TWI685266B - Vehicle network data sampling conversion method and system - Google Patents

Abstract

The invention is a vehicle network data sampling conversion method, which is executed by a vehicle host, which is used to receive an original signal and select a data signal in the original signal, and create a data table according to the selected data signal; vehicle The host further determines whether there is a data signal that has not been selected in the original signal, and when there is no data signal that has not been selected in the original signal, the information corresponding to a first time sequence is used as a reference After performing differential sampling to generate a differential data table, the information is compressed and uploaded to the differential data table. The invention greatly reduces the amount of information by performing differential sampling. In this way, the compression rate of data can be effectively improved, the amount of data can be reduced, and the delay of information transmission can be avoided.

Description

Vehicle network data sampling conversion method and system

The invention relates to a vehicle network data sampling conversion method and system, in particular to a vehicle network data sampling conversion method and system for improving data compression ratio.

Communication between multiple electronic devices is achieved through the mutual transmission of data. In the process of data transmission, due to the limitations of various hardware and software technologies, data can only be transmitted at a limited speed.

Therefore, in computer science and information theory, a data compression technology has been developed to achieve the purpose of transmitting more data at a limited speed. Data compression refers to the process of representing information according to a specific encoding mechanism and using less data than before encoding. To put it simply, the data volume of the original data is compressed through a specific coding mechanism, thereby reducing the data volume. A common example is the ZIP file format, which not only provides compression, but also serves as an archiving tool that can store many files into the same file.

However, with the development of technology, the amount of data is becoming larger and larger, such as high-quality digital audio and video files, and the existing data compression methods are not enough to compress the data to the extent that it can be smoothly transmitted without delay. The time will be too long, resulting in a delay in the transmission of information, so that users must wait, and wait until the data transmission is completed before they can continue to use the electronic device to perform subsequent functions.

For example, the US patent US8010704B2 discloses a compression method for in-vehicle CAN (Controller Area Network) data, which mainly encodes CAN Data and uses Byte as a single Grouping (Biting), and then performing subtraction for each Grouping to perform compression, but the previous case still has the following problems:

1. It can only compress CAN data and cannot be used for optical radar (Light Detection And Ranging; Lidar) data compression.

2. Because the data must be grouped in bytes (Byte) and then compressed, it will increase the complexity of the processor's pre-calculation.

In view of the shortcomings of the existing data compression method that cannot be applied to optical radar data compression and complex pre-calculation, the present invention provides a vehicle network data sampling conversion method and system, which can improve the application of optical radar data and reduce the pre-calculation the complexity.

The vehicle network data sampling and conversion method includes the following steps: receiving an original signal, wherein the original signal is an optical radar signal; selecting one of the data signals from the original signal; creating a data table based on the selected data signal, It further includes the following sub-steps: according to the selected data signal, create a distance data table and a reflection data table corresponding to the time information in the selected data signal; determine whether there are still unselected ones in the original signal Data signal; when there is a data signal that has not been selected in the original signal, select another data signal that has not been selected from the original signal; when there is no data signal in the original signal that has not been selected, Based on the information corresponding to the first time series, performing differential sampling on the information corresponding to the remaining time series to generate a differential data table, and after compressing the differential data table, upload the compressed differential data table, When there is no data signal that has not been selected in the original signal, discrete cosine conversion is performed on the distance data table and the reflection data table first, and then the corresponding time information corresponding to the time information in the first time sequence is used. Based on the distance data table and the reflection data table, perform differential sampling on the distance data table and the reflection data table corresponding to the time information corresponding to the rest of the time series to generate a sampled distance differential data table and the sampled A reflection difference data table, and the distance difference data table after sampling and the reflection difference data table after sampling are compressed and uploaded.

The vehicle network data sampling and conversion system includes a vehicle bus, a vehicle host and a network module.

The vehicle bus is connected to a controller area network and receives a raw data.

The vehicle-mounted host is connected to the vehicle bus to receive the original data and select one of the data signals from the original signal; wherein the vehicle-mounted host creates a data according to the selected data signal and determines whether the original signal There are also data signals that have not been selected. When there is a data signal that has not yet been selected in the original signal, the vehicle-mounted host selects another data signal that has not been selected from the original signal. When there is no data signal that has not been selected in the original signal, the vehicle-mounted host performs differential sampling on the information corresponding to the first time sequence to generate a differential data table based on the information corresponding to the first time sequence to generate a differential data table. Difference data sheet.

