Angle measurement method, automatic driving method, electronic device, system and vehicle

Abstract

The present invention relates to the field of vehicle control technology, and specifically to a turning angle measurement method, an automatic driving method, an electronic device, a system and a vehicle. The method includes obtaining a curve model corresponding to a first vehicle in an initial state and a target point cloud of the first vehicle in a current state, wherein an acquisition device is fixed on a second vehicle, and the second vehicle is used to connect with the first vehicle; rotating the curve model to fit the target point cloud to determine a target curve model; and determining a relative rotation angle based on the relationship between a target normal vector of a key point in the target curve model and a normal vector of a key point in the curve model. The relative rotation angle is determined by fitting the curve model with the target point cloud. Since the curve model can characterize various forms of the first vehicle, the specific processing process is not affected by the vehicle form and acquisition conditions, thereby improving the accuracy of the measurement result.

Description

Technical Field

[0001] The present invention relates to the field of vehicle control technology, and in particular to a turning angle measurement method, an automatic driving method, an electronic device, a system and a vehicle. Background Art

[0002] At present, trailers play an important role in the transportation of people and goods in scenes such as highways, closed parks, ports and terminals, and mining plants. Overall, the trailer consists of two parts: the tractor in front and the trailer in the back. The trailer part is usually divided into multiple carriages. The tractor and the trailer are connected together by a connecting mechanism, and each carriage is also connected by a connecting mechanism. Due to its mechanical and electrical structural characteristics, when the trailer turns, the carriages (or the tractor and the trailer) will rotate relative to each other, resulting in a sweeping area, which poses a risk factor. For a tractor with multiple trailer carriages, the control of the relative turning angle of each trailer carriage is very important for the safe driving of the vehicle. Among them, the relative turning angle here is the turning angle of the rear trailer carriage relative to the front trailer carriage or tractor. Especially in the autonomous driving scenario, it is necessary to determine the angle between each trailer carriage, so as to obtain the relative position between the trailer carriages, and provide a reliable basis input for the path planning and driving control of the tractor.

[0003] The existing turning angle measurement is generally to use the camera of the front vehicle to shoot the connection between each trailer, fix a long chessboard or long straight line on the towing arm of the rear vehicle, and directly calculate the inclination of the marking line to get the turning angle of the trailer. However, if the camera is in strong light or dark environment, the measurement result of this method will be unstable, resulting in low measurement accuracy. Summary of the invention

[0004] In view of this, embodiments of the present invention provide a turning angle measurement method, an automatic driving method, an electronic device, a system and a vehicle to solve the problem of low turning angle measurement accuracy.

[0005] According to a first aspect, an embodiment of the present invention provides a rotation angle measurement method, comprising:

[0006] Acquire a curve model corresponding to the first vehicle in an initial state and a target point cloud of the first vehicle in a current state, wherein the acquisition device is fixed on a second vehicle, and the second vehicle is used to connect with the first vehicle;

[0007] Rotating the curve model to fit the target point cloud to determine a target curve model;

[0008] The relative rotation angle is determined based on the relationship between the target normal vector of the key point in the target curve model and the normal vector of the key point in the curve model.

[0009] The turning angle measurement method provided in the embodiment of the present invention adopts the method of fitting the curve model with the target point cloud to determine the relative turning angle. Since the curve model can characterize various forms of first vehicles, the specific processing process is not affected by the vehicle form and the acquisition conditions, thereby improving the accuracy of the measurement results.

[0010] In combination with the first aspect, in a first implementation of the first aspect, rotating the curve model to fit the target point cloud to determine the target curve model includes:

[0011] Matching the curve model with the target point cloud at key points, wherein the key points are points of a marker disposed on the first vehicle in the curve model and the target point cloud respectively;

[0012] Based on the matching result of the key point, the curve model is rotated along the key point to obtain an updated curve model;

[0013] The updated curve model is fitted with the target point cloud to determine the target curve model.

[0014] In the turning angle measurement method provided by the embodiment of the present invention, since the relative position of the marker in the first vehicle is fixed, the marker is used as a key point, which can ensure the fixation of the reference object and the reliability of the measurement result.

