Driving device controller, driving system, data processing method and intelligent device

Abstract

The present application relates to the field of intelligent driving, and specifically provides a driving device controller, a driving system, a data processing method and an intelligent device, aiming to solve the problems of high cost of laser radar main control hardware and low software iteration efficiency. The driving device control of the present application is connected to the laser radar system for communication, and receives the original laser radar point cloud data sent by the laser radar system; the driving device controller includes a first point cloud computing module and a data processing module; the first point cloud computing module is configured to perform point cloud computing on the original laser radar point cloud data to obtain first point cloud data; the data processing module is configured to process the first point cloud data to obtain corresponding scene information. Through the above implementation, the original sampling signal of the laser radar system can be transmitted to the driving device controller for calculation, which reduces the memory and CPU resource consumption of the laser radar components, and gives full play to the rapid iteration capability of the driving device controller data closed loop and software.

Description

Technical Field

[0001] The present application relates to the field of intelligent driving technology, and specifically to a driving device controller, a driving system, a data processing method and an intelligent device. Background Art

[0002] During autonomous driving or assisted driving, traditional distributed computing sensors rely on high-cost main control chips and peripheral hardware. Each sensor acts as an independent electronic control unit (ECU) to complete its own signal processing and control.

[0003] Due to the high sampling frequency of the original signal of the LiDAR point cloud and the large amount of single-frame point cloud data, the traditional LiDAR hardware and software architecture needs to be responsible for the control, communication, fault diagnosis and other parts of each LiDAR submodule, and needs to process the huge and high-frequency LiDAR original signal, which often requires a high-performance main control chip and a large memory. At the same time, under the traditional distributed architecture, the suppliers of each sensor ECU and driving system independently develop, test and upgrade the software, which has a long software development cycle and low iteration efficiency. Summary of the invention

[0004] In order to overcome the above-mentioned defects, the present application is proposed to solve or at least partially solve the current technical problems of high cost of laser radar main control hardware, low software iteration efficiency and difficulty in optimizing system performance. In a first aspect, a driving device controller is provided, the driving device controller being communicatively connected to a laser radar system, the driving device controller receiving raw laser radar point cloud data of a target scene sent by the laser radar system;

[0005] The driving device controller includes a first point cloud computing module and a data processing module;

[0006] The first point cloud computing module is configured to perform point cloud computing on the original laser radar point cloud data to obtain first point cloud data;

[0007] The data processing module is configured to process the first point cloud data to obtain scene information corresponding to the original laser radar point cloud data.

[0008] In one technical solution of the above driving device controller, the driving device controller further includes a second point cloud computing module;

[0009] The second point cloud computing module is configured to perform point cloud computing on the original laser radar point cloud data to obtain second point cloud data;

[0010] The data processing module is further configured to process the second point cloud data to obtain scene information corresponding to the original laser radar point cloud data;

[0011] The second point cloud computing module and the first point cloud computing module perform point cloud computing on the original lidar point cloud data based on different algorithms.

[0012] In a technical solution of the above driving device controller, the first point cloud computing module is configured to perform at least one of distance calculation, angle calculation, and reflectivity calculation on the original laser radar point cloud data based on a first preset algorithm to obtain the first point cloud data;

[0013] The second point cloud computing module is configured to perform at least one of the distance calculation, the angle calculation, and the reflectivity calculation on the original lidar point cloud data based on a second preset algorithm to obtain the second point cloud data.

[0014] In a technical solution of the above driving device controller, the driving device controller further includes a noise filtering module, and the noise filtering module is configured to perform noise filtering processing on the first point cloud data and / or the second point cloud data;

[0015] The data processing module includes a data parsing submodule, a coordinate conversion submodule and a perception calculation submodule.

[0016] In a second aspect, a driving system is provided, the driving system comprising a laser radar system and a driving device controller as described in any one of the technical solutions of the above-mentioned driving system, the laser radar system being communicatively connected with the driving device controller; the laser radar system comprising a main control module, a transmitting module, a receiving module and a scanning module;

[0017] Wherein, the main control module includes a communication submodule;

[0018] The communication submodule is configured to send the original lidar point cloud data of the target scene acquired by the lidar system to the driving device controller.

[0019] In a technical solution of the above driving system, the laser radar system further includes a power module; the main control module further includes a system control submodule and a storage submodule;

[0020] Wherein, the system control submodule is configured to at least control the operation of the transmitting module, the receiving module and the scanning module;

[0021] The storage submodule is configured to store control parameters of the lidar system and the original lidar point cloud data.

[0022] In a technical solution of the above driving system, the main control module further includes a signal processing submodule;

[0023] The signal processing submodule is configured to filter the raw laser radar point cloud data before sending the raw laser radar point cloud data to the driving device controller.

[0024] In a third aspect, a data processing method is provided, which is applied to a driving device controller, and the method comprises:

[0025] Obtain the original LiDAR point cloud data of the target scene sent by the LiDAR system;

[0026] Performing a first point cloud computing on the original laser radar point cloud data to obtain first point cloud data;

[0027] The first point cloud data is processed to obtain scene information corresponding to the original laser radar point cloud data.

[0028] In a technical solution of the above data processing method, the method further includes:

[0029] Perform a second point cloud computing on the original laser radar data to obtain second point cloud data;

[0030] The second point cloud data is processed to obtain scene information corresponding to the original laser radar point cloud data.

