Intelligent driving control method and related apparatus

Abstract

An intelligent driving control method and a related apparatus. The method comprises: checking the deviation of the pitch angle of a laser radar in a vehicle; and, if the occurrence of a deviation of the pitch angle of the laser radar is detected, instructing the vehicle to exit an intelligent driving function. Using the present solution can improve intelligent driving safety protection and reduce safety risks.

Description

Intelligent driving control method and related device

[0001] The present application claims priority to the Chinese patent application No. 202411383144.X, filed on September 27, 2024, with the State Intellectual Property Office of China, and entitled "Intelligent driving control method and related device", the entire content of which is incorporated herein by reference. TECHNICAL FIELD

[0002] The present application relates to the technical field of vehicles, in particular to an intelligent driving control method and related device. BACKGROUND

[0003] Laser radar is widely used. For example, it can be applied to the fields of intelligent driving, robots, security monitoring, unmanned aerial vehicles, map surveying and mapping, mining, forestry, archaeology, geology, seismology, topographic surveying, forestry surveying, or disaster warning, etc. In particular, the laser radar applied in intelligent driving, the normal working of the laser radar is the key to guarantee safe driving in the process of intelligent driving. Therefore, how to improve the safety protection of intelligent driving and reduce the safety risk is a problem to be solved urgently. SUMMARY

[0004] The present application provides an intelligent driving control method and related device, which can improve the safety protection of intelligent driving and reduce the safety risk.

[0005] In a first aspect, the present application provides an intelligent driving control method, which comprises: detecting the deviation of the pitch angle of a laser radar in a vehicle; if it is detected that the pitch angle of the laser radar deviates, instructing the vehicle to exit the intelligent driving function.

[0006] In the above scheme, for the intelligent driving vehicle applying the laser radar, the normal working of the laser radar is the key to guarantee safe driving in the process of intelligent driving. The prerequisite for the normal working of the laser radar is to keep it within a reasonable extrinsic parameter pose range. The deviation of the pitch angle of the laser radar will affect the extrinsic parameter, and then affect the detection result of the laser radar, bring large error or even wrong driving control to the vehicle, and thus cause safety hazards. In the present scheme, the deviation of the pitch angle of the laser radar can be detected. If it is detected that the pitch angle deviates, the intelligent driving function is instructed to exit. The safety protection of intelligent driving is improved, and the safety risk is reduced.

[0007] In a possible implementation manner, the detection of the deviation of the pitch angle of the laser radar in the vehicle comprises:

[0008] acquiring target point cloud data; the target point cloud data comprises point cloud data collected by one or more laser line beams shooting to the ground in one frame or N frames of point cloud of the laser radar; N is an integer greater than 1;

[0009] determine a first angle based on the target point cloud data and the installation height of the laser radar;

[0010] determine a pitch angle deviation of the laser radar based on the first angle and a second angle; the second angle is determined based on preset pitch angles of the one or more laser beam bundles directed to the ground.

[0011] Optionally, the second angle is a mean value of the pitch angles of the one or more laser beam bundles directed to the ground; and the determination of the pitch angle deviation of the laser radar based on the first angle and the second angle comprises: if a difference between the first angle and the second angle is greater than or equal to a threshold value, determining that the pitch angle of the laser radar deviates.

[0012] In the above scheme, the point cloud data collected by the beam bundles directed to the ground and the installation height of the laser radar are selected to calculate the actual pitch angles of the beam bundles (represented by the first angle), and then the first angle and the preset pitch angles of the beam bundles (represented by the second angle) are used to determine whether the pitch angle of the laser radar deviates. In this way, after determining that the pitch angle of the laser radar deviates, a response can be quickly made to reduce the adverse effects caused by the pitch angle deviation. For example, the vehicle can be instructed to exit the intelligent driving function to reduce the safety risk.

[0013] In a possible implementation, the target point cloud data comprises point cloud data collected within a preset horizontal detection angle range.

[0014] In the above scheme, the data collected within the preset horizontal angle range is further limited, so that the influence of abnormal point cloud on the subsequent calculation result can be eliminated, and the accuracy of the pitch angle deviation determination of the laser radar is improved.

[0015] In a possible implementation, when the target point cloud data comprises point cloud data collected by the one or more laser beam bundles directed to the ground in the N frames of point cloud data, the determination of the first angle based on the target point cloud data and the installation height of the laser radar comprises:

[0016] determine a first distance set for each frame of data in the target point cloud data to obtain N first distance sets; an i-th first distance set in the N first distance sets is determined based on i-th frame of data in the target point cloud data, the i-th first distance set comprises distances of detection points corresponding to one or more detection data in the i-th frame of data, i is an integer from 1 to N; and the first angle is determined based on the N first distance sets and the installation height of the laser radar.

[0017] Optionally, the determining the first angle based on the N first distance sets and the installation height of the laser radar comprises: determining N third angles based on the installation height and the N first distance sets, wherein an i-th third angle is determined based on the installation height and an i-th first distance set; and determining the first angle based on the N first distance sets and the N third angles.

[0018] Optionally, the determining the N third angles based on the installation height and the N first distance sets comprises: calculating a ratio of the installation height and each distance in each first distance set to obtain N ratio sets, wherein an i-th ratio set is determined based on the installation height and the i-th first distance set; and performing inverse sine function solving on each ratio in the i-th ratio set to obtain an i-th pitch angle set, and determining the i-th third angle by solving a mean value of the i-th pitch angle set, to obtain the N third angles.

[0019] In the above scheme, the point cloud data collected by the line bundle directed to the ground and the installation height of the laser radar can be used to calculate the actual pitch angle of the line bundle (i.e., the first angle), and then the deviation of the pitch angle of the laser radar can be determined based on the first angle and the preset pitch angle of the line bundle. This method does not rely on any external devices, laser point cloud semantic segmentation, and feature extraction. Based on the laser raw point cloud, the pitch angle deviation in the vertical direction can be detected based on the existing laser radar field angle and the laser installation height information, without any initial value requirement. Thus, the calculation resources are greatly saved, the calculation speed is fast, and the timeliness is strong.

[0020] In a possible implementation manner, the determining N first distance sets based on the data of each frame in the target point cloud data respectively comprises:

[0021] The determining N second distance sets based on the data of each frame respectively comprises: calculating a distance of a detection point corresponding to each detection data included in the data of each frame to obtain N second distance sets, wherein an i-th second distance set is determined based on the i-th frame of data.

[0022] In a case where a discrete degree of the distances in the i-th second distance set meets a first preset condition, performing first screening processing on the i-th second distance set to determine the i-th first distance set, to obtain the N first distance sets; and the first screening processing comprises discarding outliers and / or selecting distances meeting a second preset condition in value size.

[0023] In the foregoing scheme, by screening the distances in the N second distance sets, some abnormal data can be excluded, and the influence on the subsequent calculation results is reduced. The accuracy of the laser radar pitch angle deviation judgment is improved.

[0024] In a possible implementation, the foregoing determining the first angle based on the N first distance sets and the N third angles includes: calculating a mean value of each of the N first distance sets to obtain N distance mean values; performing second screening processing on the N distance mean values and the N third angles respectively to obtain M1 distance mean values and M2 third angles; M1 and M2 are integers greater than 0 and less than N; the second screening processing includes discarding outliers; in a case where a dispersion degree of the M1 distance mean values meets a second preset condition and a dispersion degree of the M2 third angles meets a third preset condition, determining a mean value of the M2 third angles as the first angle.

[0025] In the foregoing scheme, the obtained third angles are screened based on the N first distance sets, and finally the first angle is obtained based on one or more third angles screened. This screening and processing can exclude some abnormal data, reduce the influence on the subsequent calculation results, and improve the accuracy of the laser radar pitch angle deviation judgment.

[0026] In a possible implementation, in a case where the target point cloud data includes point cloud data collected by one or more laser line beams directed to the ground in a frame of point cloud data, the foregoing determining the first angle based on the target point cloud data and the installation height of the laser radar includes: determining a first distance set based on the target point cloud data; the first distance set includes distances of detection points corresponding to one or more detection data included in the target point cloud data; and determining the first angle based on the first distance set and the installation height of the laser radar.

[0027] For example, the foregoing determining the first angle based on the first distance set and the installation height of the laser radar includes: calculating a ratio of the installation height and each distance in the first distance set; performing inverse sine function solving based on each calculated ratio to obtain a pitch angle set; and calculating a mean value of the pitch angle set to obtain the first angle.

[0028] In the foregoing scheme, the actual pitch angle (i.e., the first angle) of the line bundle collected by the point cloud data directed to the ground and the installation height of the laser radar can be calculated, and then whether the pitch angle of the laser radar deviates can be determined based on the first angle and the preset pitch angle of the line bundle. It can not depend on any external device, laser point cloud semantic segmentation and feature extraction, etc. Based on the laser original point cloud, the pitch angle deviation in the vertical direction can be detected based on the existing laser radar field angle and the laser installation height information, without any initial value requirement. Thus, the calculation resource is greatly saved, the calculation speed is fast, and the timeliness is strong.

[0029] In a possible implementation, the foregoing determining the first distance set based on the target point cloud data comprises: calculating distances of detection points corresponding to each detection data included in the target point cloud data based on the target point cloud data to obtain a second distance set; and performing screening processing on the second distance set to obtain the first distance set in a case where a discrete degree of distances in the second distance set meets a first preset condition; and the screening processing comprises discarding outliers and / or selecting distances whose numerical values meet a second preset condition.

[0030] In the foregoing scheme, by screening the distances in the second distance set, some abnormal data can be excluded, and the influence on the subsequent calculation result is reduced. The accuracy of the pitch angle deviation judgment of the laser radar is improved.

[0031] In a possible implementation, before the target point cloud data is acquired, the foregoing further comprises: determining that the vehicle travels on a road with an unobstructed field of view based on position information and speed information of the vehicle or based on an image captured by a camera of the vehicle.

[0032] In the foregoing scheme, detecting the pitch angle deviation of the laser radar on the road with the unobstructed field of view can improve the final detection result.

[0033] In a possible implementation, before the target point cloud data is acquired, the foregoing further comprises: determining that the vehicle travels on a flat road based on a body pitch angle of the vehicle.

[0034] In the foregoing scheme, detecting the pitch angle deviation of the laser radar on the flat road can improve the final detection result.

