-time oscilloscope channel 1, and the collected data is transmitted to the upper computer as a reference signal for the cross-correlation cyclic convolution method, named RS; the output of the avalanche photodetector in the receiving device is collected through a real-time oscilloscope channel 2, and the collected data is transmitted to the upper computer as a return signal for the cross-correlation cyclic convolution method, named S.

[0038] Step S12: performing cross-correlation calculation based on cyclic convolution according to the reference signal and the return signal to obtain a cross-correlation sequence.

[0039] In the embodiment of the present application, the hyperparameters are first input, including sampling frequency, sampling point number, cross-correlation point number, and cross-correlation starting point. The specific values can be set as needed. For example, the sampling frequency is consistent with the sampling frequency of the real-time oscilloscope, which is 20 GHz, the sampling point number N is set to 20001, and the cross-correlation point number n is set to 10001 or 1000000.

[0040] In step S12, optionally, the reference signal and the return signal are first sampled respectively to obtain a reference signal sequence and a return signal sequence. Specifically, the reference signal RS and the return signal S are sampled by N+n points and N points respectively according to the preset sampling point number N, cross-correlation point number, and cross-correlation starting point, to obtain a reference signal sequence rs and a return signal sequence s, n being the cross-correlation point number.

[0041] Then, the return mean normalized sequence is obtained according to the return signal sequence. Specifically, the return signal sequence s can be subjected to mean centering processing, and n zeros are added at the end to obtain the return mean normalized sequence. The return signal sequence s is subjected to mean centering processing by applying the following relationship:

[0042]

[0043] wherein i, j = 0, 1, …, N. Then, n zeros are added at the end to obtain the return mean normalized sequence Es, ES(i) being the i-th element of the return mean normalized sequence Es, and s(i) being the i-th element of the return signal sequence s.

[0044] Finally, cross-correlation calculation based on circular convolution is performed based on the normalized echo mean sequence and the reference signal sequence to obtain the cross-correlation sequence. In the embodiment of the invention, matrix operation method can be applied to perform cross-correlation calculation based on circular convolution. In this case, the sampling frequency is consistent with the sampling frequency of the real-time oscilloscope, which is 20 GHz, the number of sampling points N is set to 20001, and the number of cross-correlation points n is set to 10001. Specifically, a circular matrix is ​​constructed based on the reference signal sequence; a standard sequence ONE is constructed using N 1s and n 0s, and a standard circular matrix one is constructed based on the standard sequence ONE, where N is the number of sampling points and n is the number of cross-correlation points; cross-correlation calculation based on circular convolution is performed based on the following cross-correlation relationship using the normalized echo mean sequence Es, the circular matrix, and the standard circular matrix one:

[0045]

[0046] Where s and rs are the sampled echo signal and the reference signal, respectively. This is a circular convolution operation of length n, where one is a sequence of length N consisting entirely of 1s, and xcor is a cross-correlation sequence. The circular matrix form is constructed from the echo mean-normalized sequence Es as follows:

[0047]

[0048] Among them, ES i Let be the i-th element in the echo mean normalized sequence Es.

[0049] The traditional formula for calculating cross-correlation is:

[0050]

[0051] This invention's embodiment of cross-correlation based on circular convolution replaces the traditional for-loop structure. Each circular convolution operation involves only one matrix multiplication. Traditional cross-correlation calculations require calculating the correlation coefficient n times, with each correlation requiring N*N operations, consuming significant time. This invention's embodiment, using circular convolution in parallel processing, effectively overcomes the time consumption problem of traditional cross-correlation. (Circular convolution) The calculation formula is as follows:

[0052]

[0053] The corresponding matrix multiplication is:

[0054]

[0055] The complete flow of the method for cross-correlation calculation by applying the matrix operation of the embodiment of the application is shown in Figure 3

[0056] Step 100: Start.

[0057] Step 101: Input reference signal RS and echo signal S.

[0058] The reference signal RS and the echo signal S are obtained by using the method described above, which will not be described here.

[0059] Step 102: Define parameters cross-correlation length N and cross-correlation point number n.

[0060] The sampling frequency is consistent with that of the real-time oscilloscope, which is 20 GHz, the sampling point number N is 20001, and the cross-correlation point number n is 10001.

[0061] Step 103: Sample the reference signal for N+n points.

[0062] The reference signal RS is sampled for N+n points to obtain the reference signal sequence rs.