The network module is connected to the vehicle host, and after the vehicle host compresses the differential data table, the differential data table is uploaded to a cloud server through the network module.

Because the method and system for sampling and conversion of vehicle network data of the present invention compresses the data, the information corresponding to the first time sequence is used as a reference, differential sampling is performed on the information corresponding to the remaining time sequence, and then the difference is compressed and uploaded The data table, and the amount of information after differential sampling will be greatly reduced. In this way, it will effectively improve the data compression rate, reduce the complexity of pre-calculation, and avoid the delay of information transmission. In addition, the present invention can also be applied to optical radar data.

11‧‧‧ Car host

12‧‧‧Bus bus

13‧‧‧Network Module

14‧‧‧Controller Area Network

15‧‧‧ optical radar

20‧‧‧ cloud server

400‧‧‧Controller area network signal

401‧‧‧Data signal

4011‧‧‧Content signal

500‧‧‧Identification data sheet

600‧‧‧Content Data Sheet

600a‧‧‧content difference data table

900‧‧‧ optical radar signal

901‧‧‧Data signal

9011‧‧‧Content information

9012‧‧‧Direction content information

1000‧‧‧Distance data table

1100‧‧‧Reflectance data sheet

1000a, 1000b, 1000c ‧‧‧ distance difference data table converted into frequency domain

FIG. 1 is a schematic flowchart of a first preferred embodiment of a method for sampling and converting vehicle network data according to the present invention.

FIG. 2 is a schematic flowchart of a second preferred embodiment of a method for sampling and converting vehicle network data according to the present invention.

FIG. 3 is a block diagram of the vehicle network data sampling and conversion system of the present invention.

FIG. 4 is a schematic diagram of the format of the controller area network signal and content information of the second preferred embodiment of the vehicle network data sampling conversion method of the present invention.

FIG. 5 is a schematic diagram of the format of the identification data table of the second preferred embodiment of the vehicle network data sampling conversion method of the present invention.

FIG. 6 is a format diagram of a content data table of the second preferred embodiment of the vehicle network data sampling and conversion method of the present invention.

FIG. 7 is a format diagram of a content differential data table of the second preferred embodiment of the vehicle network data sampling conversion method of the present invention.

FIG. 8 is a schematic flowchart of a third preferred embodiment of a method for sampling and converting vehicle network data according to the present invention.

9 is a schematic diagram of the format of the optical radar signal, content information and direction angle content information of the third preferred embodiment of the vehicle network data sampling conversion method of the present invention.

10 is a schematic diagram of a format of a distance data table of a third preferred embodiment of a method for sampling and converting vehicle network data according to the present invention.

FIG. 11 is a schematic diagram of a reflection data table format of the third preferred embodiment of the vehicle network data sampling conversion method of the present invention.

12 is a schematic diagram of a format of a distance data table in the third preferred embodiment of the method for sampling and conversion of vehicle network data according to the present invention.

13 is a schematic diagram of the format of a distance difference data table after discrete cosine conversion according to the third preferred embodiment of the vehicle network data sampling conversion method of the present invention.

In the following, in conjunction with the drawings and preferred embodiments of the present invention, the technical means adopted by the present invention to achieve the intended purpose of the invention will be further described.

Referring to FIG. 1, the present invention is a vehicle network data sampling conversion method. A first preferred embodiment of the vehicle network data sampling conversion method includes the following steps: receiving an original signal (S101); Select one of the data signals from the original signal (S102); create a data table based on the selected data signal (S103); determine whether there is a data signal that has not been selected in the original signal (S104); when the original signal If there is a data signal that has not yet been selected, select another data signal that has not been selected from the original signal (S105); when there is no data signal that has not yet been selected in the original signal, a first After the information corresponding to the time series is used as a reference, after performing differential sampling on the information corresponding to the remaining time series to generate a differential data table (S106), the differential data table is compressed (S107), and the differential data table is uploaded (S108).

Since the data sampling and conversion method of the vehicle network of the present invention compresses the data, the information corresponding to the first time series is used as a reference, after performing differential sampling on the information corresponding to the remaining time series, and then compressing and uploading the differential data table The amount of information after differential sampling will be greatly reduced. In this way, the data compression rate can be effectively improved, the complexity of pre-calculation can be reduced, and the delay of information transmission can be avoided.