[0015] In combination with the first implementation manner of the first aspect, in the second implementation manner of the first aspect, the curve model is rotated along the key point based on the matching result of the key point to obtain an updated curve model, including:

[0016] Taking the key point as a starting point, dividing the rotatable interval of the curve model into intervals based on a first angle, and obtaining first branch boundaries of each first rotation interval;

[0017] The curve model is rotated at the first angle along the key point to obtain a first updated curve model, wherein the normal vector of the first updated curve model is located at the first branch boundary.

[0018] The rotation angle measurement method provided by the embodiment of the present invention divides the rotatable interval into intervals, updates the curve model by means of rotation, and subsequently gradually narrows the rotatable interval, thereby improving the measurement efficiency.

[0019] In combination with the second implementation manner of the first aspect, in the third implementation manner of the first aspect, fitting the updated curve model with the target point cloud to determine the target curve model includes:

[0020] For each of the first rotation intervals, calculating a first distance between the target point cloud and the first updated curve model;

[0021] determining a first rotation interval corresponding to the smallest first distance as a second rotation interval;

[0022] The first updated curve model is rotated based on the second rotation interval and the second angle to fit the target point cloud to determine the target curve model, and the second angle is smaller than the first angle.

[0023] The rotation angle measurement method provided by the embodiment of the present invention determines the target curve model by continuously narrowing the rotation range and angle, that is, gradually narrowing the fitting range, reducing the amount of data processing, and improving the measurement efficiency.

[0024] In combination with the first aspect, in a fourth implementation manner of the first aspect, obtaining a curve model corresponding to the first vehicle in an initial state includes:

[0025] Acquire initial point cloud data of the first vehicle in the initial state;

[0026] Performing curve fitting on the initial point cloud data to determine the curve model and normal vectors of key points in the curve model;

[0027] The acquisition device is corrected for zero bias based on the normal vector of the key point in the curve model to determine the zero bias angle.

[0028] The turning angle measurement method provided by the embodiment of the present invention not only obtains the curve model of the first vehicle in the initial state, but also performs zero bias correction to correct the installation error of the acquisition equipment, thereby further improving the accuracy of the measurement result.

[0029] In combination with the fourth implementation manner of the first aspect, in the fifth implementation manner of the first aspect, determining the relative rotation angle based on a relationship between a target normal vector of a key point in the target curve model and a normal vector of a key point in the curve model includes:

[0030] Obtaining a target distance between a target curve model and the target point cloud;

[0031] When the target distance is less than a preset value, calculating the angle between the target normal vector of the key point in the target curve model and the normal vector of the key point in the curve model;

[0032] The relative rotation angle is determined based on the zero deflection angle and the included angle.

[0033] The angle measurement method provided by the embodiment of the present invention, after determining the target curve model, further uses the target distance to screen the target curve model, and can further screen interference objects to improve the accuracy of the determined relative rotation angle.

[0034] In combination with the first aspect, in a sixth implementation of the first aspect, obtaining a target point cloud of the first vehicle in a current state includes:

[0035] Acquire the collected data of the first vehicle collected by the collection device in the current state to obtain the point cloud data in the current state;

[0036] Obtaining a point cloud threshold corresponding to the first vehicle;

[0037] Filtering the point cloud data based on the point cloud threshold;

[0038] The region of interest is extracted from the filtered point cloud to determine the target point cloud.

[0039] The turning angle measurement method provided in the embodiment of the present invention, after the point cloud threshold is screened, since the point cloud data scanned from the first vehicle will only appear within a certain range in the coordinate system of the acquisition device, an area of ​​interest is set according to this principle, and only the point cloud of the area of ​​interest is retained, thereby improving measurement efficiency.

[0040] According to a second aspect, an embodiment of the present invention provides an automatic driving method, including:

[0041] Acquire a relative rotation angle between a first vehicle and a second vehicle in the target vehicle, wherein the relative rotation angle is determined according to the rotation angle measurement method described in the first aspect of the present invention or any one of the embodiments of the first aspect;

[0042] The target vehicle is controlled based on the relative rotation angle.