[0063] Step 104: Construct the cyclic matrix rs of rs. Then jump to execute step 108.

[0064] The cyclic matrix rs is constructed according to the reference signal sequence rs.

[0065] Step 105: Sample the echo signal S for N points.

[0066] The echo signal S is sampled for N points to obtain the echo signal sequence s.

[0067] Step 106: Calculate the mean normalized sequence ES of the echo signal sequence s. Then jump to execute step 108.

[0068] The mean centering processing is sequentially performed on each element in the echo signal sequence s to obtain the mean normalized sequence ES.

[0069] Step 107: Construct a sequence one with N elements all being 1. Then execute step 108.

[0070] Step 108: Perform cross-correlation calculation based on cyclic convolution by applying the cross-correlation relationship.

[0071] The cross-correlation calculation based on cyclic convolution is performed by applying the cross-correlation relationship according to the echo mean normalized sequence Es, the cyclic matrix and the standard cyclic matrix one to obtain the cross-correlation sequence xcor.

[0072] Step 109: Output.

[0073] ​Output cross-correlation sequence xcor.

[0074] Step 110: End.

[0075] The matrix multiplication is used to calculate the cyclic convolution, and when the cross-correlation point data amount does not exceed the display memory, there is obvious acceleration effect.

[0076] In the embodiment of the application, the method of fast Fourier transform on the echo mean normalized sequence and the reference signal sequence can also be used to calculate the cyclic convolution. At this time, the sampling frequency is consistent with the sampling frequency of the real-time oscilloscope, which is 20GHz, the sampling point number N is set to 20001, and the cross-correlation point number n is set to 1000000. In order to prevent the array capacity overflow in the program, the cross-correlation point number n is automatically adjusted according to the sample point number N under the premise of ensuring that the samples can be completely cross-correlated, that is, N+n is less than the total length of the signal. Specifically, the echo mean normalized sequence is obtained by using the same method as before. The standard sequence is constructed by using N 1s and n 0s, where N is the sampling point number and n is the cross-correlation point number; the fast Fourier transform is performed on the echo mean normalized sequence and the standard sequence to obtain the frequency domain echo mean normalized sequence and the frequency domain standard sequence; the fast Fourier transform is performed on the reference signal sequence and the square of the reference signal sequence to obtain the first reference frequency domain sequence and the second reference frequency domain sequence; the cross-correlation calculation based on the cyclic convolution is performed according to the frequency domain echo mean normalized sequence, the frequency domain standard sequence, the first reference frequency domain sequence and the second reference frequency domain sequence to obtain the cross-correlation sequence. Specifically, the cross-correlation calculation of the cyclic convolution is performed according to the frequency domain echo mean normalized sequence FES, the frequency domain standard sequence Fone, the first reference frequency domain sequence FSr and the second reference frequency domain sequence FSr2 by using the following cross-correlation relationship:

[0077]

[0078] Wherein, xcor is the cross-correlation sequence, and IFFT represents the inverse Fourier transform.

[0079] The complete flow of the application of the method of fast Fourier transform for cross-correlation calculation in the embodiment of the application is as shown in Figure 4 , which comprises:

[0080] Step 200: Start.

[0081] Step 201: Input the reference signal RS and the echo signal S.

[0082] The reference signal RS and the echo signal S are obtained by using the method described above, which will not be described here.

[0083] Step 202: Define the parameters cross-correlation length N and cross-correlation point number n. Then, step 203 and step 205 are executed.

[0084] The sampling frequency is consistent with that of the real-time oscilloscope, which is 20GHz, the sampling point number N is 20001, and the cross-correlation point number n is 1000000.

[0085] Step 203: N+n points are intercepted in the reference signal center position as correlation calculation.

[0086] N+n points are intercepted in the reference signal center position to obtain a reference signal sequence Sr, for subsequent correlation calculation.

[0087] Step 204: FFT transform is performed on Sr and Sr 2 to obtain FSr. Then jump to step 212.

[0088] FFT transform is performed on the reference signal sequence Sr and the square of the reference signal sequence Sr to obtain a first reference frequency domain sequence FSr and a second reference frequency domain sequence FSr2.

[0089] Step 205: n points are intercepted in the echo signal center position as correlation calculation.

[0090] n points are intercepted in the echo signal center position to obtain an echo signal sequence, for subsequent correlation calculation.

[0091] Step 206: Mean normalization is performed on the echo signal sequence.

[0092] Each element in the echo signal sequence is processed by mean centering to obtain a mean normalized sequence ES.