[0035] In a second aspect, the present application provides an intelligent driving control device, which comprises:

[0036] a detection unit configured to detect a deviation of a pitch angle of a laser radar in a vehicle;

[0037] An indicating unit is configured to instruct the vehicle to exit the intelligent driving function if a deviation in the pitch angle of the laser radar is detected.

[0038] In one possible implementation, the detection unit is specifically configured to:

[0039] obtain target point cloud data, the target point cloud data including point cloud data collected by one or more laser line beams directed to the ground in one frame or N frames of point cloud data of the laser radar, N being an integer greater than 1;

[0040] determine a first angle based on the target point cloud data and the installation height of the laser radar;

[0041] determine a deviation in the pitch angle of the laser radar based on the first angle and a second angle, the second angle being determined based on a preset pitch angle of the one or more laser line beams directed to the ground.

[0042] In one possible implementation, the target point cloud data includes point cloud data collected within a preset horizontal detection angle range.

[0043] In one possible implementation, when the target point cloud data includes point cloud data collected by one or more laser line beams directed to the ground in the N frames of point cloud data, the detection unit is specifically configured to:

[0044] determine N first distance sets respectively based on data of each frame in the target point cloud data, the i-th first distance set in the N first distance sets being determined based on data of the i-th frame in the target point cloud data, the i-th first distance set including distances of detection points corresponding to one or more detection data included in the data of the i-th frame, i being an integer from 1 to N;

[0045] determine the first angle based on the N first distance sets and the installation height of the laser radar.

[0046] In one possible implementation, the detection unit is specifically configured to:

[0047] calculate distances of detection points corresponding to each detection data included in the data of each frame respectively based on the data of each frame, and obtain N second distance sets, the i-th second distance set in the N second distance sets being determined based on the data of the i-th frame.

[0048] In a case where a dispersion degree of distances of the i-th second distance set meets a first preset condition, performing a first screening process on the i-th second distance set to determine the i-th first distance set, so as to obtain the N first distance sets; the first screening process comprises discarding outliers and / or selecting distances with numerical values meeting a second preset condition.

[0049] In a possible implementation, the detection unit is specifically configured to:

[0050] determining N third angles based on the installation height and the N first distance sets, wherein an i-th third angle in the N third angles is determined based on the installation height and the i-th first distance set;

[0051] determining the first angle based on the N first distance sets and the N third angles.

[0052] In a possible implementation, the detection unit is specifically configured to:

[0053] calculating ratios of the installation height and each distance in each first distance set to obtain N ratio sets, wherein an i-th ratio set in the N ratio sets is determined based on the installation height and the i-th first distance set;

[0054] performing inverse sine function solving based on each ratio in the i-th ratio set to obtain an i-th pitch angle set, and determining the i-th third angle by solving a mean value of the i-th pitch angle set, so as to obtain the N third angles.

[0055] In a possible implementation, the detection unit is specifically configured to:

[0056] calculating a mean value of each first distance set in the N first distance sets to obtain N distance mean values;

[0057] performing a second screening process on the N distance mean values and the N third angles respectively to obtain M1 distance mean values and M2 third angles; M1 and M2 are integers greater than 0 and less than N; the second screening process comprises discarding outliers;

[0058] in a case where a dispersion degree of the M1 distance mean values meets a second preset condition and a dispersion degree of the M2 third angles meets a third preset condition, determining a mean value of the M2 third angles as the first angle.

[0059] In a possible implementation, in a case where the target point cloud data comprises point cloud data collected by one or more laser line beams directed to the ground in a frame of point cloud data, the detection unit is specifically configured to:

[0060] determine a first distance set based on the target point cloud data; the first distance set comprises distances of the detection points corresponding to one or more detection data included in the target point cloud data;

[0061] determine the first angle based on the first distance set and the installation height of the lidar.

[0062] In a possible implementation, the detection unit is specifically configured to:

[0063] calculate distances of the detection points corresponding to each detection data included in the target point cloud data based on the target point cloud data, to obtain a second distance set;

[0064] perform screening processing on the second distance set to obtain the first distance set in a case where a dispersion degree of the distances in the second distance set meets a first preset condition; the screening processing comprises discarding outliers and / or selecting distances with numerical values meeting a second preset condition.

[0065] In a possible implementation, the detection unit is specifically configured to:

[0066] calculate a ratio of the installation height and each distance in the first distance set;

[0067] perform inverse sine function solving based on each calculated ratio, to obtain an elevation angle set;

[0068] obtain the first angle by solving a mean value of the elevation angle set.

[0069] In a possible implementation, the second angle is a mean value of the elevation angles of the one or more laser beams directed to the ground; and the detection unit is specifically configured to: determine that the elevation angle of the lidar deviates in a case where a difference between the first angle and the second angle is greater than or equal to a threshold value.

[0070] In a possible implementation, the detection unit is specifically configured to: before the target point cloud data is obtained, determine that the vehicle travels on a road with an unobstructed field of view based on position information and speed information of the vehicle, or based on an image captured by a camera of the vehicle.

[0071] In a possible implementation, the detection unit is specifically configured to: before the target point cloud data is obtained, determine that the vehicle travels on a flat road based on a body elevation angle of the vehicle.

[0072] In a third aspect, the present application provides a processing device in a vehicle, comprising a processor and a memory, wherein the memory is configured to store computer programs or computer instructions, and the processor is configured to execute the computer programs or computer instructions stored in the memory, so that the processing device executes the method of any one of the first aspect.

[0073] In a fourth aspect, the present application provides a vehicle, comprising a lidar and the processing device of the third aspect.

[0074] In a fifth aspect, the present application provides a computer readable storage medium, which stores computer programs or computer instructions, and the computer programs or computer instructions are executed by a processor to implement the method of any one of the first aspect.

[0075] In a sixth aspect, the present application provides a computer program product, when executed by a processor, the method of any one of the first aspect will be implemented.

[0076] The advantages of the second aspect to the third aspect described above can be referred to the advantages described above with respect to the first aspect, which will not be repeated here. BRIEF DESCRIPTION OF DRAWINGS

[0077] FIG. 1 shows a coordinate system of a lidar;

[0078] FIG. 2 shows a multi-line beam emitted by a lidar;

[0079] FIG. 3 shows a system architecture diagram;

[0080] FIGS. 3A and 4 show method flow diagrams;

[0081] FIGS. 5 and 5A show a multi-line beam emitted by a lidar;

[0082] FIG. 6 shows a relative position between a lidar and a ground;

[0083] FIG. 7 shows a possible algorithm framework diagram;

[0084] FIGS. 8 and 9 show structural diagrams. DETAILED DESCRIPTION

[0085] In the embodiments of the present application, "multiple" refers to two or more than two. In the embodiments of the present application, "and / or" is used to describe the association relationship of the associated objects, which represents three independent existing relationships, for example, A and / or B, which can represent: A exists alone, B exists alone, or A and B exist together. The description such as "at least one of a1, a2, … and an (or at least one)" adopted in the embodiments of the present application includes any one of a1, a2, … and an existing alone, and also includes any combination of a1, a2, … and an, and each case can exist independently; for example, the description of "at least one of a, b and c" includes the cases of a alone, b alone, c alone, a and b combination, a and c combination, b and c combination, or abc three combination.

[0086] In various embodiments of the present application, the terms and / or descriptions between various embodiments are consistent and can be mutually referred to if there is no special description and logical conflict, and the technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationship.

[0087] First, introduce the technical terms involved in the embodiments of the present application.

[0088] Point cloud: the point cloud refers to a set of points formed by obtaining the spatial coordinates of each sampling point on the surface of an object. The point cloud used for three-dimensional target detection is usually obtained by laser radar scanning. Illustratively, a laser beam emitted by a laser radar irradiates the surface of an object, and the laser reflected by the surface of the object is received by the receiver of the laser radar to form a detection point of detection data. The detection data of a detection point can form a laser point. If the laser beam scans according to a certain trajectory, the information of the laser point can be recorded while scanning, so as to obtain a large number of laser points. These laser points can form a point cloud. Illustratively, the detection data of each laser point, i.e. each detection point, can include three-dimensional coordinates of the detection point in the laser radar coordinate system. Optionally, it can also include laser reflection intensity and color information, etc.

[0089] Illustratively, in another implementation, the coordinates of the laser points in the point cloud can be polar coordinates in a polar coordinate system, or can be Cartesian coordinates in a Cartesian coordinate system. The polar coordinates, Cartesian coordinates and three-dimensional coordinates in the laser radar coordinate system can be converted into each other. In order to facilitate the description, the embodiments of the present application mainly take the coordinates of the laser points in the point cloud as the three-dimensional coordinates in the laser radar coordinate system as an example for introduction.

[0090] Single-frame point cloud: Generally, the point cloud collected by the laser radar within a preset time length can be regarded as a frame point cloud (i.e., a single-frame point cloud). Alternatively, the point cloud within a preset range including a certain point in the point cloud map can be regarded as a single-frame point cloud, for example, the point cloud within a preset range centered on the certain point can be regarded as a single-frame point cloud. In a specific implementation, the value of the preset time length or the preset range can be set according to actual needs, and the embodiments of the present application do not make any limitation. The single-frame point cloud can also be referred to as a point cloud frame.

[0091] Field of view angle of laser radar: The size of the field of view angle (FOV) determines the range that can be detected by the laser radar. The field of view angle of the laser radar includes a horizontal field of view angle and a vertical field of view angle. The size of the horizontal field of view angle determines the angle range that can be detected by the laser radar in the horizontal direction. For example, if the horizontal field of view angle is 360°, the laser radar can rotate a full circle (360°) in the horizontal direction for detection. The vertical field of view angle, which can also be referred to as the pitch field of view angle, determines the angle range that can be detected by the laser radar in the vertical direction. For example, if the vertical field of view angle is 40°, the laser radar can detect within a 40° range in the vertical direction. It can be understood that the values of the horizontal field of view angle and the vertical field of view angle of the laser radar can be determined according to actual conditions, and the embodiments of the present application do not make any limitation.

[0092] Channel of laser radar: The channel refers to a channel in the horizontal direction. Each channel covers a certain horizontal detection angle range. If each channel covers the same horizontal detection angle range, the number of channels is equal to the horizontal field of view angle of the laser radar within the horizontal detection angle range covered by each channel. For example, assuming that the horizontal field of view angle of the laser radar is 120° and each channel covers a horizontal detection angle range of 24°, the number of channels is 120 / 24=5.