[0043] The automatic driving method provided by the embodiment of the present invention ensures the reliability of automatic driving and improves the safety of the target vehicle because the control of the target vehicle is performed based on an accurate relative rotation angle.

[0044] According to the third aspect, an embodiment of the present invention provides an electronic device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the angle measurement method described in the first aspect or any one of the embodiments of the first aspect, or executing the automatic driving method described in the second aspect, by executing the computer instructions.

[0045] According to the fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable the computer to execute the angle measurement method described in the first aspect or any one of the embodiments of the first aspect, or to execute the automatic driving method described in the second aspect.

[0046] According to a fifth aspect, an embodiment of the present invention further provides a rotation angle measurement system, comprising:

[0047] A collection device is arranged on a second vehicle, the collection device is used to collect point cloud data corresponding to the first vehicle, and the second vehicle is used to connect to the first vehicle;

[0048] The electronic device described in the third aspect of the present invention is connected to the acquisition device.

[0049] In conjunction with the fifth aspect, in a first implementation of the fifth aspect, the system further includes:

[0050] A marker is arranged on the first vehicle, and the marker is used to determine key points of the curve model and the target point cloud.

[0051] In the turning angle measurement system provided by the embodiment of the present invention, since the relative position of the marker in the first vehicle is fixed, the measurement of the relative turning angle in combination with the marker can ensure the reliability of the measurement result.

[0052] According to a sixth aspect, an embodiment of the present invention further provides a vehicle, comprising:

[0053] a first vehicle and a second vehicle, wherein the second vehicle is configured to be connected to the first vehicle;

[0054] The fifth aspect of the present invention, or the rotation angle measurement system described in any embodiment of the fifth aspect.

[0055] It should be noted that, for the corresponding beneficial effects of the angle measurement system, electronic device, computer-readable storage medium and vehicle provided in the embodiments of the present invention, please refer to the description of the corresponding beneficial effects of the angle measurement method and the automatic driving method above, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0057] Figure 1 is a structural block diagram of a rotation angle measurement system according to an embodiment of the present invention;

[0058] Figure 2 is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention;

[0059] Figure 3a is a front view of a vehicle according to an embodiment of the present invention;

[0060] Figure 3b is a top view of a vehicle according to an embodiment of the present invention;

[0061] Figure 4 is a flow chart of a rotation angle measurement method according to an embodiment of the present invention;

[0062] Figure 5 is a schematic diagram of relative rotation angles according to an embodiment of the present invention;

[0063] Figure 6 is a flow chart of a rotation angle measurement method according to an embodiment of the present invention;

[0064] Figure 7 is a schematic diagram of key point matching according to an embodiment of the present invention;

[0065] Figure 8 is a schematic diagram of interval division according to an embodiment of the present invention;

[0066] Fig. 9 is a flow chart of a rotation angle measurement method according to an embodiment of the present invention;

[0067] Fig.10 is a schematic diagram of zero bias correction according to an embodiment of the present invention;

[0068] Fig.11 is a flowchart of an automatic driving method according to an embodiment of the present invention;

[0069] Fig.12 is a structural block diagram of a rotation angle measuring device according to an embodiment of the present invention;

[0070] Fig.13 is a structural block diagram of an automatic driving device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0071] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0072] The embodiment of the present invention provides a rotation angle measurement system, such as Figure 1 As shown, it includes a collection device 10 and an electronic device 20, wherein the collection device 10 is arranged on a second vehicle and is used to collect point cloud data corresponding to the first vehicle. The second vehicle is connected to the first vehicle.

[0073] It should be noted that the first vehicle and the second vehicle are two connected vehicle parts, for example, the first vehicle is a tractor and the second vehicle is a trailer; or the first vehicle is a trailer and the second vehicle is a tractor; or the first vehicle and the second vehicle are both trailer compartments. Along the lateral travel direction of the vehicles, the first vehicle is the front vehicle and the second vehicle is the rear vehicle; or the first vehicle is the rear vehicle and the second vehicle is the front vehicle. The specific form and position of the first vehicle and the second vehicle are set according to actual needs, and there is no restriction on them here, as long as the first vehicle and the second vehicle are connected to each other.