[0031] In a technical solution of the above data processing method, performing a first point cloud computing on the original laser radar point cloud data includes:

[0032] Perform at least one of distance calculation, angle calculation, and reflectivity calculation on the original laser radar data based on a first preset algorithm;

[0033] The performing second point cloud computing on the original laser radar point cloud data comprises:

[0034] Based on a second preset algorithm, at least one of the distance calculation, the angle calculation, and the reflectivity calculation is performed on the original laser radar data.

[0035] In a technical solution of the above data processing method, the processing of the first point cloud data or the processing of the second point cloud data includes:

[0036] Perform data analysis, coordinate transformation, and perception calculation on the first point cloud data or the second point cloud data.

[0037] In a technical solution of the above data processing method, the method further includes:

[0038] Performing abnormality diagnosis on the first point cloud data and / or the second point cloud data;

[0039] When the first point cloud data or the second point cloud data is abnormal, obtaining original laser radar point cloud data corresponding to the first point cloud data or the second point cloud data;

[0040] The abnormality cause is obtained based on the original laser radar point cloud data.

[0041] In a technical solution of the above data processing method, the method further includes:

[0042] comparing the first point cloud data and the second point cloud data;

[0043] The first preset algorithm or the second preset algorithm is optimized based on the comparison result.

[0044] In a technical solution of the above data processing method, the method further includes:

[0045] Perform noise filtering on the first point cloud data and / or the second point cloud data.

[0046] In a fourth aspect, a smart device is provided, which includes a smart device body and a driving system described in the technical solution of the above-mentioned driving system.

[0047] Solution 1. A driving device controller, characterized in that the driving device controller is communicatively connected to a laser radar system, and the driving device controller receives original laser radar point cloud data of a target scene sent by the laser radar system;

[0048] The driving device controller includes a first point cloud computing module and a data processing module;

[0049] The first point cloud computing module is configured to perform point cloud computing on the original laser radar point cloud data to obtain first point cloud data;

[0050] The data processing module is configured to process the first point cloud data to obtain scene information corresponding to the original laser radar point cloud data.

[0051] Solution 2. The driving device controller according to Solution 1, characterized in that the driving device controller also includes a second point cloud computing module;

[0052] The second point cloud computing module is configured to perform point cloud computing on the original laser radar point cloud data to obtain second point cloud data;

[0053] The data processing module is further configured to process the second point cloud data to obtain scene information corresponding to the original laser radar point cloud data;

[0054] The second point cloud computing module and the first point cloud computing module perform point cloud computing on the original lidar point cloud data based on different algorithms.

[0055] Solution 3. The driving device controller according to Solution 2 is characterized in that the first point cloud computing module is configured to perform at least one of distance calculation, angle calculation, and reflectivity calculation on the original laser radar point cloud data based on a first preset algorithm to obtain the first point cloud data;

[0056] The second point cloud computing module is configured to perform at least one of the distance calculation, the angle calculation, and the reflectivity calculation on the original lidar point cloud data based on a second preset algorithm to obtain the second point cloud data.

[0057] Solution 4. The driving device controller according to Solution 2 is characterized in that the driving device controller further comprises a noise filtering module, and the noise filtering module is configured to perform noise filtering processing on the first point cloud data and / or the second point cloud data;

[0058] The data processing module includes a data parsing submodule, a coordinate conversion submodule and a perception calculation submodule.

[0059] Solution 5. A driving system, characterized in that the driving system comprises a laser radar system and a driving device controller according to any one of claims 1 to 4, wherein the laser radar system is communicatively connected with the driving device controller; the laser radar system comprises a main control module, a transmitting module, a receiving module and a scanning module;

[0060] Wherein, the main control module includes a communication submodule;

[0061] The communication submodule is configured to send the original lidar point cloud data of the target scene acquired by the lidar system to the driving device controller.

[0062] Solution 6. The driving system according to Solution 5 is characterized in that the laser radar system further includes a power module; the main control module further includes a system control submodule and a storage submodule;

[0063] Wherein, the system control submodule is configured to at least control the operation of the transmitting module, the receiving module and the scanning module;

[0064] The storage submodule is configured to store control parameters of the lidar system and the original lidar point cloud data.

[0065] Solution 7. The driving system according to Solution 6 is characterized in that the main control module also includes a signal processing submodule;

[0066] The signal processing submodule is configured to filter the raw laser radar point cloud data before sending the raw laser radar point cloud data to the driving device controller.

[0067] Solution 8. A data processing method, applied to a driving device controller, characterized in that the method comprises:

[0068] Obtain the original LiDAR point cloud data of the target scene sent by the LiDAR system;

[0069] Performing a first point cloud computing on the original laser radar point cloud data to obtain first point cloud data;

[0070] The first point cloud data is processed to obtain scene information corresponding to the original laser radar point cloud data.

[0071] Solution 9. The data processing method according to Solution 8, characterized in that the method further comprises:

[0072] Perform a second point cloud computing on the original laser radar data to obtain second point cloud data;

[0073] The second point cloud data is processed to obtain scene information corresponding to the original laser radar point cloud data.

[0074] Solution 10. The data processing method according to Solution 9 is characterized in that performing a first point cloud computing on the original laser radar point cloud data comprises:

[0075] Perform at least one of distance calculation, angle calculation, and reflectivity calculation on the original laser radar data based on a first preset algorithm;

[0076] The performing second point cloud computing on the original laser radar point cloud data comprises:

[0077] Based on a second preset algorithm, at least one of the distance calculation, the angle calculation, and the reflectivity calculation is performed on the original laser radar data.