[0093] Direction angle of laser beam: The direction angle of the laser beam includes a horizontal direction angle and a vertical direction angle. The horizontal direction angle refers to the emission angle of the laser beam in the horizontal direction. The vertical direction angle refers to the emission angle of the laser beam in the vertical direction. The vertical direction angle is also referred to as the pitch angle. The angle sizes of the horizontal direction angle and the vertical direction angle can be determined in combination with the defined laser radar coordinate system, and the angle sizes of the horizontal direction angle and the vertical direction angle determined by different definitions of the laser radar coordinate system can be different. For ease of understanding, as an example, reference can be made to FIG. 1.

[0094] In FIG. 1, a coordinate system of a laser radar is exemplarily shown. The coordinate system is a three-dimensional coordinate system, and the coordinate origin is point O, which includes an x-axis, a y-axis, and a z-axis. The coordinate origin may, for example, be a center point of the laser radar or a laser emission center point, etc. Exemplarily, in one possible implementation, the laser emission center point is the center point of the laser radar. Point P in FIG. 1 is a detection point, and a laser beam emitted by the laser radar can hit the point P. The projection of the point P in the xy plane of the coordinate system is represented as point P'. The line connecting the point P' and the point O (which can be represented as line OP') is located in the xy plane. Then, the included angle φ between the line OP' and the x-axis is the horizontal direction angle of the laser beam. The included angle θ between the laser beam and the line OP' is the vertical direction angle of the laser beam, which is also the pitch angle. The pitch angle of the laser beam is actually the included angle between the laser beam and the xy plane.

[0095] It can be understood that FIG. 1 is only an example, and in specific implementations, the angle of the horizontal direction angle and the vertical direction angle of the laser beam can also be determined according to different definitions, which are not limited in the embodiments of the present application.

[0096] The pitch angle of the laser beam described above is also the pitch angle of the laser radar. That is, the pitch angle of the laser radar refers to the included angle between the laser beam and the horizontal plane (for example, the xy plane) when the laser radar emits the laser beam.

[0097] Line number of laser radar: The laser radar includes a single-line laser radar and a multi-line laser radar. The single-line laser radar refers to a radar in which the line beam emitted by the laser source is a single line. At present, it is mainly applied in the field of robots, and service robots are the most common, which can help the robots to avoid obstacles, and has fast scanning speed, high resolution, and high reliability. The single-line laser radar can rotate in the horizontal direction and the vertical direction to realize scanning of the surrounding environment.

[0098] The multi-line laser radar refers to a radar that can emit multiple laser beams at the same time. At present, there are 4-line, 8-line, 16-line, 32-line, 64-line, 96-line, or 128-line multi-line laser radars on the market. The multi-line laser radar has multiple laser emitters and receivers in the vertical direction, and can generate multiple laser beams during each scan. The more the line number is, the more perfect the obtained object surface profile is. Exemplarily, the pitch angle of each laser beam emitted by the multi-line laser radar is fixed. For example, see FIG. 2. FIG. 2 exemplarily shows a multi-beam diagram of a laser radar. As can be seen in FIG. 2, the angle of each beam emitted by the laser radar in the vertical direction is fixed. Then, under the driving of the mechanical structure, the laser radar can rotate in the horizontal direction to realize scanning of the surrounding environment.

[0099] The embodiments of the present application will be exemplarily introduced below with reference to the accompanying drawings.

[0100] In practical applications, the normal operation of the laser radar is the key to guarantee safe driving in the process of intelligent driving in a vehicle that applies laser radar to assist in realizing intelligent driving functions. Therefore, how to improve intelligent driving safety protection and reduce safety risks is a problem to be solved. Exemplarily, the premise of normal operation of the laser radar is to keep within a reasonable range of external parameter poses. Deviation of the pitch angle of the laser radar will affect the external parameter, and further affect the detection result of the laser radar. For example, if there is a large deviation in the pitch angle of the laser radar, it will cause a large error in the estimated distance when detecting objects, thereby causing adverse effects. Taking the field of intelligent driving as an example, the laser radar, as an important sensor of the intelligent driving system, plays an important role in the perception and positioning of the surrounding environment of the vehicle. If there is a large deviation in the pitch angle of the laser radar in the process of intelligent driving, it will cause a large error in the estimated distance when detecting objects, which will affect the decision of intelligent driving, or cause a collision, or trigger the automatic emergency brake system, etc., which will have a huge impact on the safety of intelligent driving. Therefore, in order to reduce the impact of the pitch angle deviation of the laser radar on the safety of intelligent driving, the embodiments of the present application provide an intelligent driving control method and related device, which can improve intelligent driving safety protection and reduce safety risks.

[0101] In a possible implementation manner, in order to implement the intelligent driving control method provided by the embodiments of the present application, the embodiments of the present application provide a system architecture as shown in FIG. 3. The system architecture includes a laser radar 310 and a processing device 320. Exemplarily, the laser radar 310 can be a single-line laser radar or a multi-line laser radar. The point cloud data collected by the laser radar 310 can be sent to the processing device 320, and the processing device 320 analyzes and judges whether the laser radar 310 has a pitch angle deviation based on the point cloud data.

[0102] Exemplarily, the laser radar 310 can be installed on a terminal device, and the processing apparatus 320 can be a controller, a processor or a chip system or the like processing function apparatus or module in the terminal device. For example, taking a vehicle as the terminal device. The laser radar 310 can be installed on the vehicle to assist the vehicle to realize the intelligent driving function. The processing apparatus 320 can be a controller, a processor or a chip system or the like processing function apparatus or module in the vehicle. For example, it can be a vehicle domain controller (VDC), a cockpit domain controller (CDC), a vehicle integrated / integration unit (VIU), a mobile data center (MDC) or a central computing architecture (CCA) and the like in the vehicle. The MDC belongs to the intelligent driving computing power module of the vehicle, and can also be called a motion domain control (MDC). It can be understood that the introduction of the controller in the vehicle here is only an example, and in another implementation, these controllers can be called by other names, and the embodiments of the present application do not limit this.

[0103] Exemplarily, the terminal device is not limited to be a vehicle, but can also be a drone, a robot, a building surveying device or an exploration device and the like. Or it can be a ship or an airplane or other traffic tools. The embodiments of the present application do not repeat them one by one.

[0104] The following exemplarily introduces an intelligent driving control method provided by the embodiments of the present application. The laser radar can be the laser radar 310 shown in FIG. 3. The method can be executed by the processing apparatus 320 shown in FIG. 3. Exemplarily, referring to FIG. 3A, the method includes but is not limited to the following steps S301-S302.

[0105] S301, detecting the deviation of the pitch angle of the laser radar in the vehicle.

[0106] The specific implementation of this step can refer to the exemplarily introduced in the following FIG. 4 and its possible implementation, which is not described in detail here.

[0107] S302, if the deviation of the pitch angle of the laser radar is detected, instructing the vehicle to exit the intelligent driving function.

[0108] Exemplarily, in a specific implementation, if the vehicle is using the intelligent driving function, after determining that the pitch angle of the lidar deviates, the processing device can instruct the vehicle to exit the intelligent driving function, so as to reduce the risk of traffic accidents and ensure the driving safety of the vehicle.

[0109] In a possible implementation, the detection of the deviation of the pitch angle of the lidar in the vehicle can include, but is not limited to, steps S401 to S403, which can be exemplarily seen from FIG. 4.

[0110] S401, obtaining target point cloud data; the target point cloud data includes point cloud data collected by one or more laser beams shooting to the ground in one frame or N frames of point cloud of the lidar; N is an integer greater than 1.

[0111] Exemplarily, in a specific implementation, the lidar can perform laser scanning on the surrounding environment in real time to collect point cloud data. The quantity unit of the point cloud data can be a frame. The description of one frame of point cloud data can refer to the description of the single frame of point cloud in the foregoing description, and details are not repeated here. One frame of point cloud data can also be referred to as a point cloud frame. The one frame or N frames of point cloud of the lidar are point cloud frames obtained by the lidar scanning the surrounding environment.

[0112] Then, the lidar can send the collected point cloud frame to the processing device (for example, the processing device 320 shown in FIG. 3). After receiving the point cloud data, the processing device can process the received point cloud data to obtain the target point cloud. The processing process is exemplarily introduced below.

[0113] In a possible implementation, if the lidar is a multi-line lidar, based on the foregoing description of the multi-line lidar, the pitch angle of each laser beam emitted by the multi-line lidar is fixed. Then, the processing device can select the point cloud data collected by the preset beam from the received one frame or N frames of point cloud to obtain the target point cloud data. The preset beam is, for example, one or more laser beams shooting to the ground. For ease of understanding, FIG. 5 and FIG. 5A are exemplarily shown.

[0114] As can be seen in FIG. 5, the multi-line laser radar can emit a plurality of laser beams with fixed elevation angles in the vertical direction. Some of the laser beams are directed to the ground, and some are directed to other directions. In FIG. 5, the laser beams directed to the ground are exemplarily shown. Then, one or more beams are selected from the laser beams directed to the ground to obtain a preset beam. Exemplarily, the preset beam can be one or more beams in the middle part of the laser beams directed to the ground in the vertical direction, for example, as shown in FIG. 5. For example, in the laser beams directed to the ground, one or more beams close to the ground in the vertical direction and one or more beams far away from the ground in the vertical direction are excluded to obtain the preset beam. Alternatively, for example, one or more beams can be randomly selected from the middle part of the laser beams directed to the ground as the preset beam.

[0115] Alternatively, exemplarily, the preset beam described above can be one or more beams close to the ground in the vertical direction of the laser beams directed to the ground, for example, as shown in FIG. 5A. For example, in the laser beams directed to the ground, one or more beams close to the ground are selected as the preset beam.

[0116] It can be understood that the above-described FIG. 5 and FIG. 5A are only an example and do not constitute a limitation on the embodiments of the present application.

[0117] Exemplarily, in a possible implementation, the detection data including a plurality of detection points in a point cloud frame, that is, including a plurality of laser points. The laser points collected by different laser beams can be distinguished by the coordinates of the laser points. For example, the coordinates of the laser points are the three-dimensional coordinates of the detection points corresponding to the laser points in the laser radar coordinate system. Based on this, taking a frame of point cloud as an example, after the processing device receives the point cloud frame sent by the laser radar, based on the preset beam described above, in combination with the coordinates of each laser point in the point cloud frame, the point cloud data collected by the preset beam in the point cloud frame can be obtained.