In the preferred embodiment, the on-vehicle network data sampling and conversion method of the present invention is mainly executed by an on-board host, which is used to compress the on-board data, increase the compression ratio of the on-board data, and increase the on-board host to upload the on-board data to the cloud server s speed. In addition, in the step of compressing the difference data table (S107), the difference data table is encoded by Huffman (Huffman), arithmetic (Arithmetic), Number (Symbol) coding or codebook (Codebook) coding and other technologies to compress data to encode the two-bit data of the differential data table.

For example, the original data is [ y ₁ y ₂ y ₃ … y _n ] ^T , where the information corresponding to the first time series is [ y ₁ ] ^T , and the data after differential sampling is [ y ₁ y ₂ - y ₁ y ₃ - y ₂ … y _n - y _{n -1} ] ^T , and because the amount of data y _t - y _{t -1} << y _t , the amount of data after differential sampling will be greatly reduced, thereby effectively improving the data The compression ratio reduces the amount of data and avoids the delay of information transmission. For example, the original data is [10000 10005 10006 10010 10012] ^T , and the data after differential sampling is [10000 5 1 4 2] ^T.

Please refer to FIG. 2, a second preferred embodiment of the vehicle network data sampling conversion method is the same as the first embodiment, but in this preferred embodiment, the original signal is a controller Area network (Controller Area Network) signal. And in the step of creating the data table based on the selected data signal (S103), it further includes the following sub-steps: decoding a piece of identification information in the selected data signal (S2031); judging whether the decoded identification information has Record in an identification data table (S2032); when the decoded identification information is recorded in the identification data table, select a content data table corresponding to the decoded identification information (S2033), and select the selected The time information and content information in the data signal are stored in the selected content data table (S2034); when the decoded identification information is not recorded in the identification data table, the decoded identification information is stored in In the identification data table, a new content data table corresponding to the decoded identification information is created, and after selecting the newly created content data table (S2035), the time information and content information in the selected data signal are stored in The selected content data table (S2034).

In addition, in the preferred embodiment, when there is no data signal that has not been selected in the original signal, the information corresponding to a first time sequence is used as a reference, and differential sampling is performed on the information corresponding to the remaining time sequence to generate After a difference data table (S106), a content difference data is generated After generating the content differential data table, the content differential data table is compressed (S207), and the identification data table and the content differential data table are uploaded (S208).

Referring to FIG. 3, the second preferred embodiment of the vehicle network data sampling and conversion method of the present invention is executed by a vehicle network data sampling and conversion system. The vehicle network data sampling and conversion system includes a vehicle host 11, A bus 12 for vehicles and a network module 13. The vehicle-mounted host 11 is connected to a controller area network 14 through the vehicle bus 12 to receive the controller area network signal. The in-vehicle host 11 is used to execute the in-vehicle network data sampling conversion method, and upload the differential data table to a cloud server 20 through the network module 13.

For example, please further refer to FIG. 4, the controller area network signal 400 includes a plurality of data signals 401, and each data signal includes a time information, an identification information, a length information and a content information 4011. And the content information 4011 is composed of a plurality of content information, for example, includes a 0th content information, a 1st content information, a 2nd content information, ..., a nth content information.

When the vehicle host 11 receives the controller local area network signal 400 and selects one of the data signals 401 from the controller local area network signal 400, the vehicle host 11 first decodes the identification information in the data signal 401, And judge whether the identification information is recorded in the identification data table. The identification data table 500, as shown in FIG. 5, records the content of each identification information ( CID ₀ , CID ₁ , CID ₂ , ...) and the content of the corresponding length information ( DLC ₀ , DLC ₁ , DLC ₂ ,). ..). The identification information represents the identification codes of each controller in each controller area network, that is, the vehicle-mounted host 11 determines which controller generated the next data. Therefore, when the vehicle-mounted host 11 judges that the identification information is recorded in the identification data table 500, it means that the identification information is generated by the controller that has been identified, and its corresponding data has been recorded. . Then, the vehicle-mounted host 11 selects the content data table corresponding to the identification information, and stores the time information and content information in the selected data signal in the selected content data table.

For example, when the vehicle-mounted host 11 determines that the identification information is CID ₀ , since CID ₀ has been recorded in the identification data table 500, the vehicle-mounted host 11 directly selects the content data table corresponding to the CID ₀ , such as As shown in FIG. 6, CID ₀ corresponding to the content information table 600 is already generated previously recorded information through a corresponding control CID _0, for example, the time information (Timestamp ₀₎ and the content information (data ₀₀ ~ data _07). Therefore, the vehicle-mounted host 11 can directly store the time information ( Timestamp _n ) and content information ( Data _{n 0} ~ Data _{n 7} ) corresponding to CID ₀ in the data signal 101 received this time into the content data table 600.