[0093] Step 207: The mean normalized sequence ES is followed by n zeros.

[0094] Step 208: FFT transform is performed on the zero-padded mean normalized sequence ES to obtain FES. Then jump to step 212.

[0095] FFT transform is performed on the zero-padded mean normalized sequence ES to obtain a frequency domain echo mean normalized sequence FES.

[0096] Step 209: A sequence one with N rows and all elements being 1 is constructed.

[0097] Step 210: n zeros are appended to the sequence one.

[0098] Step 211: FFT transform is performed on the zero-padded one to obtain Fone. Then step 212 is performed.

[0099] FFT transform is performed on the zero-padded sequence one to obtain a frequency domain standard sequence Fone

[0100] Step 212: applying a cross-correlation formula to perform a cross-correlation calculation based on cyclic convolution.

[0101] According to the frequency domain echo mean normalized sequence FES, the frequency domain standard sequence Fone, the first reference frequency domain sequence FSr and the second reference frequency domain sequence FSr2, a cross-correlation formula is applied to perform a cross-correlation calculation based on cyclic convolution to obtain a cross-correlation sequence xcor.

[0102] Step 213: output.

[0103] The cross-correlation sequence xcor is output.

[0104] Step 214: end.

[0105] The embodiment of the application performs fast Fourier transform on the initial signals, i.e., the reference signal and the echo signal, converts the time domain signal to the frequency domain, multiplies the reference signal and the echo signal in the frequency domain to obtain the corresponding frequency domain sequence, and then converts to the time domain through inverse fast Fourier transform, and then performs formula calculation to obtain the corresponding cross-correlation sequence. The error of the cross-correlation time delay calculated by the fast Fourier transform method is zero, and the calculation speed under large data is greatly improved.

[0106] Step S13: constructing a cross-correlation time delay graph based on the cross-correlation sequence and a time sequence, and obtaining the time of flight of the laser according to the cross-correlation time delay graph to perform laser radar ranging, wherein the time sequence is generated by a preset cross-correlation point number and a sampling frequency.

[0107] In the embodiment of the application, the time sequence is generated according to the cross-correlation point number and the sampling frequency, and then the cross-correlation sequence corresponds to the generated time sequence, and the cross-correlation time delay graph is constructed. Based on the time sequence and Figure 3 The cross-correlation time delay graph constructed by the cross-correlation sequence obtained by the method shown in Figure 5 , based on the time sequence and Figure 4 The cross-correlation time delay graph constructed by the cross-correlation sequence obtained by the method shown in Figure 6 In the cross-correlation time delay graph, the time point corresponding to the peak value is the time of flight of the laser, and then the laser radar ranging can be performed according to the time of flight.

[0108] Figure 7The point number and time relationship diagram for applying the FFT operation in the embodiment of the present application can be seen that, although the overall trend rises with the increase of the cross-correlation point number, the rising amplitude is very small, it can be seen that the running time of most data volume is below 1s, and the average value of the running time of the cross-correlation from 10,000 to 200,000 is 0.65s. Even near the maximum cross-correlation point number, there are still very small running time cases. Under the premise of fast calculation, the signal-to-noise ratio (SNR) under different sample quantities and different cross-correlation point numbers is analyzed, Figure 8 The change trend of the signal-to-noise ratio (SNR) under different sample quantities is shown in the figure; the more the sample quantity, the higher the signal-to-noise ratio (SNR). Figure 9 The change trend of the signal-to-noise ratio (SNR) under different cross-correlation point numbers is shown in the figure, and the signal-to-noise ratio (SNR) is relatively stable and only fluctuates in a very small range with the change of the cross-correlation point number.

[0109] The cyclic convolution fast positioning method for laser radar ranging of the embodiment of the present application generates a laser signal by applying a laser radar device, and obtains a reference signal and a return signal based on the laser signal; cross-correlation calculation based on cyclic convolution is performed according to the reference signal and the return signal to obtain a cross-correlation sequence; a cross-correlation time delay diagram is constructed based on the cross-correlation sequence and a time sequence, and the time of flight of laser is obtained according to the cross-correlation time delay diagram to perform laser radar ranging, wherein the time sequence is generated by a preset cross-correlation point number and a sampling frequency, which can efficiently improve the calculation speed and improve the efficiency and accuracy of laser radar ranging.