[0118] In a possible implementation, the laser radar is a single-line laser radar. As introduced above, the single-line laser radar can rotate and scan in the vertical direction. That is, the laser beam of the single-line laser radar can change in the pitch angle. Then, the processing apparatus can select the point cloud data collected within a preset pitch angle range from a received frame or N frames of point cloud, to obtain the target point cloud data. The laser beam emitted within the preset pitch angle range is directed to the ground. The laser beam emitted within the preset pitch angle range can be exemplarily seen from the preset beam shown in FIG. 5. Different from the multi-line laser radar, the preset beam shown in FIG. 5 is emitted by the single-line laser radar in the vertical direction. Similarly, the laser points collected by laser beams with different pitch angles can be distinguished by the coordinates of the laser points. Based on this, taking one frame of point cloud as an example, after receiving the point cloud frame sent by the laser radar, the processing apparatus can obtain the point cloud data collected within the preset pitch angle range in the point cloud frame, based on the preset pitch angle range and the coordinates of the laser points in the point cloud frame.

[0119] In another possible implementation, the target point cloud data can also be point cloud data collected within a preset horizontal detection angle range. That is, the target point cloud data is point cloud data collected by one or more laser beams directed to the ground in one frame or N frames of point cloud of the laser radar, and is point cloud data collected within the preset horizontal detection angle range.

[0120] Exemplarily, in a possible implementation, the point cloud data collected within the preset horizontal detection angle range can be selected by a preset channel and / or a preset horizontal detection angle range. As introduced above, each channel of the laser radar covers a certain horizontal detection angle range. After the channel is determined, the corresponding horizontal detection angle range can be determined. In a possible implementation, the laser points collected at different horizontal detection angles can be distinguished by the coordinates of the laser points. Based on this, taking one frame of point cloud as an example, the processing apparatus can obtain the point cloud data collected within the preset horizontal detection angle range, based on the preset channel and / or the preset horizontal detection angle range and the coordinates of the laser points in the point cloud frame.

[0121] Exemplarily, the preset beam and / or the preset horizontal detection angle range can be used to select point cloud data with a smaller abnormal probability, so as to eliminate some abnormal point cloud data, thereby reducing the influence of the abnormal point cloud on the subsequent calculation result and improving the accuracy of the result. In addition, for laser radars with different field angles of view, the beam selection and / or the horizontal detection angle selection can be used to obtain suitable point cloud data for subsequent pitch angle offset judgment. This solution can be applied to any field angle of view or any type of laser radar, and has a wide range of applications. Moreover, the solution does not depend on complex calculations such as laser point cloud semantic segmentation and feature extraction. That is, the solution has low computing resource consumption and fast calculation speed.

[0122] S402, determine a first angle based on the target point cloud data and the installation height of the lidar.

[0123] Exemplarily, the first angle can be regarded as an actual determined pitch angle of the lidar.

[0124] Exemplarily, the installation height of the lidar can be, for example, a height between a mounting position of the lidar and the ground. For example, it can be a height between a center point of the fixedly mounted lidar and the ground.

[0125] Exemplarily, based on the foregoing description, the target point cloud data is selected based on one frame or N frames of point cloud of the lidar. The following exemplarily introduces two cases that the target point cloud data is selected based on N frames of point cloud of the lidar and the target point cloud data is selected based on one frame of point cloud of the lidar.

[0126] The following exemplarily introduces the implementation process of determining the first angle based on the target point cloud data and the installation height of the lidar in the case that the target point cloud data is selected based on N frames of point cloud of the lidar.

[0127] Exemplarily, in a specific implementation, the point cloud data collected by the preset beam is selected from each frame of point cloud of the N frames of point cloud in the manner of the above step S401. Or, the point cloud data collected by the preset beam within the preset horizontal detection angle range is selected from each frame of point cloud of the N frames of point cloud. For the convenience of subsequent description, the point cloud data selected from each frame is simply referred to as a target point cloud frame. Then, N target point cloud frames can be obtained based on the N frames of point cloud. The N target point cloud frames constitute the target point cloud data.

[0128] Exemplarily, one first distance set can be determined based on the data of each frame of the N target point cloud frames respectively, and N first distance sets are obtained. The i-th first distance set in the N first distance sets can be determined based on the data of the i-th frame in the N target point cloud frames. The i-th first distance set includes distances of detection points corresponding to one or more detection data included in the data of the i-th frame, and i is an integer from 1 to N. For the convenience of understanding, the following takes the i-th first distance set determined based on the i-th frame in the N target point cloud frames as an example for introduction.

[0129] Exemplarily, the distance of each probe point corresponding to the probe data included in the i-th frame can be calculated first to obtain a i-th second distance set. Exemplarily, the distance of the probe point is the distance between the probe point and the origin of the laser radar coordinate system. As known from the foregoing description, the probe data of the probe point includes the three-dimensional coordinates of the probe point in the laser radar coordinate system. Based on the three-dimensional coordinates, the distance between the probe point and the origin of the laser radar coordinate system can be calculated, and details are not described herein.

[0130] In a possible embodiment, the dispersion degree of the i-th second distance set can be calculated. For example, the standard deviation, average difference or range of the i-th second distance set can be calculated. If the dispersion degree of the i-th second distance set meets a first preset condition, the i-th second distance set is further subjected to a first screening process to obtain the i-th first distance set. If the dispersion degree of the i-th second distance set does not meet the first preset condition, the i-th frame data is discarded. Exemplarily, the first preset condition is that the dispersion degree is less than or equal to a preset threshold. Taking the standard deviation as an example of the dispersion degree, the first preset condition is that the standard deviation is less than or equal to a first standard deviation threshold. If the standard deviation of the i-th second distance set is less than or equal to the first standard deviation threshold, the i-th second distance set is further subjected to the first screening process to obtain the i-th first distance set. Otherwise, if the standard deviation of the i-th second distance set is greater than the first standard deviation threshold, the i-th frame data is discarded. Exemplarily, the value range of the first standard deviation threshold can be any value between 0.5 meters and 2 meters, for example. It can be understood that the value range shown here is only an example and does not limit the embodiments of the present application. If the dispersion degree is represented by the average difference or the range, the same applies to the standard deviation, and details are not described herein. This implementation can discard unsuitable data, reduce the use of abnormal data for subsequent calculation, improve the accuracy of the calculation result, and improve the stability and reliability of the overall scheme.

[0131] Exemplarily, the first screening processing can include discarding outliers and / or selecting distances with numerical values satisfying a second preset condition. For example, in one possible implementation, in a case where the dispersion degree of the i-th second distance set satisfies a first preset condition, the distance values in the i-th second distance set can be sorted in ascending order, and then one or more distance values with the smallest distances and one or more distance values with the largest distances can be discarded. Alternatively, outliers in the i-th second distance set can be discarded by a clustering method, and the present embodiment does not limit the specific implementation means of discarding outliers. Then, after discarding outliers, a preset proportion of distance values are selected from the remaining distance values. The selected distance values with the preset proportion are the i-th first distance set. Exemplarily, the selected distance values with the preset proportion can be, for example, distance values with larger distances selected from the remaining distance values in a preset proportion. For example, the first 10% or 20% or the like of distance values with larger distances are selected. Alternatively, a preset proportion of distances can also be randomly selected from the remaining distance values. The present embodiment does not limit this. In another possible implementation, outliers can not be discarded, and a preset proportion of distance values can be directly selected from the i-th second distance set.

[0132] With reference to the implementation process of determining the i-th first distance set based on the i-th frame, each frame of the N target point cloud frames is processed to obtain N second distance sets, and each second distance set is subjected to the first screening processing to obtain the N first distance sets.

[0133] In another possible implementation, in a case where the dispersion degree of the i-th second distance set satisfies the first preset condition, the first screening processing can not be performed. Then, the i-th second distance set is the i-th first distance set. Based on this, the N first distance sets can also be obtained.

[0134] Exemplarily, further, the first angle can be determined based on the N first distance sets and the installation height of the lidar. For ease of understanding, an example is exemplarily introduced below. For ease of description, the installation height of the lidar is denoted as installation height H.

[0135] Exemplarily, a ratio of the installation height H to each of the N first distance sets can be calculated to obtain N ratio sets. The i-th ratio set in the N ratio sets is determined based on the installation height H and the i-th first distance set. For example, taking obtaining the i-th ratio set as an example. In a specific implementation, the installation height H is divided by each distance value in the i-th first distance set to obtain the i-th ratio set. Exemplarily, the ratio of the installation height H to the distance value in the first distance set is actually equal to a sine function value of the pitch angle of the laser beam that detects the distance value. Therefore, the pitch angle of the laser beam can be obtained by solving the inverse sine function based on the ratio. For ease of understanding, an example is introduced below in combination with FIG. 6.

[0136] Exemplarily, the coordinate system shown in FIG. 6 is the coordinate system of the laser radar. It can be seen that the installation height H of the laser radar is the distance between the center point of the laser radar, which is also the origin O point of the coordinate system, and the ground. It is assumed that a laser beam 1 emitted by the laser radar hits a P1 point on the ground, and the laser beam reflected by the P1 point is received by the laser radar to form detection data of the P1 point. For example, the detection data of the P1 point is the detection data used to obtain any distance value in the i-th first distance set in the i-th point cloud frame. The distance value is, for example, the distance d1 shown in FIG. 6. In addition, in FIG. 6, the projection point of the detection point P1 on the xy plane is P1’. Based on the introduction of the pitch angle of the laser beam in FIG. 1, it can be known that the included angle θ1 between the line OP1 and the line OP1’ in FIG. 6 is the pitch angle of the laser beam 1 shown in FIG. 6. Since the line OP1’ is parallel to the ground, the size of the included angle between the laser beam 1, that is, the line OP1, and the ground is equal to the included angle θ1. Based on this, sin θ1 = H / d1. Then θ1 = acsin(H / d1).

[0137] Based on the above introduction, the inverse sine function is solved based on each ratio in the i-th ratio set. Each ratio can be solved to obtain a pitch angle, so that an i-th pitch angle set corresponding to the i-th ratio set can be obtained. Further, a mean value or a weighted average value of the pitch angles in the i-th pitch angle set can be calculated. For ease of subsequent description, the mean value or the weighted average value of the pitch angles in the pitch angle set is referred to as a third angle. The mean value or the weighted average value of the pitch angles in the i-th pitch angle set is referred to as the i-th third angle. The i-th third angle can be regarded as the actual pitch angle of the laser radar determined based on the i-th frame in the N target point cloud frames.