When the vehicle-mounted host 11 determines that the identification information is not recorded in the identification data table, it means that the identification information is generated by the controller that has not been identified, that is, no corresponding data has been recorded. . Therefore, the vehicle-mounted host 11 first stores the identification information and the length information in the selected data signal in the identification data table, such as CID _n and DLC _n , and creates a new content corresponding to the decoded identification information Data table 600, and after selecting the newly created content data table 600, a time information ( Timestamp ₀ ) and a content information ( Data ₀₀ ~ Data ₀₇ ) in the selected data signal are stored as the first data in the selected The content data table 600 corresponds to the newly created content data table 600 corresponding to CID _n .

Through the method of creating the content data tables 600 described above, the original data can be converted to a data space based on time, space, and data codes, thereby effectively reducing data disorder.

In addition, the data in the content data tables 600 are arranged in the order of time information. Because the data generated by each controller in the controller area network changes with time, such as the speed of the vehicle. The data recorded in the content data table 600 corresponds to the same controller. Therefore, the data generated by the same controller is arranged in chronological order according to the time information, and the data of each controller can be clearly recorded.

Further, when the vehicle-mounted host 11 performs differential sampling, the information corresponding to the first time series can be used as a reference, and the information corresponding to the remaining time series can be differentially recorded and then recorded again, thereby effectively reducing the amount of data. When the vehicle-mounted host 11 performs differential sampling, it is performed based on the content data table 600 shown in FIG. 6. For example, the data corresponding to the first time sequence in the content data table 600 shown in FIG. 6 is time information ( Timestamp ₀ ) and content information ( Data ₀₀ ~ Data ₀₇ ), so when performing differential sampling on the content data table 600, as shown in FIG. 7, the data corresponding to the first time sequence is first time information ( Timestamp ₀ ) And content information ( Data ₀₀ ~ Data ₀₇ ) are retained, and the remaining time sequence time information ( Timestamp ₁ ~ Timestamp ₃ ) and content information ( Data ₁₀ ~ Data ₃₇ ) are differentially sampled in order to generate a content differential data table 600a.

For example, the differential data of time information is: Δ t ₁ = Timestamp ₁ - Timestamp ₀ , Δ t ₂ = Timestamp ₂ - Timestamp ₁ , ..., and the differential data of content information is: Δ D ₁₀ = Data ₁₀ - Data ₀₀ , Δ D ₂₀ = Data ₂₀ - Data ₁₀ , ....

Please refer to FIG. 8, a third preferred embodiment of the vehicle network data sampling conversion method is the same as the first embodiment, but in this preferred embodiment, the original signal is an optical radar ( Light Detection And Ranging) signal. And in the step of creating the data table according to the selected data signal (S103), the method further includes the following sub-steps: according to the selected data signal, a distance data table corresponding to a time information in the selected data signal is created And a reflection data table (S8031).

In addition, when there is no data signal in the original signal that has not been selected, discrete cosine transform (Discrete Cosine Transform) (S806) is performed on the distance data table and the reflection data table, and then the first The distance data table and the reflection amount data table corresponding to the time information corresponding to the time series are used as a reference, and differential sampling is performed on the distance data table and the reflection amount data table corresponding to the time information corresponding to the remaining time series (S807) to generate A distance difference data table after sampling and a reflection difference data table after sampling, and the distance difference data table after sampling and the reflection difference data table after sampling (S808), (S809) are compressed and uploaded.

Referring to FIG. 3, the third preferred embodiment of the vehicle network data sampling and conversion method of the present invention is also executed by the vehicle network data sampling and conversion system, and the vehicle network data sampling and conversion system further includes an optical Radar 15. The optical radar 15 generates the optical radar signal for the vehicle-mounted host 11 to receive, that is, the original signal is the optical radar signal generated by the optical radar 15. In this preferred embodiment, the on-board host 11 is used to perform the third preferred embodiment of the on-vehicle network data sampling conversion method, and the distance differential data table after sampling and compression is sampled through the network module 13 And the sampled difference data table is uploaded to the cloud server 20.