[0110] The above describes a specific embodiment of the present application. In some cases, the actions or steps described in the embodiment of the present application can be performed in an order different from that in the embodiment and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are possible or can be advantageous.

[0111] Based on the same concept, the embodiment of the present application also provides a cyclic convolution fast positioning device for laser radar ranging. It is applied to a server. Figure 10 As shown in the figure, the cyclic convolution fast positioning device for laser radar ranging comprises a signal acquisition unit, a cross-correlation calculation unit and a time of flight acquisition unit. Among them,

[0112] The signal acquisition unit is used to generate a laser signal by applying a laser radar device, and obtain a reference signal and a return signal based on the laser signal;

[0113] The cross-correlation calculation unit is used to perform cross-correlation calculation based on cyclic convolution according to the reference signal and the return signal to obtain a cross-correlation sequence;

[0114] a time-of-flight acquisition unit configured to construct a cross-correlation time-delay map based on the cross-correlation sequence and the time sequence, and to acquire a time of flight of the laser light for laser radar ranging according to the cross-correlation time-delay map, wherein the time sequence is generated by a preset number of cross-correlation points and a sampling frequency.

[0115] For the convenience of description, the above apparatus is described in various modules in terms of functions. Of course, the functions of the modules can be implemented in one or more software and / or hardware when implementing the embodiments of the present application.

[0116] The apparatus of the above embodiments is applied to the corresponding method in the above embodiments, and has the beneficial effects of the corresponding method embodiments, which are not described herein again.

[0117] Based on the same inventive concept, the embodiments of the present application also provide an electronic device, which comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the method according to any one of the above embodiments when executing the program.

[0118] The embodiments of the present application provide a non-volatile computer storage medium, which stores at least one executable instruction, and the computer executable instruction can execute the method according to any one of the above embodiments.

[0119] Figure 11 A more specific hardware structure schematic diagram of an electronic device provided by the present embodiment is shown, which can include a processor 1101, a memory 1102, an input / output interface 1103, a communication interface 1104 and a bus 1105. The processor 1101, the memory 1102, the input / output interface 1103 and the communication interface 1104 are connected to each other through the bus 1105 for communication within the device.

[0120] The processor 1101 can be implemented in the form of a general-purpose CPU (Central Processing Unit, central processor), a microprocessor, an application specific integrated circuit (ASIC), or one or more integrated circuits, etc., for executing related programs to implement the technical solutions provided by the method embodiments of the present application.

[0121] The memory 1102 can be implemented in the form of a ROM (Read Only Memory), a RAM (Random Access Memory), a static storage device, a dynamic storage device, etc. The memory 1102 can store an operating system and other application programs, and when the technical solutions provided by the embodiments of the application are implemented by software or firmware, the related program codes are stored in the memory 1102 and are called and executed by the processor 1101.

[0122] The input / output interface 1103 is configured to connect an input / output module to realize information input and output. The input / output module can be configured in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. The input device can include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device can include a display, a speaker, a vibrator, an indicator light, etc.

[0123] The communication interface 1104 is configured to connect a communication module (not shown in the figure) to realize the communication interaction between the device and other devices. The communication module can realize communication through a wired manner (such as a USB, a network cable, etc.) or a wireless manner (such as a mobile network, WIFI, Bluetooth, etc.).

[0124] The bus 1105 includes a channel to transmit information between various components (such as the processor 1101, the memory 1102, the input / output interface 1103, and the communication interface 1104) of the device.

[0125] It should be noted that although the above device only shows the processor 1101, the memory 1102, the input / output interface 1103, the communication interface 1104, and the bus 1105, in the specific implementation process, the device can also include other components necessary for normal operation. In addition, those skilled in the art can understand that the above device can also only include the components necessary to implement the embodiments of the application, and does not have to include all the components shown in the figure.

[0126] Those skilled in the art should understand that the above discussion of any embodiment is only exemplary and is not intended to limit the scope of the application to these examples; under the idea of the present application, the above embodiments or technical features in different embodiments can also be combined, the steps can be implemented in any order, and there are many other changes of different aspects of the application as described above. In order to be brief, they are not provided in detail.

[0127] This application is intended to cover any alternatives, modifications, and variations of the embodiments of the present application falling within the scope of the present application. Accordingly, any and all such alternatives, modifications, variations, omissions, and equivalents are included within the scope of the present application.