[0138] Based on the above processing procedure, each of the N ratio sets can be processed to obtain a set of pitch angles, and then N sets of pitch angles can be obtained. Each of the N sets of pitch angles can be processed to obtain a third angle. Then, N third angles can be obtained.

[0139] In a possible implementation, after the N third angles are obtained, a mean value or a weighted mean value of the N third angles can be directly calculated. The calculated mean value or weighted mean value is the first angle.

[0140] In another possible implementation, after the N third angles are obtained, the first angle can be determined in combination with the N sets of first distances. The following exemplary description is provided.

[0141] Exemplarily, in a possible implementation, a mean value of each of the N sets of first distances can be calculated, and a distance mean value is obtained for each set of first distances, thereby obtaining N distance mean values. Then, the N distance mean values and the N third angles are respectively subjected to second screening processing to obtain M1 distance mean values and M2 third angles. M1 and M2 are integers greater than 0 and less than N. The second screening processing includes discarding outliers. For example, the N distance mean values can be sorted in ascending order, and then a small value part and a large value part are discarded, and the remaining middle part data are the M1 distance mean values. For example, 10% of the small value part and 10% of the large value part are discarded, and the remaining 80% of the middle part data are the M1 distance mean values. It can be understood that this is only an example and does not constitute a limitation on the embodiments of the present application. Alternatively, in another implementation, the N distance mean values can also be subjected to outlier discarding processing by using a clustering algorithm, and the embodiments of the present application do not limit this.

[0142] Similarly, the N third angles can be sorted in ascending order, and then a small value part and a large value part are discarded, and the remaining middle part data are the M2 third angles. For example, 10% of the small value part and 10% of the large value part are discarded, and the remaining 80% of the middle part data are the M2 third angles. It can be understood that this is only an example and does not constitute a limitation on the embodiments of the present application. Alternatively, in another implementation, the N third angles can also be subjected to outlier discarding processing by using a clustering algorithm, and the embodiments of the present application do not limit this.

[0143] Then, in a case that the dispersion degree of the M1 distance means satisfies a second preset condition, and the dispersion degree of the M2 third angles satisfies a third preset condition, the mean of the M2 third angles is determined as the first angle. Conversely, if the dispersion degree of the M1 distance means does not satisfy the second preset condition, or the dispersion degree of the M2 third angles does not satisfy the third preset condition, it indicates that the reliability of this group of data is low. The subsequent pitch angle deviation judgment is no longer based on this group of data. Optionally, new point cloud data can be used to continue the pitch angle deviation judgment. This implementation can discard unsuitable data, reduce the use of abnormal data for subsequent calculation, improve the accuracy of the calculation result, and improve the stability and reliability of the overall scheme.

[0144] Exemplarily, the dispersion degree of the M1 distance means can be represented by the standard deviation, average difference or range of the M1 distance means. Similarly, the dispersion degree of the M2 third angles can be represented by the standard deviation, average difference or range of the M2 third angles. Taking the standard deviation as an example of the dispersion degree, the second preset condition is that the standard deviation of the M1 distance means is less than or equal to a second standard deviation threshold. Exemplarily, the value range of the second standard deviation threshold can be any value between 0.5 meters and 2 meters. It can be understood that the value range shown here is only an example and does not limit the embodiments of the present application. The third preset condition is that the standard deviation of the M2 third angles is less than or equal to a third standard deviation threshold. Exemplarily, the value range of the third standard deviation threshold can be any value between 0.1 degrees and 1 degree. It can be understood that the value range shown here is only an example and does not limit the embodiments of the present application. If the dispersion degree is represented by the average difference or range, the same applies to the standard deviation, which will not be repeated.

[0145] In another possible implementation, after obtaining the N third angles and the N first distance sets, the second screening process described above can not be performed. It can be directly determined whether the dispersion degree of the N distance means satisfies the second preset condition, and whether the dispersion degree of the N third angles satisfies the third preset condition. If both satisfy, the mean of the N third angles is determined as the first angle. Conversely, if at least one does not satisfy, it indicates that the reliability of this group of data is low. The subsequent pitch angle deviation judgment is no longer based on this group of data. Optionally, new point cloud data can be used to continue the pitch angle deviation judgment.

[0146] The implementation process of determining the first angle based on the target point cloud data and the installation height of the laser radar is exemplarily introduced below in a case that the target point cloud data is obtained based on one frame of point cloud of the laser radar.

[0147] Exemplarily, in a specific implementation, the point cloud data collected by the preset beam is selected from the one frame of point cloud in the manner of step S401 described above. Alternatively, the point cloud data collected by the preset beam within the preset horizontal detection angle range is selected from the one frame of point cloud. The selected data is the target point cloud data.

[0148] Exemplarily, a first distance set can be determined based on the target point cloud data. The first distance set includes distances of the detection points corresponding to one or more detection data included in the target point cloud data. Exemplarily, the distance of each detection point corresponding to the detection data included in the target point cloud data can be calculated first to obtain a second distance set. Exemplarily, the distance of the detection point is the distance between the detection point and the origin of the laser radar coordinate system. As known from the foregoing description, the detection data of the detection point includes three-dimensional coordinates of the detection point in the laser radar coordinate system. Based on the three-dimensional coordinates, the distance between the detection point and the origin of the laser radar coordinate system can be calculated, which will not be described herein.

[0149] In a possible embodiment, the dispersion degree of the second distance set can be calculated. If the dispersion degree of the second distance set meets the first preset condition, the first distance set is obtained by further performing a first screening process on the second distance set. If the dispersion degree of the second distance set does not meet the first preset condition, the i-th frame of data is discarded. For details, reference can be made to the description of the dispersion degree of the i-th second distance set and the related processing, which will not be described herein. This implementation can discard unsuitable data, reduce the use of abnormal data for subsequent calculation, improve the accuracy of the calculation result, and improve the stability and reliability of the overall scheme.

[0150] In another possible implementation, if the dispersion degree of the second distance set meets the first preset condition, the first screening process can not be performed. Then, the second distance set is the first distance set.

[0151] Exemplarily, further, a first angle can be determined based on the first distance set and the installation height of the laser radar. For ease of understanding, an exemplary description is given below. For ease of description, the installation height of the laser radar is denoted as installation height H.

[0152] Exemplarily, a ratio of the installation height H to each of the first distance sets can be calculated. The ratio of the installation height H to the distance value in the first distance set is actually equal to the sine function value of the pitch angle of the laser beam from which the distance value is detected. Therefore, the pitch angle of the laser beam can be obtained by solving the inverse sine function based on the ratio. The specific implementation can refer to the related description of FIG. 6, which is not repeated here. The inverse sine function is solved based on each calculated ratio. Each ratio can be solved to obtain a pitch angle, so that a set of pitch angles corresponding to the set of ratios can be obtained.

[0153] Further, a mean or weighted mean of the pitch angles in the set of pitch angles can be calculated. The mean obtained by solving is the first angle described above.

[0154] S403, determine the pitch angle deviation of the laser radar based on the first angle and a second angle; the second angle is determined based on a preset pitch angle of one or more laser beams directed to the ground.

[0155] Exemplarily, the second angle can be regarded as the pitch angle of the laser radar in an ideal installation scenario. Since the first angle is determined based on the point cloud data collected by the one or more laser beams directed to the ground (for example, the preset beam shown in FIG. 5), the second angle is also determined based on the preset pitch angle of the one or more laser beams directed to the ground.

[0156] Exemplarily, in a possible implementation, if the first angle is determined based on the point cloud data collected by the one laser beam directed to the ground. The second angle can be the pitch angle of the laser radar emitting the laser beam under ideal installation conditions.

[0157] Exemplarily, in another possible implementation, if the first angle is determined based on the point cloud data collected by the plurality of laser beams directed to the ground. The second angle can be the mean or weighted mean of the pitch angles of the laser radar emitting the plurality of laser beams under ideal installation conditions.

[0158] Exemplarily, in another possible implementation, the second angle can be preset in the processing device.

[0159] After the first angle and the second angle are obtained, the first angle and the second angle can be compared. If a difference between the first angle and the second angle is less than a threshold value, it is determined that the tilt angle of the laser radar does not deviate. Or, the deviation is small and can be ignored. Otherwise, if the difference between the first angle and the second angle is greater than or equal to the threshold value, it is determined that the tilt angle of the laser radar deviates. Exemplarily, the threshold value may, for example, be any value greater than or equal to 3 degrees. It can be understood that the value of the threshold value shown here is only an example and does not limit the embodiments of the present application.

[0160] In a possible implementation, after the processing device determines that the tilt angle of the laser radar deviates, the user can be warned. The user is informed that the laser radar is malfunctioning or is not suitable for continued use, and the like. Exemplarily, if the laser radar is a laser radar on a vehicle. Then, the user can be warned through a vehicle-mounted display screen or voice reminder, and the like. The embodiments of the present application do not limit this.

[0161] In a possible implementation, in order to further improve the accuracy of the judgment result of the tilt angle deviation of the laser radar, the point cloud data can be collected under the condition that the direction in which the laser radar emits the laser beam is unobstructed. Exemplarily, taking the application scenario that the laser radar is applied to a vehicle as an example. Before the target point cloud data is obtained in step S401, in an implementation, the position information and the speed information of the vehicle can be obtained first, and then it is determined whether the direction in which the laser radar emits the laser beam is unobstructed based on the position information and the speed information of the vehicle. Exemplarily, if the speed of the vehicle is greater than a preset speed threshold value, it is considered that the vehicle is driving outdoors. Driving outdoors and reaching a certain threshold value of speed can be considered as driving on a road with an unobstructed field of view. It is further determined that the direction in which the laser radar emits the laser beam is unobstructed. Then, the step S401 can be performed. Otherwise, if the speed of the vehicle is less than or equal to the preset speed threshold value, it is considered that the vehicle is driving on a road section with a large number of people or in a place with a large number of obstructions such as a parking lot. It is further considered that the direction in which the laser radar emits the laser beam is obstructed. Further, the position information of the vehicle can be used to further assist in judging the environment around the vehicle, and then it can be determined whether the direction in which the laser radar emits the laser beam is unobstructed. Exemplarily, the preset speed threshold value may, for example, be any value between 30 km / h and 60 km / h. It can be understood that the value of the threshold value shown here is only an example and does not limit the embodiments of the present application.