[0078] Solution 11. The data processing method according to Solution 9 is characterized in that the processing of the first point cloud data or the processing of the second point cloud data comprises:

[0079] Perform data analysis, coordinate transformation, and perception calculation on the first point cloud data or the second point cloud data.

[0080] Solution 12. The data processing method according to Solution 9, characterized in that the method further comprises:

[0081] Performing abnormality diagnosis on the first point cloud data and / or the second point cloud data;

[0082] When the first point cloud data or the second point cloud data is abnormal, obtaining original laser radar point cloud data corresponding to the first point cloud data or the second point cloud data;

[0083] The abnormality cause is obtained based on the original laser radar point cloud data.

[0084] Solution 13. The data processing method according to Solution 10, characterized in that the method further comprises:

[0085] comparing the first point cloud data and the second point cloud data;

[0086] The first preset algorithm or the second preset algorithm is optimized based on the comparison result.

[0087] Solution 14. The data processing method according to Solution 9, characterized in that the method further comprises:

[0088] Perform noise filtering on the first point cloud data and / or the second point cloud data.

[0089] Solution 15. An intelligent device, characterized in that the intelligent device includes an intelligent device body and a driving system described in any one of Solutions 5 to 7.

[0090] The above one or more technical solutions of the present application have at least one or more of the following beneficial effects:

[0091] In the implementation of the technical solution provided by this application, the driving device controller is provided with a point cloud computing module and a data processing module, which can directly transmit the original sampling signal of the laser radar system to the driving device controller for calculation, without the need to set up a high-performance main control chip and a large memory in the laser radar system, reducing the memory and CPU resource consumption of the laser radar components, reducing the main control cost and the volume of the main control board of the laser radar system, and realizing a lightweight laser radar hardware and software architecture. In addition, by embedding the point cloud computing part of the laser radar into the software module of the driving device controller, the signal processing algorithm can be moved up, and through the reasonable allocation of the central integrated computing power, the overall cost of the laser radar system and the cost of the vehicle-mounted intelligent perception system can be effectively reduced, and system resources can be saved. At the same time, the driving device controller can give full play to the rapid iteration capability of the data closed loop and software, and improve the performance of data processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0092] The disclosure of the present application will become easier to understand with reference to the accompanying drawings. It is easy for those skilled in the art to understand that these drawings are only for illustrative purposes and are not intended to limit the scope of protection of the present application. Among them:

[0093] Figure 1 It is a schematic diagram of the traditional LiDAR system architecture;

[0094] Figure 2 It is a schematic diagram of the traditional driving system architecture;

[0095] Figure 3 is a schematic diagram of the main structure of a laser radar system according to an embodiment of the present application;

[0096] Figure 4 A schematic diagram of the main structure of a laser radar system according to another embodiment of the present application;

[0097] Figure 5 A schematic diagram of the main structure of a driving device controller according to an embodiment of the present application;

[0098] Figure 6 A schematic diagram of the main structure of a driving device controller according to another embodiment of the present application;

[0099] Figure 7 A schematic diagram of the main structure of a driving system according to an embodiment of the present application;

[0100] Figure 8 It is a flowchart of the main steps of a data processing method according to an embodiment of the present application.

[0101] Reference numerals:

[0102] 31: main control module; 32: transmitting module; 33: receiving module; 34: scanning module 34; 35: power module 35; 311: communication submodule; 312: system control submodule; 313: storage submodule; 51: first point cloud computing module; 52: data processing module; 53: second point cloud computing module; 54: noise filtering module; 71: lidar system; 72: driving device controller; 721: first point cloud computing module; 722: data processing module; 723: second point cloud computing module; 724: noise filtering module. DETAILED DESCRIPTION

[0103] Some embodiments of the present application are described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principles of the present application and are not intended to limit the protection scope of the present application.

[0104] In the description of this application, "module" may include hardware, software or a combination of the two. A module may include hardware circuits, various suitable sensors, communication ports, and memories. It may also include software parts, such as program codes, or it may be a combination of software and hardware.

[0105] The term "A and / or B" means all possible combinations of A and B, such as only A, only B, or A and B. The term "at least one of A or B" or "at least one of A and B" has a similar meaning to "A and / or B" and may include only A, only B, or A and B. The singular forms "a", "the" and "the" may also include plural forms.

[0106] Currently, the driving device controller, i.e., the ECU of the driving device's Advanced Driver-Assistance Systems (ADAS) or Autonomous Driving (AD), is usually designed and developed independently from the ECU of the LiDAR system.

[0107] See attached Figure 1 , Figure 1 This is a schematic diagram of the traditional LiDAR system architecture. Figure 1 As shown, the laser radar system needs to complete the entire process of laser physical signal transmission and reception, photoelectric signal conversion, electrical signal processing and point cloud computing through a transmitting module, a receiving module, a scanning module, a control module and a power module.

[0108] Due to the high sampling frequency of the original signal of the lidar point cloud and the large amount of single-frame point cloud data, the traditional lidar system is responsible for the control, communication, fault diagnosis and other parts of each lidar sub-module, and needs to process the huge and high-frequency lidar original signal. The control module often requires a high-performance main control chip (Central Processing Unit, CPU) and a large memory that can store a large amount of point cloud cache data.

[0109] For further information, see Appendix Figure 2 , Figure 2 This is a schematic diagram of the traditional driving system architecture. Figure 2 As shown in the figure, after completing the whole process of signal transmission and reception, photoelectric signal conversion, electrical signal processing and point cloud computing, the LiDAR system transmits the LiDAR point cloud data with spatial coordinate information to the ADAS / AD ECU, and the ADAS / AD ECU parses the LiDAR point cloud data with spatial coordinate information.