For example, referring to FIG. 9, the optical radar signal 900 includes a plurality of data signals 901, and each data signal includes a time information, an identification information, a length information, and a content information 9011. And the content information 9011 is composed of a plurality of direction angle content information 9012, for example, including a 0th direction angle content information, a first direction angle content information, a second direction angle content information, ..., a first n Direction angle content information, etc. And each of the direction angle content information is composed of a flag information, a direction angle information, a complex distance information and a complex reflection amount information, for example, the 0th direction angle content information includes a flag information, a 0th direction angle Information, layer 0 distance information, layer 0 reflection information, ..., m layer distance information and m layer reflection information, etc.

When the car host 11 receives the optical radar signal 900, since each content information 9011 is composed of a plurality of directional angle content information 9012, the car host 11 first selects one of the content information 9011 to process, The distance data table 1000 shown in FIG. 10 and the reflection data table 1100 shown in FIG. 11 are generated.

Taking the distance data table 1000 as an example, the content information 9001 includes a plurality of direction angle content information 9012, and each direction angle content information 9012 represents data of different direction angles, and each direction angle content information 9012 and Contains the distance information and reflection information of each layer, so a single content information 9011 can create a complete distance data table and a complete reflection data table. For example, in the content information of the 0th direction angle, the distance information of the 0th layer is included. Information, layer 1 distance information, ..., m layer distance information, and layer 0 reflection information, layer 1 reflection information, ..., m layer reflection information.

Therefore, when creating the distance data table, the layer of the distance data is used as the vertical axis, and the direction angle of the distance data is used as the horizontal axis. For example, the 0th layer distance information in the 0th direction angle content information is recorded as D _0,0 in the distance data table 1000, and the 15th layer distance information in the 0th direction angle content information That is, it is recorded in the distance data table 1000 as D _15,0 , and the 15th layer distance information in the 364th direction angle content information is recorded in the distance data table 1000 as D _15,364 . In addition, since one piece of content information 9011 corresponds to only one piece of time information, each distance data table 1000 corresponds to one piece of time information t ₀ ~ t _{2, respectively} .

The reflection data table 1100 is created in the same way as the distance data table 1000, and will not be repeated here.

When all the data signals have been selected and their corresponding distance data tables 1000 and reflection data tables 1100 are established, the vehicle-mounted host 11 also executes the distance data tables 1000 and reflection data tables 1100 respectively. Discrete cosine conversion, which transforms time-domain data into frequency-domain data.

In the preferred embodiment, the discrete cosine transformation is calculated according to the following formula:

Where f ( i , j ) is each data in the distance data table 1000 or the reflection data table 1100 as input.

For example, please refer to FIG. 12 and FIG. 13, since the last data of the distance data table 1000 is D _15,364 , so N-1=15, M-1=364, and N=16, M=365, the After N and M are brought into the formula, each piece of data in the distance data table 1100a converted into the frequency domain can be obtained. For example, the last _{piece of} data F _{15,364 is}

In this way, each piece of data F _0,0 ~ F _15,364 of the distance data table 1100a converted into the frequency domain can be calculated _separately .

Since one piece of content information 9011 corresponds to only one piece of time information, and each piece of content information 9011 can generate a distance data table 1000 and a reflection data table 1100, respectively, as shown in FIGS. 10 and 11, each distance data table 1000 is Each corresponds to a piece of time information t ₀ ~ t ₂ , and each reflection data table 1100 also corresponds to a piece of time information t ₀ ~ t ₂ .

And since each distance data table 1000 corresponds to a piece of time information t ₀ ~ t ₂ , the distance data tables 1000a, 1000b, and 1000c converted into the frequency domain also correspond to a piece of time information t ₀ ~ t _{2, respectively} . As shown in FIG. 13, when the vehicle-mounted host 11 performs differential sampling, the vehicle-mounted host 11 is based on the distance data table 1000a converted into the frequency domain corresponding to the time information t ₀ corresponding to the first time sequence, and the remaining time The time information t ₁ ~ t ₂ corresponding to the sequence corresponds to the distance data tables 1000b and 1000c converted into the frequency domain, and then re-recorded to generate the distance difference data tables 1000b and 1000c converted into the frequency domain, thereby effectively reducing the data the amount.

For example, the data Δ Fa _0,0 ~Δ Fa _{N -1, M -1} corresponding to the time difference t ₁ converted into the frequency domain distance difference data table 1000b is: Δ Fa _0,0 = Fb _0,0- Fa _0,0 , Δ Fa _0,1 = Fb _0,1 - Fa _0,1 ,...