[0074] The acquisition device 10 includes but is not limited to a binocular camera, a lidar sensor, a depth camera, etc. The acquisition device 10 is arranged on the second vehicle, and is used to collect data of the first vehicle, and send the collected data to the electronic device 20. It should be noted that for different types of acquisition devices, the collection results are different. For example, for a lidar sensor, what is collected is point cloud data; for a camera, what is collected is image data. If the collected data obtained by the electronic device 20 is image data, it is necessary to convert the image data into point cloud data using camera parameters, etc. The electronic device 20 determines the relative rotation angle between the first vehicle and the second vehicle by executing the angle measurement method described in the embodiment of the present invention.

[0075] The electronic device 20 uses the curve model of the first vehicle in the initial state to fit the point cloud data in the current state to determine the relative rotation angle. The specific processing process will be described in detail in the embodiment of the rotation angle measurement method below.

[0076] In some optional embodiments, the turning angle measurement system further includes a marker, which is used to determine the key points in the curve model and the point cloud data of the first vehicle. The marker includes but is not limited to a reflective strip, or other identification that can be distinguished from the first vehicle. Since the relative position of the marker in the first vehicle is fixed, the measurement of the relative turning angle in combination with the marker can ensure the reliability of the measurement result.

[0077] The present invention also provides an electronic device having the following Fig.12 Angle measuring device as shown, or Fig.13 The autopilot device shown.

[0078] See also Figure 2 , Figure 2 is a schematic diagram of the structure of an electronic device provided by an optional embodiment of the present invention, such as Figure 2 As shown, the electronic device may include: at least one processor 201, such as a CPU (Central Processing Unit), at least one communication interface 203, a memory 204, and at least one communication bus 202. The communication bus 202 is used to realize the connection and communication between these components. The communication interface 203 may include a display screen (Display), a keyboard (Keyboard), and the optional communication interface 203 may also include a standard wired interface and a wireless interface. The memory 204 may be a high-speed RAM memory (Random Access Memory) or a non-volatile memory (non-volatile memory), such as at least one disk storage. The memory 204 may optionally be at least one storage device located away from the aforementioned processor 201. The processor 201 may be combined with Fig.12 or Fig.13 In the described device, the memory 204 stores an application program, and the processor 201 calls the program code stored in the memory 204 to execute any of the above method steps.

[0079] The communication bus 202 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The communication bus 202 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 2 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0080] Among them, the memory 204 may include a volatile memory (English: volatile memory), such as a random access memory (English: random-access memory, abbreviated: RAM); the memory may also include a non-volatile memory (English: non-volatile memory), such as a flash memory (English: flash memory), a hard disk drive (English: hard disk drive, abbreviated: HDD) or a solid-state drive (English: solid-state drive, abbreviated: SSD); the memory 204 may also include a combination of the above types of memory.

[0081] The processor 201 may be a central processing unit (CPU), a network processor (NP), or a combination of a CPU and a NP.

[0082] The processor 201 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The PLD may be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof.

[0083] Optionally, the memory 204 is also used to store program instructions. The processor 201 can call the program instructions to implement the angle measurement method or the automatic driving method as shown in any embodiment of the present application.

[0084] An embodiment of the present invention provides a vehicle, including a first vehicle, a second vehicle, and a turning angle measurement system. The second vehicle is used to connect to the first vehicle. For example, the first vehicle is a tractor, the second vehicle is a trailer, and so on. When the first vehicle is a tractor, the number of the second vehicle is not limited to it, and it is specifically set according to actual needs. For the specific structural details of the turning angle measurement system, please refer to the above description, which will not be repeated here.

[0085] For example, the first vehicle is a vehicle that provides power, and the second vehicle is a vehicle towed by the first vehicle. Figure 3a The front view of the vehicle is shown, and the vehicle includes a first vehicle 1, a second vehicle 2 and a turning angle measurement system. The turning angle measurement system includes a collection device 10, a marker 30 and an electronic device ( Figure 3a (not shown). Accordingly, Figure 3b A top view of the vehicle is shown.