[0110] exist Figure 2In the traditional distributed architecture shown in the figure, each ECU supplier independently develops, tests, and upgrades software, resulting in a long software development cycle and low iteration efficiency.

[0111] In order to solve the above problems, the present application provides a driving device controller, a driving system, a data processing method and an intelligent device.

[0112] Specifically, this application embeds the point cloud computing part of the lidar system into the software module of the driving device controller, moves the signal processing algorithm upward, and effectively reduces the overall cost of the lidar system and the cost of the on-board intelligent perception system and saves system resources through the reasonable allocation of central integrated computing power.

[0113] See attached Figure 3 , Figure 3 1 is a schematic diagram of the main structure of a laser radar system according to an embodiment of the present application. The laser radar system in the embodiment of the present application is connected to the driving device controller in communication. Figure 3 As shown, the laser radar system mainly includes a main control module 31, a transmitting module 32, a receiving module 33 and a scanning module 34.

[0114] The main control module 31 includes a communication submodule 311. The communication submodule 311 is configured to send the original laser radar point cloud data acquired by the laser radar system to the driving device controller, so that the driving device controller processes the original laser radar point cloud data and controls the driving device to perform corresponding operations based on the processed data.

[0115] Specifically, the communication submodule 311 includes a communication interface (such as an Ethernet interface, a CAN interface, etc.), which is a channel for the laser radar system to exchange data and information with external devices. It can transmit the data acquired by the laser radar system to other systems, such as the vehicle's automatic driving system, the control center of an industrial robot, etc., and can also receive control instructions from external devices.

[0116] The transmitting module 32 is the signal source of the laser radar system, which is configured to transmit laser pulse signals to the target area to detect the position, distance and other information of the target object.

[0117] The receiving module 33 is configured to receive the laser signal reflected from the target object, convert the reflected laser signal into an electrical signal, and perform signal amplification and other processing on the electrical signal.

[0118] The scanning module 34 is configured to control the scanning direction and scanning pattern of the laser beam so that the laser beam emitted by the laser radar system can scan the target area in a predetermined manner.

[0119] The above is Figure 3 Illustration of the lidar system shown.

[0120] Further, in some embodiments, see the attached Figure 4 , Figure 4 A schematic diagram of the main structure of a laser radar system according to another embodiment of the present application. Figure 4 As shown, the laser radar system may include not only a main control module 31, a transmitting module 32, a receiving module 33 and a scanning module 34, but also a power module 35. In addition, the main control module 31 may also include a system control submodule 312 and a storage submodule 313.

[0121] Among them, the power module 35 is configured to provide a stable power supply to each module and component within the laser radar system.

[0122] It should be noted that the LiDAR system can be set Figure 4 The power module 35 shown in the figure provides a stable power supply to various modules and components inside it through the power module 35. The power module may also not be set, and a stable power supply may be provided to various modules and components inside it through an external power supply. This is not limited here.

[0123] The system control submodule 312 in the main control module 31 is configured to at least control the operation of the transmitting module, the receiving module and the scanning module.

[0124] For example, the system control submodule 312 can control the transmission frequency and power of the transmission module, the scanning mode and speed of the scanning module, and manage the work of the receiving module and the signal processing module.

[0125] In some embodiments, the system control submodule 312 can also convert the electrical signal of the receiving module into a digital signal through an analog to digital converter (ADC) to achieve a sampling function. In addition, the system control submodule 312 can also adjust and optimize the signal, and perform pulse selection on multiple pulse signals involved in the transmission and reception process to obtain a valid pulse signal, etc., which is not limited here.

[0126] The storage submodule 313 in the main control module 31 is configured to store control parameters of the laser radar system and original laser radar point cloud data.

[0127] Specifically, the storage submodule 313 may include a random access memory (RAM) and a read-only memory (ROM). The RAM is mainly used to store temporary data during the operation of the system, including the original laser radar point cloud data received by the laser radar system through the receiving module before being sent to the driving device controller, as well as some intermediate data, etc. The ROM is used to store the control parameters of the laser radar system, including firmware program parameters, calibration parameters, etc.

[0128] This application Figure 4 The storage submodule 313 shown needs to store less data, so only a small memory capacity is required to meet actual application requirements. Compared with the memory of the traditional laser radar system, the main control cost and the size of the main control board are reduced, and it is more convenient in the installation, layout and subsequent maintenance of the laser radar equipment.

[0129] Furthermore, in some embodiments, the main control module 31 may further include a signal processing submodule. The signal processing submodule may be configured to filter the raw laser radar point cloud data before sending the raw laser radar point cloud data to the driving device controller.

[0130] Specifically, the signal processing submodule can filter out those point cloud data with stronger signals from a large amount of raw data, streamline the original lidar point cloud data, and send the filtered original lidar point cloud data to the driving device controller, thereby reducing the bandwidth pressure during data transmission and reducing the time required for data transmission.

[0131] Through the above implementation, the original sampling signal of the lidar system can be directly transmitted to the driving device controller for calculation. There is no need to set up a high-performance main control chip and a large memory in the lidar system, which reduces the memory and CPU resource consumption of the lidar components, reduces the main control cost and main control board size of the lidar system, and realizes a lightweight lidar hardware and software architecture.