[0162] In another possible implementation, before the target point cloud data is acquired in step S401, the situation in the vehicle's advancing direction can also be analyzed through an image captured by a camera of the vehicle. For example, if it is analyzed through the image that there is no occlusion in a preset range in the vehicle's advancing direction, it can be determined that there is no occlusion in the direction in which the laser radar emits the laser beam. Conversely, if it is analyzed through the image that there is an occlusion in the preset range in the vehicle's advancing direction, it can be determined that there is an occlusion in the direction in which the laser radar emits the laser beam. Then, the above step S401 can be performed.

[0163] Exemplarily, in a possible implementation, if it is determined that there is an occlusion in the direction in which the laser radar emits the laser beam, the operations in steps S401 to S403 can not be performed temporarily.

[0164] It can be understood that the above determination of whether there is an occlusion in the direction in which the laser radar emits the laser beam is only an example and does not constitute a limitation on the embodiments of the present application. In other possible implementations, the vehicle can also communicate with other devices, such as a cloud server or other vehicles, to acquire the situation of the field of view in the advancing direction, and then determine whether there is an occlusion in the direction in which the laser radar emits the laser beam. The embodiments of the present application will not be described one by one.

[0165] In a possible implementation, since the present application is based on the installation height of the laser radar and the distance of the detected point on the ground to calculate the pitch angle of the laser beam emitted to the detected point. Then, the flatter the ground, the more accurate the pitch angle of the laser beam calculated, and thus the accuracy of the determination result of the pitch angle offset of the laser radar can be further improved. Based on this, before the target point cloud data is acquired in step S401, the flatness of the ground can be determined first. Taking the application scenario of the laser radar applied to the vehicle as an example.

[0166] Exemplarily, before the target point cloud data is acquired in step S401, the vertical angular velocity of the vehicle can be acquired first. Then, whether the road on which the vehicle travels is flat or not is determined based on the vertical angular velocity. Exemplarily, if the vertical angular velocity of the vehicle is greater than a preset angular velocity threshold 1, it is considered that the vehicle travels on a relatively bumpy road. Conversely, if the vertical angular velocity of the vehicle is less than or equal to the preset angular velocity threshold 1, it is considered that the vehicle travels on a flat road. Then, the above step S401 can be performed. Exemplarily, the preset angular velocity threshold 1 can be any value between 1 rad / s and 3 rad / s. It can be understood that the value of the threshold shown here is only an example and does not constitute a limitation on the embodiments of the present application.

[0167] Exemplarily, whether the vehicle travels on a flat road can also be determined by the horizontal angular velocity of the vehicle together with the vertical angular velocity of the vehicle. For example, if the vertical angular velocity of the vehicle is greater than a preset angular velocity threshold 1 and the horizontal angular velocity of the vehicle is greater than a preset angular velocity threshold 2, it is considered that the vehicle travels on a bumpy road. Exemplarily, the preset angular velocity threshold 2 can be any value between 1 rad / s and 3 rad / s. It can be understood that the values of the thresholds shown herein are only examples and do not limit the embodiments of the present application.

[0168] Further, the position information of the vehicle can also be used to further assist in determining the environment around the vehicle, so as to more accurately determine whether the vehicle travels on a flat road.

[0169] Exemplarily, in a possible implementation, if it is determined that the vehicle travels on a bumpy road, the operations of steps S401 to S403 can not be performed temporarily.

[0170] Exemplarily, in a possible implementation, the image of the ground can also be obtained by the vehicle camera, and then the flatness of the ground can be analyzed.

[0171] It can be understood that the above implementation of determining the flatness of the road traveled by the vehicle is only an example and does not limit the embodiments of the present application. In specific implementation, any way can be used to determine the flatness of the ground, and the embodiments of the present application do not limit this.

[0172] In a possible implementation, exemplary reference can be made to FIG. 7. FIG. 7 exemplarily shows a possible algorithm framework provided by the embodiments of the present application. The algorithm framework can be used to implement the method described in FIG. 4 and possible implementations thereof. As shown in FIG. 7, the algorithm framework can include a scene detection layer, a data processing layer and a result statistics layer.

[0173] Exemplarily, the scene detection layer can be used to determine whether there is no obstruction in the direction of the laser radar emitting the laser beam, and / or to determine whether the ground is flat, etc. The specific implementation process can be referred to the foregoing description, which is not repeated here.

[0174] Exemplarily, the data processing layer can be configured to process the point cloud frames collected by the lidar frame by frame, and finally output the distance mean value and the angle based on the processing of each frame. For example, the data processing layer can select the target point cloud frame for each frame of point cloud as described above, and then determine the first distance set based on the selected target point cloud frame. And calculate the distance mean value of the first distance set. Then, based on the first distance set and the installation height of the lidar, the corresponding third angle is determined. The third angle can be regarded as the actual pitch angle of the lidar determined based on the target point cloud frame. The distance mean value and the third angle are the output data of the data processing layer after processing each frame of point cloud. The specific implementation process of each step of the data processing layer can refer to the corresponding description of the preceding steps S401 and S402, which will not be described here.

[0175] Exemplarily, the result statistics layer can be configured to perform cumulative frame statistical processing on the processing results of each frame of point cloud output by the data processing layer. For example, two sliding windows of a certain length can be set to select the processing results for judging the pitch angle deviation of the lidar. Based on the description in the preceding step S401, the target point cloud data for judging the pitch angle deviation of the lidar is selected from one or N frames of point cloud of the lidar. The sliding window selects the processing results corresponding to the one or N frames. If the target point cloud data is selected based on the one frame of point cloud, the length of the sliding window can be one frame. If the target point cloud data is selected based on the N frames of point cloud, the length of the sliding window can be N frames. Based on the sliding window, one or N frames of point cloud corresponding processing results can be obtained from each frame of processing result output by the data processing layer. Further, based on the obtained one or N frames of point cloud corresponding processing results, i.e., based on the corresponding distance mean value and the third angle, or based on N distance mean values and N third angles, the first angle is determined. The specific implementation can refer to the related description of the preceding step S402, which will not be described here. Optionally, after determining the first angle, the result statistics layer can also judge whether the pitch angle of the lidar deviates based on the first angle and the second angle. The specific implementation can refer to the related description of the preceding step S403, which will not be described here.

[0176] To sum up, in the scheme of the embodiment of the application, the actual pitch angle (i.e., the first angle) of the line bundle can be calculated by selecting the point cloud data collected by the line bundle and the installation height of the laser radar, and then whether the pitch angle of the laser radar deviates can be determined based on the first angle and the preset pitch angle of the line bundle. It can not depend on any external device, laser point cloud semantic segmentation and feature extraction, etc. Based on the laser original point cloud, the pitch angle deviation in the vertical direction can be detected based on the existing laser radar field of view angle and laser installation height information, without any initial value requirement. Thus, the calculation resources are greatly saved, the calculation speed is fast, and the timeliness is strong. For the application scenario of vehicle intelligent driving, the pitch angle deviation of the laser radar can be found in time, and the safety risk is reduced. The safety and stability of the intelligent driving commercial vehicle are significantly improved. In addition, for laser radars with different field of view angles, appropriate point cloud data can be obtained through line bundle selection and / or horizontal detection angle selection for subsequent pitch angle deviation judgment. Thus, it can be applied to any field of view angle or any type of laser radar, and the application range is wide.

[0177] The above mainly introduces the method provided by the embodiment of the application. It can be understood that, in order to realize the above corresponding functions, each control unit or device contains the hardware structure and / or software module for executing each function. The units and steps of each example described in combination with the embodiments disclosed herein can be realized in the form of hardware or a combination of hardware and computer software. Whether a certain function is executed by hardware or computer software to drive hardware depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the application.

[0178] The embodiment of the application can divide the function modules of the device according to the above method examples, for example, each function module can be divided according to each function, or two or more functions can be integrated in one module. The above integrated module can be realized in the form of hardware or software function module. It should be noted that the division of the module in the embodiment of the application is illustrative, and is only a logical function division. When actually implemented, there can be another division method.

[0179] In the case of dividing each function module according to each function, the embodiment of the application also provides a device for implementing any one of the above methods, for example, providing a device including units (or means) for implementing each step in any one of the above methods.

[0180] For example, refer to FIG. 8, which is a structural schematic diagram of a processing apparatus 800 provided by an embodiment of the present application. The processing apparatus 800 shown in FIG. 8 can be a processing apparatus used to implement any of the above method embodiments. The processing apparatus 800 can include a detection unit 801 and an indication unit 802. Wherein:

[0181] The detection unit 801 is configured to detect a deviation of a pitch angle of a laser radar in a vehicle.

[0182] The indication unit 802 is configured to instruct the vehicle to exit an intelligent driving function if the detection unit 801 detects that the pitch angle of the laser radar deviates.

[0183] In a possible implementation, the detection unit 801 is specifically configured to:

[0184] obtain target point cloud data; the target point cloud data includes point cloud data collected by one or more laser line beams directed to the ground in one frame or N frames of point cloud data of the laser radar; N is an integer greater than 1;

[0185] determine a first angle based on the target point cloud data and an installation height of the laser radar;

[0186] determine a pitch angle deviation of the laser radar based on the first angle and a second angle; the second angle is determined based on a preset pitch angle of the one or more laser line beams directed to the ground.

[0187] In a possible implementation, the target point cloud data includes point cloud data collected within a preset horizontal detection angle range.

[0188] In a possible implementation, when the target point cloud data includes point cloud data collected by one or more laser line beams directed to the ground in the N frames of point cloud data, the detection unit 801 is specifically configured to:

[0189] determine N first distance sets respectively based on data of each frame in the target point cloud data; the i-th first distance set in the N first distance sets is determined based on data of the i-th frame in the target point cloud data, the i-th first distance set includes distances of detection points corresponding to one or more detection data included in the data of the i-th frame, and i is an integer from 1 to N;

[0190] determine the first angle based on the N first distance sets and the installation height of the laser radar.