Claims

1. A fast localization method using recurrent convolution for lidar ranging, characterized in that, The method includes: A laser signal is generated using a lidar device, and a reference signal and an echo signal are acquired based on the laser signal; The process involves performing cross-correlation calculations based on circular convolution on the reference signal and the echo signal to obtain a cross-correlation sequence. This includes: sampling the reference signal and the echo signal respectively to obtain a reference signal sequence and an echo signal sequence; obtaining an echo mean normalized sequence based on the echo signal sequence; and performing cross-correlation calculations based on circular convolution on the echo mean normalized sequence and the reference signal sequence to obtain the cross-correlation sequence. The step of sampling the reference signal and the echo signal respectively to obtain the reference signal sequence and the echo signal sequence includes: sampling the reference signal and the echo signal at N+n points and N points respectively according to the preset number of sampling points N, the number of cross-correlation points and the cross-correlation start point to obtain the reference signal sequence and the echo signal sequence, where n is the number of cross-correlation points; A cross-correlation delay map is constructed based on the cross-correlation sequence and the time series, and the flight time of the laser is obtained according to the cross-correlation delay map for lidar ranging. The time series is generated by a preset number of cross-correlation points and sampling frequency.

2. The method as described in claim 1, characterized in that, The step of obtaining the echo mean normalized sequence based on the reference signal sequence and the echo signal sequence includes: The echo signal sequence is mean-centered and n zeros are added to the end to obtain the echo mean-normalized sequence.

3. The method as described in claim 1, characterized in that, The step of performing cross-correlation calculation based on circular convolution based on the echo mean normalized sequence and the reference signal sequence to obtain the cross-correlation sequence includes: Construct a cyclic matrix based on the reference signal sequence; Construct a standard sequence ONE using N ones and n zeros, and construct a standard cyclic matrix ONE based on the standard sequence ONE, where N is the number of sampling points and n is the number of cross-correlation points; Based on the echo mean normalized sequence Es, the cyclic matrix, and the standard cyclic matrix one, the following relationship is applied to perform cross-correlation calculation based on circular convolution: Where s and rs are the sampled echo signal sequence and the reference signal sequence, respectively. This is a circular convolution operation of length n, where xcor is a cross-correlation sequence.

4. The method as described in claim 1, characterized in that, The step of performing cross-correlation calculation based on circular convolution based on the echo mean normalized sequence and the reference signal sequence to obtain the cross-correlation sequence includes: Construct a standard sequence using N ones and n zeros, where N is the number of sampling points and n is the number of cross-correlation points; Perform a Fast Fourier Transform on the normalized echo mean sequence and the standard sequence to obtain a frequency domain normalized echo mean sequence and a frequency domain standard sequence. A first reference frequency domain sequence and a second reference frequency domain sequence are obtained by performing a fast Fourier transform on the reference signal sequence and the square of the reference signal sequence. Cross-correlation calculation based on circular convolution is performed on the frequency domain echo mean normalized sequence, the frequency domain standard sequence, the first reference frequency domain sequence, and the second reference frequency domain sequence to obtain a cross-correlation sequence.

5. A fast positioning device using circular convolution for lidar ranging, characterized in that, The device includes: The signal acquisition unit is used to generate a laser signal using a lidar device, and to acquire a reference signal and an echo signal based on the laser signal; A cross-correlation calculation unit is used to perform cross-correlation calculation based on circular convolution based on the reference signal and the echo signal to obtain a cross-correlation sequence; including: sampling the reference signal and the echo signal respectively to obtain a reference signal sequence and an echo signal sequence; obtaining an echo mean normalized sequence based on the echo signal sequence; performing cross-correlation calculation based on circular convolution based on the echo mean normalized sequence and the reference signal sequence to obtain a cross-correlation sequence; the step of sampling the reference signal and the echo signal respectively to obtain the reference signal sequence and the echo signal sequence includes: sampling the reference signal and the echo signal at N+n points and N points respectively according to a preset number of sampling points N, a number of cross-correlation points and a cross-correlation starting point to obtain the reference signal sequence and the echo signal sequence, where n is the number of cross-correlation points; The flight time acquisition unit is used to construct a cross-correlation delay map based on the cross-correlation sequence and the time series, and to acquire the flight time of the laser according to the cross-correlation delay map for lidar ranging, wherein the time series is generated by a preset number of cross-correlation points and sampling frequency.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method as described in any one of claims 1-4.

7. A computer storage medium, characterized in that, The storage medium stores at least one executable instruction that causes the processor to perform the method as described in any one of claims 1-4.

Images