In addition, when the vehicle-mounted host 11 performs differential sampling, the vehicle-mounted host 11 can further determine whether the data in the distance differential data table 1000a after the differential sampling and the data in the reflection differential data table after the differential sampling are greater than a critical value. And when the vehicle-mounted host 11 judges the distance difference after differential sampling When the data in the sub-data table 1000a and the differential data table after the differential sampling is less than a critical value, the vehicle-mounted host 11 will record 0 as the critical value, thereby further reducing the amount of data.

Since the meaning of each content information 9011 in the optical radar signal 900 is a plane scanned by light, it can accept a certain degree of distortion and will not affect subsequent use. That is to say, when the vehicle-mounted host 11 records less than the critical value as 0, a small part of the data representing the plane will be deleted, and the data less than the critical value may represent noise or some tiny objects. Therefore, when the vehicle-mounted host 11 records less than the critical value as 0, the amount of data can be further reduced without affecting the subsequent use of the data.

In summary, the present invention is mainly through differential sampling of correlation data, or representing the data on a frequency domain basis, and detailed sampling of important information is retained, so that there is no need to record or compress the original data, thereby Improve the vehicle data sampling conversion rate and reduce the amount of data.

The above is only the preferred embodiment of the present invention, and does not limit the present invention in any form. Although the present invention has been disclosed in the preferred embodiment as above, it is not intended to limit the present invention. Anyone who is familiar with this technology Personnel, within the scope of not departing from the technical solution of the present invention, when the technical contents disclosed above can be used to make some modifications or modifications to equivalent embodiments of equivalent changes, but any content that does not depart from the technical solution of the present invention, according to the present invention Technical essence Any simple modifications, equivalent changes and modifications to the above embodiments still fall within the scope of the technical solution of the present invention.

Claims (2)

A vehicle network data sampling conversion method includes the following steps: receiving an original signal, wherein the original signal is an optical radar signal; selecting one of the data signals from the original signal; creating a data table based on the selected data signal , And further includes the following sub-steps: according to the selected data signal, create a distance data table and a reflection data table corresponding to the time information in the selected data signal; determine whether the original signal has not yet been selected Data signal; when there is a data signal that has not been selected in the original signal, select another data signal that has not been selected from the original signal; when there is no data signal in the original signal that has not been selected , Based on the information corresponding to a first time series, perform differential sampling on the information corresponding to the remaining time series to generate a differential data table, compress the differential data table, and upload the compressed differential data table, where the When there is no data signal that has not been selected in the original signal, discrete cosine conversion is performed on the distance data table and the reflection data table first, and then the distance data table corresponding to the time information corresponding to the first time sequence And the reflection data table as a reference, perform differential sampling on the distance data table and the reflection data table corresponding to the time information corresponding to the remaining time series to generate a sampled distance differential data table and a sampled reflection Difference data table, and compress and upload the sampled distance difference data table and the sampled reflection difference data table. An on-vehicle network data sampling and conversion system includes: a bus for vehicle connected to a controller area network to receive a raw data; an on-board host connected to the bus for vehicle to receive the raw data, And select one of the data signals from the original signal; wherein the vehicle-mounted host creates a data table based on the selected data signal, and judges whether there is a data signal that has not been selected in the original signal; When there is a data signal that has not been selected in the original signal, the vehicle host selects another data signal that has not been selected from the original signal; where there is no data signal that has not been selected in the original signal At this time, the vehicle-mounted host uses the information corresponding to a first time sequence as a reference, performs differential sampling on the information corresponding to the remaining time sequence to generate a differential data table, and compresses the differential data table; a network module is connected to the Onboard host; wherein after compressing the differential data table, the onboard host uploads the compressed differential data table to a cloud server through the network module; an optical radar is connected to the onboard host; wherein the original signal is An optical radar signal generated by the optical radar; wherein when the vehicle host creates the data table based on the selected data signal, the vehicle host further creates a time information corresponding to the selected data signal according to the selected data signal A distance data table and a reflection data table; where there is no data signal that has not been selected in the original signal, the vehicle-mounted host first performs discrete cosine conversion on the distance data table and the reflection data table, Then, based on the distance data table and the reflection data table corresponding to the time information corresponding to the first time series, the distance data table and the reflection data table corresponding to the time information corresponding to the remaining time series are differentially sampled To generate a sampled distance difference data table and a sampled reflection difference data table, and after compressing the sampled distance difference data table and the sampled reflection difference data table, through the network module Upload the compressed differential distance data table and the reflected differential data table after sampling to the cloud server.

Images