[0191] In a possible implementation, the detection unit 801 is specifically configured to:

[0192] The detection unit 801 is specifically configured to:

[0193] In a case where the dispersion degree of the distances in the i th second distance set meets a first preset condition, the i th second distance set is subjected to a first screening process to determine the i th first distance set, so as to obtain the N first distance sets; the first screening process includes discarding outliers and / or selecting distances with numerical values meeting a second preset condition.

[0194] In a possible implementation, the detection unit 801 is specifically configured to:

[0195] The detection unit 801 is specifically configured to:

[0196] The detection unit 801 is specifically configured to:

[0197] In a possible implementation, the detection unit 801 is specifically configured to:

[0198] The detection unit 801 is specifically configured to:

[0199] The detection unit 801 is specifically configured to:

[0200] In a possible implementation, the detection unit 801 is specifically configured to:

[0201] The detection unit 801 is specifically configured to:

[0202] The detection unit 801 is specifically configured to:

[0203] In a case where the dispersion degree of the M 1 distance means meets a second preset condition and the dispersion degree of the M 2 third angles meets a third preset condition, the mean value of the M 2 third angles is determined as the first angle.

[0204] In a possible implementation, when the target point cloud data includes point cloud data collected by one or more laser line beams shot to the ground in a frame of point cloud data, the detection unit 801 is specifically configured to:

[0205] determine a first distance set based on the target point cloud data, the first distance set including distances of the detection points corresponding to the one or more pieces of detection data included in the target point cloud data;

[0206] determine the first angle based on the first distance set and the installation height of the laser radar.

[0207] In a possible implementation, the detection unit 801 is specifically configured to:

[0208] calculate distances of the detection points corresponding to each piece of detection data included in the target point cloud data based on the target point cloud data, to obtain a second distance set;

[0209] perform screening processing on the second distance set to obtain the first distance set when a discrete degree of the distances in the second distance set meets a first preset condition, the screening processing including discarding outliers and / or selecting distances with numerical values meeting a second preset condition.

[0210] In a possible implementation, the detection unit 801 is specifically configured to:

[0211] calculate a ratio of the installation height and each distance in the first distance set;

[0212] perform arccosine function solving based on each calculated ratio, to obtain an elevation angle set;

[0213] perform mean value solving on the elevation angle set to obtain the first angle.

[0214] In a possible implementation, the second angle is a mean value of the elevation angles of the one or more laser line beams shot to the ground, and the detection unit 801 is specifically configured to: if a difference between the first angle and the second angle is greater than or equal to a threshold value, determine that the elevation angle of the laser radar deviates.

[0215] In a possible implementation, the detection unit 801 is specifically configured to: before the target point cloud data is acquired, determine, based on position information and speed information of the vehicle or based on an image captured by a camera of the vehicle, that the vehicle travels on a road with an unobstructed field of view.

[0216] In a possible implementation, the detection unit 801 is specifically configured to: before the target point cloud data is acquired, determine, based on a body elevation angle of the vehicle, that the vehicle travels on a flat road.

[0217] The specific operations and advantages of each unit in the processing apparatus 800 shown in FIG. 8 can be found in the above description of FIG. 3A and its possible embodiments, and will not be repeated here.

[0218] It should be understood that the division of each unit in the above processing apparatus is only a logical division of functions, and in actual implementation, all or part of the units can be integrated into one physical entity, or can be physically separated. In addition, the units in the apparatus can be implemented in the form of processor calling software; for example, the apparatus includes a processor, the processor is connected with a memory, the memory stores instructions, and the processor calls the instructions stored in the memory to implement any of the above methods or to implement the functions of each unit of the apparatus, wherein the processor is, for example, a general processor such as a central processing unit (CPU) or a microprocessor, and the memory is a memory within the apparatus or a memory outside the apparatus. Alternatively, the units in the apparatus can be implemented in the form of hardware circuit, and the functions of part or all of the units can be implemented by designing the hardware circuit, which can be understood as one or more processors; for example, in one implementation, the hardware circuit is an application-specific integrated circuit (ASIC), and by designing the logical relationship of elements in the circuit, the functions of part or all of the units are implemented; for another example, in another implementation, the hardware circuit is a programmable logic device (PLD), and taking a field programmable gate array (FPGA) as an example, it can include a large number of logic gate circuits, and the connection relationship between the logic gate circuits can be configured by a configuration file, so as to implement the functions of part or all of the units. All units of the above apparatus can be implemented in the form of processor calling software, or all units can be implemented in the form of hardware circuit, or part of the units can be implemented in the form of processor calling software, and the remaining part can be implemented in the form of hardware circuit.

[0219] In embodiments of the present application, the processor is a circuit with data processing capability. In one implementation, the processor can be a circuit with instruction reading and running capability, such as a CPU, a microprocessor, a graphics processing unit (GPU) (which can be understood as a microprocessor), a digital signal processor (DSP), etc. In another implementation, the processor can implement certain functions through a logical relationship of hardware circuit, which is fixed or can be reconfigured, such as an ASIC or a PLD implemented hardware circuit, such as an FPGA. In a reconfigurable hardware circuit, the processor loads a configuration document to implement hardware circuit configuration. It can be understood that the processor loads instructions to implement the functions of the above units. In addition, it can also be a hardware circuit designed for artificial intelligence, which can be understood as an ASIC, such as a neural network processing unit (NPU), a tensor processing unit (TPU), a deep learning processing unit (DPU), etc.

[0220] It can be seen that each unit in the above device can be one or more processors (or processing circuits) configured to implement the above method, such as a CPU, a GPU, an NPU, a TPU, a DPU, a microprocessor, a DSP, an ASIC, an FPGA, or a combination of at least two of these processor forms.

[0221] In addition, each unit in the above device can be integrated together or can be independently implemented. In one implementation, these units are integrated together to implement a system-on-a-chip (SOC). The SOC can include at least one processor for implementing any of the above methods or implementing the functions of the units of the device. The at least one processor can be different, such as a CPU and an FPGA, a CPU and an artificial intelligence processor, a CPU and a GPU, etc.

[0222] Exemplarily, refer to FIG. 9, which is a structural schematic diagram of a possible physical entity of a processing device provided by the present application. The processing device 900 shown in FIG. 9 can be a processing device in the method described in the above embodiments. The processing device 900 includes a processor 901, a memory 902 and a communication interface 903. The processor 901, the communication interface 903 and the memory 902 can be connected to each other or connected to each other through a bus 904.

[0223] The memory 902 is configured to store computer programs and data of the processing apparatus 900. The memory 902 can include, but is not limited to, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a compact disc read-only memory (CD-ROM), or the like.

[0224] The software or program codes required by all or part of the functions of the processing apparatus in the above method embodiments are stored in the memory 902.

[0225] In a possible implementation, if the software or program codes required by part of the functions are stored in the memory 902, the processor 901 can cooperate with other components to complete other functions described in the method embodiments, in addition to calling the program codes in the memory 902 to implement part of the functions.

[0226] The communication interface 903 can be in a plurality of forms, and is configured to support the processing apparatus 900 to communicate, for example, to receive or send data or signals.

[0227] The processor 901 can be the CPU, GPU, NPU, TPU, DPU, microprocessor, DSP, ASIC, FPGA, or a combination of at least two of these forms, and the like. The processor 901 can be configured to read the program stored in the memory 902 and execute the operations performed by the processing apparatus in the above method embodiments.

[0228] The specific operations and advantages of each unit in the processing apparatus 900 shown in FIG. 9 can be referred to the corresponding description in the above method embodiments, which will not be repeated here.

[0229] The embodiments of the present application also provide a chip, which includes a processor and a memory. The memory is configured to store computer programs or computer instructions, and the processor is configured to execute the computer programs or computer instructions stored in the memory, so that the chip performs the operations performed by the processing apparatus in the above method embodiments.

[0230] The embodiment of the present application further provides a computer readable storage medium which stores a computer program or computer instructions, and the computer program or computer instructions are executed by a processor to implement the method implemented by the processing device in the above-mentioned Figure 3A and possible embodiments thereof. Exemplarily, the computer readable storage medium can include but is not limited to a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk and various media which can store program codes.

[0231] The embodiment of the present application further provides a computer program product, and when the computer program product is read and executed by a computer, the method implemented by the processing device in the above-mentioned Figure 3A and possible embodiments thereof will be executed. Exemplarily, the computer program product includes but is not limited to a computer program which can implement the method by a computer at runtime, a code or an electronic (digital) signal for transmitting computer program instruction code and the like.

[0232] In the present application, the terms "first", "second" and the like are used to distinguish the same items or similar items with substantially the same function and action, and it should be understood that there is no logical or time sequence relationship between "first", "second", "n", and the number and execution order are not limited. It should also be understood that although the following description uses the terms first, second, and the like to describe various elements, these elements should not be limited by the terms. These terms are only used to distinguish one element from another.

[0233] It should also be understood that in various embodiments of the present application, the size of the serial number of each process does not mean the order of execution, and the execution order of each process should be determined by its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0234] It should also be understood that the term "includes" (also referred to as "includes", "including", "comprises" and / or "comprising") when used in the present specification specifies the presence of stated features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0235] It should also be understood that, throughout the specification, language such as "one embodiment," "an embodiment," "a possible implementation," and the like refers to one of possible implementations of claimed subject matter. Furthermore, this language is not intended to limit the scope of claimed subject matter in any way. Indeed, any combination of one or more features, structures, or characteristics can be a possible implementation.

[0236] Finally, it should be noted that the above-described embodiments are merely possible implementations of the present application, but are not intended to limit the present application. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the technical solutions recorded in the foregoing embodiments can still be modified, or some or all of the technical features can be replaced by equivalents. Such modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the embodiments of the present application.

Claims

1. An intelligent driving control method, characterized by, The method comprises: detecting a deviation of a pitch angle of a laser radar in a vehicle; if the deviation of the pitch angle of the laser radar is detected, instructing the vehicle to exit an intelligent driving function.

2. The method of claim 1, wherein, The detection of the deviation of the pitch angle of the laser radar in the vehicle comprises: obtaining target point cloud data; the target point cloud data comprises point cloud data collected by one or more laser beams directed to the ground in one frame or N frames of point cloud data of the laser radar; N is an integer greater than 1; determining a first angle based on the target point cloud data and the installation height of the laser radar; determining the pitch angle deviation of the laser radar based on the first angle and a second angle; the second angle is determined based on a preset pitch angle of the one or more laser beams directed to the ground.