[0132] Furthermore, the present application also provides a driving device controller. In order to transmit the original sampling signal of the laser radar system directly to the driving device controller for calculation, the driving device controller is communicatively connected with the laser radar system, so that the driving device controller receives the original laser radar point cloud data of the target scene sent by the laser radar system, and embeds the point cloud computing part of the laser radar system into the software module, which is deployed in the driving device controller as an independent application software for operation.

[0133] For details, see the attached Figure 5 , Figure 5A schematic diagram of the main structure of a driving device controller according to an embodiment of the present application. Figure 5 As shown, the driving device controller includes a first point cloud computing module 51 and a data processing module 52 .

[0134] The first point cloud computing module 51 is configured to perform point cloud computing on the original laser radar point cloud data to obtain first point cloud data.

[0135] Specifically, the first point cloud computing module 51 can perform at least one of distance calculation, angle calculation, and reflectivity calculation on the original laser radar data based on a first preset algorithm to obtain first point cloud data.

[0136] For example, the distance of the original lidar data is calculated through the time-of-flight method, triangulation method, etc., the angle of the original lidar data is calculated through the mechanical scanning angle measurement method, optical phased array angle calculation method, etc., and the reflectivity of the original lidar data is calculated through the photoelectric detector signal intensity analysis method, multi-wavelength lidar reflectivity measurement method, etc., to obtain a lidar point cloud (i.e., the first point cloud data) with spatial coordinate information.

[0137] The data processing module 52 is configured to process the first point cloud data to obtain scene information corresponding to the original laser radar point cloud data.

[0138] Specifically, the data processing module 52 may include a data parsing submodule, a coordinate conversion submodule and a perception computing submodule.

[0139] Among them, the data analysis submodule can analyze the first point cloud data to obtain the coordinate information, angle information and reflectivity information of the laser point; the coordinate conversion submodule can convert the coordinates in the first point cloud data into the driving device coordinate system or other unified coordinate systems that are more convenient for the driving device to analyze and judge according to certain rules; the perception calculation submodule can analyze the first point cloud data, identify the various target objects in the actual scene represented by the point cloud data and their mutual relationships, and use perception algorithms, such as clustering algorithms, to cluster point clouds with similar distances and features into point sets that may represent a certain object, and then use machine learning models or rule-based recognition methods to determine whether these point sets are vehicles, pedestrians, traffic signs or other obstacles.

[0140] Through the sequential processing and collaborative work of the three sub-modules in the data processing module, the scene information corresponding to the original lidar point cloud data can be obtained, such as which target objects are around the driving device, where they are, what motion state they are in, etc., so that the driving system can control the driving device to perform corresponding operations based on this scene information.

[0141] The above is Figure 5 Description of the driving device controls shown.

[0142] Further, in some embodiments, see the attached Figure 6 , Figure 6 A schematic diagram of the main structure of a driving device controller according to another embodiment of the present application. Figure 6 As shown, the driving device controller may include, in addition to the first point cloud computing module 51 and the data processing module 52 , a second point cloud computing module 53 and a noise filtering module 54 .

[0143] The second point cloud computing module 53 is configured to perform point cloud computing on the original laser radar point cloud data to obtain second point cloud data.

[0144] It should be pointed out that the second point cloud computing module 53 and the first point cloud computing module 51 perform point cloud computing on the original laser radar point cloud data based on different algorithms.

[0145] Specifically, the second point cloud computing module 53 can perform at least one of distance calculation, angle calculation, and reflectivity calculation on the original lidar data based on a second preset algorithm to obtain second point cloud data.

[0146] For example, the first point cloud computing module 51 can perform distance calculation on the original laser radar data through the time of flight method, perform angle calculation on the original laser radar data through the optical phased array angle calculation method, and perform reflectivity calculation on the original laser radar data through the photoelectric detector signal intensity analysis method, so as to obtain a laser radar point cloud with spatial coordinate information (i.e., the first point cloud data); the second point cloud computing module 53 can perform distance calculation on the original laser radar data through the triangulation method, perform angle calculation on the original laser radar data through the mechanical scanning angle measurement method, and perform reflectivity calculation on the original laser radar data through the multi-wavelength laser radar reflectivity measurement method, so as to obtain a laser radar point cloud with spatial coordinate information (i.e., the second point cloud data).

[0147] The noise filtering module 54 is configured to perform noise filtering processing on the first point cloud data and / or the second point cloud data.

[0148] Specifically, the noise filtering module 54 can filter out noise on the lidar point cloud with spatial coordinate information (i.e., the first point cloud data output by the first point cloud computing module 51, and / or the second point cloud data output by the second point cloud computing module 53), reduce or eliminate the noise components existing in the lidar point cloud data, improve the data quality, and facilitate the subsequent data processing module 52 to perform data processing.

[0149] Furthermore, the data processing module 52 is also configured to process the second point cloud data to obtain scene information corresponding to the original laser radar point cloud data.

[0150] That is, the data processing module 52 can process the first point cloud data and the second point cloud data after noise is filtered out to obtain scene information corresponding to the original lidar point cloud data, so that the driving system controls the driving device to perform corresponding operations based on the scene information corresponding to the original lidar point cloud data.

[0151] Through the above implementation method, the point cloud computing part of the lidar is embedded in the software module of the driving device controller, which can move the signal processing algorithm upward. Through the reasonable allocation of the central integrated computing power, the overall cost of the lidar system and the cost of the vehicle-mounted intelligent perception system can be effectively reduced, and system resources can be saved. At the same time, the driving device controller can give full play to the rapid iteration capability of the data closed loop and the software, thereby improving the performance of data processing.

[0152] The above is a description of the driving device controller provided in this application.

[0153] Furthermore, the present application also provides a driving system.

[0154] The driving system includes a laser radar system and a driving device controller, and the laser radar system is communicatively connected with the driving device controller.

[0155] The laser radar system is the laser radar system described in any one of the embodiments of the above laser radar system, that is, Figure 3-Figure 4 The lidar system shown.

[0156] The driving device controller is the driving device controller described in any one of the above driving device controller embodiments, that is, Figure 5-Figure 6 The driving device controller is shown.

[0157] See attached Figure 7 , Figure 7 A schematic diagram of the main structure of a driving system according to an embodiment of the present application. Figure 7 As shown, the driving system includes a laser radar system 71 and a driving device controller 72 , and the laser radar system 71 is communicatively connected with the driving device controller 72 .

[0158] The driving device controller 72 includes a first point cloud computing module 721 , a data processing module 722 , a second point cloud computing module 723 and a noise filtering module 724 .

[0159] Furthermore, the present application also provides a data processing method applied to a driving device controller.

[0160] See attached Figure 8, Figure 8 FIG. 1 is a flow chart of the main steps of a data processing method according to an embodiment of the present application. Figure 8 As shown, the data processing method in the embodiment of the present application mainly includes the following steps S801 to S803.

[0161] Step S801: Acquire original laser radar point cloud data of the target scene sent by the laser radar system;

[0162] The target scene may be a scene within a preset range around the driving device. After the laser radar system obtains the raw laser radar point cloud data, the raw laser radar point cloud data may be directly transmitted to the driving device controller via a communication network such as Ethernet or CAN.

[0163] Step S802: Perform a first point cloud computing on the original laser radar point cloud data to obtain first point cloud data;

[0164] Specifically, the driving device controller can perform a first point cloud computing on the original lidar point cloud data through a first point cloud computing module. The first point cloud computing module is provided with a first preset algorithm, which can perform at least one of distance calculation, angle calculation, and reflectivity calculation on the original lidar data to obtain a lidar point cloud with spatial coordinate information (i.e., the first point cloud data).

[0165] For example, the distance of the original lidar data is calculated by the time-of-flight method, triangulation method, etc., the angle of the original lidar data is calculated by the mechanical scanning angle measurement method, optical phased array angle calculation method, etc., and the reflectivity of the original lidar data is calculated by the photoelectric detector signal intensity analysis method, multi-wavelength lidar reflectivity measurement method, etc., without limitation here.

[0166] Step S803: Process the first point cloud data to obtain scene information corresponding to the original laser radar point cloud data.

[0167] Specifically, based on the data analysis submodule, coordinate conversion submodule and perception calculation submodule in the data processing module of the driving device controller, the first point cloud data can be analyzed, coordinate converted and perceived calculated to obtain scene information corresponding to the original lidar point cloud data, such as which target objects are around the driving device, where they are, what motion states they are in, etc., so that the driving system can control the driving device to perform corresponding operations based on the scene information.

[0168] The above data processing method is further explained below.

[0169] In some embodiments, two different sets of point cloud computing modules can be deployed in the driving device controller, and the performance differences of point cloud processing algorithms in different point cloud computing modules in the same scenario can be compared by using the autonomous driving companion system (the autonomous driving companion system refers to other driving systems that work in conjunction with the autonomous driving system during the operation of the autonomous driving vehicle), thereby accelerating the iteration speed of the point cloud computing modules.

[0170] Specifically, while, before or after executing the above step S802, a second point cloud computing module based on the driving device controller can be used to perform a second point cloud computing on the original lidar data to obtain second point cloud data.

[0171] The second point cloud computing module can perform at least one of distance calculation, angle calculation, and reflectivity calculation on the original laser radar data based on a second preset algorithm. The second preset algorithm in the second point cloud computing module is different from the first preset algorithm in the first point cloud computing module.

[0172] For example, the first point cloud computing module calculates the distance of the original laser radar data through the time of flight method, calculates the angle of the original laser radar data through the optical phased array angle calculation method, and calculates the reflectivity of the original laser radar data through the photoelectric detector signal intensity analysis method, so as to obtain a laser radar point cloud with spatial coordinate information (i.e., the first point cloud data); the second point cloud computing module calculates the distance of the original laser radar data through the triangulation method, calculates the angle of the original laser radar data through the mechanical scanning angle measurement method, and calculates the reflectivity of the original laser radar data through the multi-wavelength laser radar reflectivity measurement method, so as to obtain a laser radar point cloud with spatial coordinate information (i.e., the second point cloud data).

[0173] Furthermore, since the second preset algorithm is different from the first preset algorithm, the calculated first point cloud data and the second point cloud data may also differ in point distribution, attribute values ​​corresponding to each point (including coordinates, reflectivity, etc.), etc.

[0174] Therefore, the first point cloud data and the second point cloud data may be compared, and the first preset algorithm or the second preset algorithm may be optimized based on the comparison result.

[0175] Specifically, some comparison parameters (such as reflectivity comparison parameters, coordinate comparison parameters, etc.) can be set to compare and analyze the first point cloud data and the second point cloud data to quantify the performance of different algorithms in different aspects.