3. The method of claim 2, wherein, The target point cloud data comprises point cloud data collected within a preset horizontal detection angle range.

4. The method according to claim 2 or 3, characterized in that, In the case that the target point cloud data comprises point cloud data collected by one or more laser beams directed to the ground in the N frames of point cloud data, the determination of the first angle based on the target point cloud data and the installation height of the laser radar comprises: determining one first distance set based on the data of each frame of the target point cloud data respectively to obtain N first distance sets; the i-th first distance set in the N first distance sets is determined based on the data of the i-th frame of the target point cloud data, the i-th first distance set comprises distances of detection points corresponding to one or more detection data included in the data of the i-th frame, and i is an integer from 1 to N; determining the first angle based on the N first distance sets and the installation height of the laser radar.

5. The method of claim 4, wherein, The determination of one first distance set based on the data of each frame of the target point cloud data respectively to obtain N first distance sets comprises: calculating distances of detection points corresponding to each detection data included in the data of each frame based on the data of each frame respectively to obtain N second distance sets; the i-th second distance set in the N second distance sets is determined based on the data of the i-th frame; in the case that the dispersion degree of distances in the i-th second distance set meets a first preset condition, performing first screening processing on the i-th second distance set to determine the i-th first distance set to obtain the N first distance sets; the first screening processing comprises discarding outliers and / or selecting distances meeting a second preset condition in numerical size.

6. The method according to claim 4 or 5, characterized in that, The determination of the first angle based on the N first distance sets and the installation height of the laser radar comprises: determining N third angles based on the installation height and the N first distance sets, the i-th third angle in the N third angles is determined based on the installation height and the i-th first distance set; determining the first angle based on the N first distance sets and the N third angles.

7. The method of claim 6, wherein, The determination of N third angles based on the installation height and the N first distance sets comprises: calculating a ratio of the installation height and each distance in each of the first distance sets to obtain N ratio sets, wherein the i-th ratio set in the N ratio sets is determined based on the installation height and the i-th first distance set; performing inverse sine function solving based on each ratio in the i-th ratio set to obtain an i-th pitch angle set, and determining the i-th third angle by performing mean value solving on the i-th pitch angle set, to obtain the N third angles.

8. The method according to claim 6 or 7, characterized in that, The determining the first angle based on the N first distance sets and the N third angles comprises: performing mean value solving on each of the N first distance sets to obtain N distance means; performing second screening processing on the N distance means and the N third angles respectively to obtain M1 distance means and M2 third angles, wherein M1 and M2 are integers greater than 0 and less than N, and the second screening processing comprises discarding outliers. In a case where a discrete degree of the M1 distance means meets a second preset condition and a discrete degree of the M2 third angles meets a third preset condition, determining a mean value of the M2 third angles as the first angle.

9. The method of claim 2 or 3, wherein, In a case where the target point cloud data comprises point cloud data collected by one or more laser line beams directed to the ground in a frame of point cloud data, the determining the first angle based on the target point cloud data and the installation height of the lidar comprises: determining a first distance set based on the target point cloud data, wherein the first distance set comprises distances of detection points corresponding to one or more detection data included in the target point cloud data; determining the first angle based on the first distance set and the installation height of the lidar.

10. The method of claim 9, wherein, The determining the first distance set based on the target point cloud data comprises: calculating distances of detection points corresponding to each detection data included in the target point cloud data based on the target point cloud data to obtain a second distance set; in a case where a discrete degree of distances in the second distance set meets a first preset condition, performing screening processing on the second distance set to obtain the first distance set, wherein the screening processing comprises discarding outliers and / or selecting distances with numerical values meeting a second preset condition.

11. The method according to claim 9 or 10, characterized in that, The determining the first angle based on the first distance set and the installation height of the lidar comprises: calculating a ratio of the installation height and each distance in the first distance set; performing inverse sine function solving based on each calculated ratio to obtain a pitch angle set; performing mean value solving on the pitch angle set to obtain the first angle.

12. The method according to any one of claims 2-11, characterized in that, The second angle is a mean value of pitch angles of the one or more laser line beams directed to the ground. The determining the pitch angle deviation of the lidar based on the first angle and the second angle comprises: if a difference between the first angle and the second angle is greater than or equal to a threshold value, determining that the pitch angle of the lidar deviates.

13. The method according to any one of claims 2-12, characterized in that, Before the obtaining the target point cloud data, the method further comprises: Determine, based on position information and speed information of the vehicle, or based on an image captured by a camera of the vehicle, that the vehicle is driving on a road with no obstruction in the field of view.

14. The method according to any one of claims 2-13, characterized in that, Before the target point cloud data is acquired, the method further includes: Determine, based on a pitch angle of a body of the vehicle, that the vehicle is driving on a flat road.

15. An intelligent driving control device, characterized by comprising: The device includes: A detection unit configured to detect a deviation in the pitch angle of the laser radar in the vehicle; An indication unit configured to instruct the vehicle to exit the intelligent driving function if the deviation in the pitch angle of the laser radar is detected.

16. The apparatus of claim 15, wherein, The detection unit is specifically configured to: Acquire target point cloud data; the target point cloud data includes point cloud data collected by one or more laser line beams directed to the ground in one frame or N frames of point cloud data of the laser radar; N is an integer greater than 1; Determine a first angle based on the target point cloud data and an installation height of the laser radar; Determine a deviation in the pitch angle of the laser radar based on the first angle and a second angle; the second angle is determined based on a preset pitch angle of the one or more laser line beams directed to the ground.

17. The apparatus of claim 16, wherein, The target point cloud data includes point cloud data collected within a preset horizontal detection angle range.

18. The apparatus of claim 16 or 17, wherein, In the case where the target point cloud data includes point cloud data collected by one or more laser line beams directed to the ground in the N frames of point cloud data, the detection unit is specifically configured to: Determine a first distance set for each frame of data in the target point cloud data respectively to obtain N first distance sets; an i-th first distance set in the N first distance sets is determined based on i-th frame of data in the target point cloud data, the i-th first distance set includes distances of detection points corresponding to one or more detection data included in the i-th frame of data, i is an integer from 1 to N; Determine the first angle based on the N first distance sets and the installation height of the laser radar.

19. The apparatus of claim 18, wherein, The detection unit is specifically configured to: Calculate distances of detection points corresponding to each detection data included in each frame of data based on the each frame of data respectively to obtain N second distance sets; an i-th second distance set in the N second distance sets is determined based on the i-th frame of data; In the case where a discrete degree of distances in the i-th second distance set meets a first preset condition, perform first screening processing on the i-th second distance set to determine the i-th first distance set to obtain the N first distance sets; The first screening processing includes discarding outliers and / or selecting distances meeting a second preset condition in value size.

20. The apparatus of claim 18 or 19, wherein, The detection unit is specifically configured to: Determine N third angles based on the installation height and the N first distance sets, an i-th third angle in the N third angles is determined based on the installation height and the i-th first distance set; Determine the first angle based on the N first distance sets and the N third angles.

21. The apparatus of claim 20, wherein, The detection unit is specifically configured to: compute a ratio of the installation height and each distance in each of the first distance sets, to obtain N ratio sets; the i-th ratio set in the N ratio sets is determined based on the installation height and the i-th first distance set; perform inverse sine function solving based on each ratio in the i-th ratio set, to obtain an i-th pitch angle set, and determine the i-th third angle by performing mean value solving on the i-th pitch angle set, to obtain the N third angles.

22. The apparatus of claim 20 or 21, wherein, The detection unit is specifically configured to: perform mean value solving on each of the N first distance sets, to obtain N distance means; perform second screening processing on the N distance means and the N third angles respectively, to obtain M1 distance means and M2 third angles; M1 and M2 are integers greater than 0 and less than N; the second screening processing includes discarding outliers; in a case where a dispersion degree of the M1 distance means meets a second preset condition, and a dispersion degree of the M2 third angles meets a third preset condition, determine a mean value of the M2 third angles as the first angle.

23. The apparatus of claim 16 or 17, wherein, In a case where the target point cloud data includes point cloud data collected by one or more laser line beams directed to the ground in a frame of point cloud data, the detection unit is specifically configured to: determine a first distance set based on the target point cloud data; the first distance set includes distances of detection points corresponding to one or more detection data included in the target point cloud data; determine the first angle based on the first distance set and an installation height of the lidar.

24. The apparatus of claim 23, wherein, The detection unit is specifically configured to: compute distances of detection points corresponding to each detection data included in the target point cloud data based on the target point cloud data, to obtain a second distance set; in a case where a dispersion degree of distances in the second distance set meets a first preset condition, perform screening processing on the second distance set to obtain the first distance set; the screening processing includes discarding outliers and / or selecting distances with numerical values meeting a second preset condition.

25. The apparatus of claim 23 or 24, wherein, The detection unit is specifically configured to: compute a ratio of the installation height and each distance in the first distance set; perform inverse sine function solving based on each computed ratio, to obtain a pitch angle set; determine the first angle by performing mean value solving on the pitch angle set.

26. The apparatus of any one of claims 16-25, wherein, The second angle is a mean value of pitch angles of the one or more laser line beams directed to the ground; the detection unit is specifically configured to: if a difference between the first angle and the second angle is greater than or equal to a threshold value, determine that a pitch angle of the lidar deviates.

27. The device of any of claims 16-26, wherein, The detection unit is specifically configured to, before obtaining the target point cloud data, determine, based on position information and speed information of the vehicle, or based on an image captured by a camera of the vehicle, that the vehicle travels on a road with an unobstructed field of view.

28. The device of any of claims 16-27, wherein, The detection unit is specifically configured to, before obtaining the target point cloud data, determine, based on a body pitch angle of the vehicle, that the vehicle travels on a flat road.

29. A processing device in a vehicle, characterized by The processing device comprises a processor and a memory, wherein the memory is configured to store computer programs or computer instructions, and the processor is configured to execute the computer programs or computer instructions stored in the memory, so that the processing device executes the method according to any one of claims 1-14.

30. A vehicle characterized by The vehicle comprises a lidar and the processing device according to claim 29.

31. A computer readable storage medium, characterized in that, The computer readable storage medium stores computer programs or computer instructions, and the computer programs or computer instructions are executed by the processor to implement the method according to any one of claims 1-14.

32. A computer program product, characterised in that, When the computer program product is executed by the processor, the method according to any one of claims 1-14 is implemented.

Images