[0176] Furthermore, the first preset algorithm or the second preset algorithm can be optimized through the comparison results. For example, when the comparison shows that the first point cloud algorithm is highly accurate in calculating the distance of the target object, but is insufficient in the reflectivity calculation, the reflectivity algorithm in the first preset algorithm can be optimized. Alternatively, when the comparison shows that the reflectivity calculation of the second point cloud algorithm is relatively accurate, but the angle calculation of the target in complex scenes is prone to deviation, the angle algorithm in the second preset algorithm can be optimized to improve the processing quality of the lidar point cloud data.

[0177] In some embodiments, abnormality diagnosis can also be performed on the first point cloud data and / or the second point cloud data; and when the first point cloud data or the second point cloud data is abnormal, the original lidar point cloud data corresponding to the first point cloud data or the second point cloud data is obtained; and the cause of the abnormality is obtained based on the original lidar point cloud data.

[0178] Specifically, abnormality diagnosis may be performed on the first point cloud data and / or the second point cloud data through multiple dimensions.

[0179] For example, from the perspective of the spatial distribution characteristics of the data, the point cloud corresponding to the complete surface of an object should be relatively evenly distributed. If the point cloud data is obviously sparse or the density is unreasonable in a certain area, it indicates that the point cloud data may be abnormal. From the perspective of the attribute dimension of the point cloud data, such as the reflectivity value of the point cloud, if there are maximum or minimum values ​​beyond the normal range, or if the reflectivity corresponding to the same object is too different, it may indicate that the point cloud data is abnormal. In addition, from the perspective of the continuity and integrity of the point cloud data, if the outline of an object that should be continuous is broken or missing in the point cloud, it indicates that the point cloud data may be abnormal.

[0180] Since the first point cloud data or the second point cloud data may have lost some original information after computational processing, when the first point cloud data or the second point cloud data is diagnosed to be abnormal, the cause of the abnormality can be obtained based on the corresponding original lidar point cloud data.

[0181] Specifically, the signal strength analysis can be performed on the original LiDAR point cloud data to see if there is a sudden change in the signal strength in a local area. If the signal strength in a certain area is significantly reduced, the abnormal reason may be that the target is blocked; if the signal strength suddenly increases, the abnormal reason may be that a strong reflective object (such as a metal sign, a reflector, etc.) has produced multiple reflections.

[0182] The original lidar point cloud data can also be analyzed for point cloud density. If the point cloud density in a certain area is too low, the abnormality may be that the target object is too far away, beyond the effective detection range of the lidar, or the target object has an extremely low reflectivity to the laser. If the point cloud density is too high, the abnormality may be due to multiple reflections of the laser in a narrow space or scanning of an object with a complex shape.

[0183] In addition, the spatial distribution and geometric shape analysis of the original lidar point cloud data can be performed. If there are multiple consecutive frames of original point cloud data, time series analysis can also be performed to obtain the corresponding abnormal causes, which is not limited here.

[0184] When a traditional LiDAR system encounters anomalies or failures caused by point cloud performance problems, it can only be solved through a large number of tests to reproduce the problem and capture the original LiDAR signal data. Under the new driving system architecture provided by this application, since the driving device controller can directly obtain the original LiDAR data of the LiDAR system, the driving system can immediately obtain the cause of the anomaly through the original LiDAR point cloud data when the anomaly occurs, which speeds up the data processing speed.

[0185] In addition, after the driving device controller obtains the original lidar point cloud data, it can directly perform abnormal diagnosis on the original lidar point cloud data. If it is diagnosed that the original lidar point cloud data has abnormalities (such as missing data, high noise, etc.), the data can be discarded and new original lidar point cloud data can be obtained.

[0186] In some embodiments, the first point cloud data and / or the second point cloud data can also be subjected to noise filtering processing based on the noise filtering module of the driving device controller to reduce or eliminate the noise components existing in the point cloud data, improve the data quality, and facilitate subsequent data processing.

[0187] Furthermore, in some embodiments of step S803, data analysis, coordinate conversion and perception calculation can be performed on the first point cloud data and the second point cloud data based on the data analysis submodule, coordinate conversion submodule and perception calculation submodule in the data processing module to obtain scene information corresponding to the original lidar point cloud data, so that the driving system can control the driving device to perform corresponding operations based on the scene information.

[0188] Based on the data processing method described in steps S801 to S803 above, data processing can be performed through the driving device controller, giving full play to the rapid iteration capability of the driving system data closed loop and software, and improving the data processing performance.

[0189] The above is an explanation of the data processing method provided in this application.

[0190] It should be pointed out that although the various steps in the above embodiments are described in a specific order, those skilled in the art can understand that in order to achieve the effect of the present application, different steps do not have to be executed in such an order. They can be executed simultaneously (in parallel) or in other orders. These adjusted schemes are equivalent to the technical schemes described in this application, and therefore will also fall within the scope of protection of this application.

[0191] It is understood by those skilled in the art that all or part of the processes in the method for implementing the above-mentioned embodiment of the present application can also be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable storage medium may include: any entity or device, medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory, random access memory, electrical carrier signal, telecommunication signal and software distribution medium, etc. that can carry the computer program code.

[0192] Furthermore, the present application also provides a smart device.

[0193] In an embodiment of an intelligent device according to the present application, the intelligent device may include a driving device body and a driving system described in the technical solution of the above-mentioned driving system.

[0194] In some embodiments of the present application, the smart device may further include at least one processor; and a memory communicatively connected to the at least one processor.

[0195] The processor may include hardware, software or a combination of the two. The processor may be a central processing unit, a microprocessor, an image processor, a digital signal processor or any other suitable processor. The processor has data and / or signal processing functions. The processor may be implemented in software, hardware or a combination of the two. Non-temporary computer-readable storage media include any suitable medium that can store program code, such as a disk, a hard disk, an optical disk, a flash memory, a read-only memory, a random access memory, etc.

[0196] A computer program is stored in the memory, and when the computer program is executed by at least one processor, the data processing method described in any of the above embodiments is implemented.

[0197] The smart devices described in this application may include driving devices, smart cars, robots and other devices.

[0198] In some embodiments of the present application, the smart device may further include at least one sensor, which is used to sense information. The sensor is communicatively connected to any type of processor mentioned in the present application. The driving system is used to guide the smart device to drive itself or assist in driving. The processor communicates with the sensor and / or the driving system to complete the method described in any of the above embodiments.

[0199] It should be noted that the relevant user personal information that may be involved in the various embodiments of the present application is strictly in accordance with the requirements of laws and regulations, following the principles of legality, legitimacy and necessity, based on the reasonable purposes of business scenarios, to process the personal information that users actively provide during the use of products / services or generated due to the use of products / services, as well as the personal information obtained with the user's authorization.

[0200] The user personal information processed by this application will vary depending on the specific product / service scenario, and shall be based on the specific scenario in which the user uses the product / service. It may involve the user's account information, sensor information, driving information, vehicle information or other related information. This application will treat the user's personal information and its processing with a high degree of diligence.

[0201] This application attaches great importance to the security of user personal information and has taken reasonable and feasible security protection measures that meet industry standards to protect user information and prevent personal information from being accessed, disclosed, used, modified, damaged or lost without authorization.

[0202] So far, the technical solution of the present application has been described in conjunction with an embodiment shown in the accompanying drawings, but it is easy for those skilled in the art to understand that the protection scope of the present application is obviously not limited to these specific embodiments. Without departing from the principles of the present application, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present application.

Claims

1. A driving device controller, characterized in that: The driving device controller is in communication connection with the laser radar system, and the driving device controller receives the original laser radar point cloud data of the target scene sent by the laser radar system; The driving device controller includes a first point cloud computing module and a data processing module; The first point cloud computing module is configured to perform point cloud computing on the original laser radar point cloud data to obtain first point cloud data; The data processing module is configured to process the first point cloud data to obtain scene information corresponding to the original laser radar point cloud data.

2. The driving device controller according to claim 1, characterized in that: The driving device controller also includes a second point cloud computing module; The second point cloud computing module is configured to perform point cloud computing on the original laser radar point cloud data to obtain second point cloud data; The data processing module is further configured to process the second point cloud data to obtain scene information corresponding to the original laser radar point cloud data; The second point cloud computing module and the first point cloud computing module perform point cloud computing on the original lidar point cloud data based on different algorithms.

3. The driving device controller according to claim 2, characterized in that: The first point cloud computing module is configured to perform at least one of distance calculation, angle calculation, and reflectivity calculation on the original laser radar point cloud data based on a first preset algorithm to obtain the first point cloud data; The second point cloud computing module is configured to perform at least one of the distance calculation, the angle calculation, and the reflectivity calculation on the original lidar point cloud data based on a second preset algorithm to obtain the second point cloud data.

4. The driving device controller according to claim 2, characterized in that: The driving device controller further comprises a noise filtering module, wherein the noise filtering module is configured to perform noise filtering processing on the first point cloud data and / or the second point cloud data; The data processing module includes a data parsing submodule, a coordinate conversion submodule and a perception calculation submodule.

5. A driving system, characterized in that: The driving system comprises a laser radar system and a driving device controller according to any one of claims 1 to 4, wherein the laser radar system is communicatively connected with the driving device controller; the laser radar system comprises a main control module, a transmitting module, a receiving module and a scanning module; Wherein, the main control module includes a communication submodule; The communication submodule is configured to send the original lidar point cloud data of the target scene acquired by the lidar system to the driving device controller.

6. The driving system according to claim 5, characterized in that: The laser radar system also includes a power module; the main control module also includes a system control submodule and a storage submodule; Wherein, the system control submodule is configured to at least control the operation of the transmitting module, the receiving module and the scanning module; The storage submodule is configured to store control parameters of the lidar system and the original lidar point cloud data.

7. The driving system according to claim 6, characterized in that: The main control module also includes a signal processing submodule; The signal processing submodule is configured to filter the raw laser radar point cloud data before sending the raw laser radar point cloud data to the driving device controller.

8. A data processing method, applied to a driving device controller, characterized in that: The method comprises: Obtain the original LiDAR point cloud data of the target scene sent by the LiDAR system; Performing a first point cloud computing on the original laser radar point cloud data to obtain first point cloud data; The first point cloud data is processed to obtain scene information corresponding to the original laser radar point cloud data.

9. The data processing method according to claim 8, characterized in that: The method further comprises: Perform a second point cloud computing on the original laser radar data to obtain second point cloud data; The second point cloud data is processed to obtain scene information corresponding to the original laser radar point cloud data.

10. The data processing method according to claim 9, characterized in that: The performing a first point cloud computing on the original laser radar point cloud data comprises: Perform at least one of distance calculation, angle calculation, and reflectivity calculation on the original laser radar data based on a first preset algorithm; The performing second point cloud computing on the original laser radar point cloud data comprises: Based on a second preset algorithm, at least one of the distance calculation, the angle calculation, and the reflectivity calculation is performed on the original laser radar data.

Images