[0086] In order to facilitate the subsequent processing of point cloud data, the marker is posted vertically at the midline position behind the first vehicle. The marker has two functions: 1) as a marker for the first vehicle, it prevents other vehicles from interfering with the detection in the field of view; 2) because its location is the key point of the curve model (the midpoint of the curve), the relative rotation angle can be calculated by comparing the normal vector at this point with the normal vector corresponding to the 0° angle during calibration.

[0087] Specifically, the curve model is a model of the first vehicle, which is obtained by collecting data from the first vehicle and modeling it; or, based on the collected data, a polynomial is used to fit it to obtain a curve model, etc. There is no limitation on the method of obtaining the curve model, and it is specifically set according to actual needs.

[0088] In a specific application example, the acquisition device is a laser radar sensor and the marker is a reflective strip. Figure 3a As shown, the reflective strip is vertically posted in the middle position behind the front car, and the laser radar sensor is installed in the middle position in front of the rear car. Among them, the measurement system includes a laser radar sensor, a marker, and the electronic device described above. The laser radar sensor is connected to the electronic device by a wiring harness, and the electronic device is connected to the electronic and electrical system on the vehicle through a CAN bus. The laser radar sensor is used to send the collected data to the electronic device, and the electronic device obtains the relative rotation angle through calculation, and sends the relative rotation angle to the electronic and electrical system through the CAN bus, so that the electronic and electrical system can perform corresponding control based on the relative rotation angle. The acquisition, transmission, calculation and output of point cloud data are realized through the above devices.

[0089] As an optional application scenario of an embodiment of the present invention, in the automatic driving scenario of the above-mentioned vehicle, the turning angle measurement device is a part of the environmental perception unit of the vehicle automatic driving system. The turning angle measurement device detects the relative rotation angle between two adjacent first vehicles and the second vehicle, and sends the detection result to the decision-making and control unit of the automatic driving system. The decision-making and control unit performs motion control on the first vehicle or the second vehicle based on the relative rotation angle to reduce the relative rotation angle and ensure automatic and safe driving of the vehicle.

[0090] As another optional application scenario of the embodiment of the present invention, during the driving process of the above-mentioned vehicles, since a tractor may be mounted with multiple trailer carriages, and since there is a rotation deviation between the carriages and between the trailer carriages and the tractor, the trajectory of the trailer behind the driver may appear in the blind spot during the driving process, which may cause driving safety problems. Based on this, during the driving process of the vehicle, the relative rotation angle between the two adjacent first vehicles and the second vehicle is measured based on the angle measurement method described in the embodiment of the present invention, and the relative rotation angle is displayed on the driving screen through the CAN network to prompt the driver to pay attention.

[0091] Of course, the above application scenario is only an example of an embodiment of the present invention and does not limit the protection scope of the present invention. It is specifically configured according to actual needs.

[0092] According to an embodiment of the present invention, a method for measuring an angle and an embodiment of an automatic driving method are provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in an order different from that shown here.

[0093] In this embodiment, a rotation angle measurement method is provided, which can be used for the above-mentioned electronic device. Figure 4 is a flow chart of a method for measuring a rotation angle according to an embodiment of the present invention. Figure 4 As shown, the process includes the following steps:

[0094] S11, obtaining a curve model corresponding to the first vehicle in an initial state and a target point cloud of the first vehicle in a current state.

[0095] Wherein, the acquisition device is fixed on a second vehicle, and the second vehicle is used to be connected to the first vehicle.

[0096] The initial state is that the first vehicle and the second vehicle are on the same straight line. Before measuring the turning angle, an initial calibration is performed to determine the curve model corresponding to the first vehicle. The curve model can be obtained by modeling it through image acquisition; it can also be obtained by collecting the initial point cloud data of the first vehicle and fitting the curve to obtain the curve model; it can also be obtained by the electronic device from a third-party device, for example, from the manufacturer of the first vehicle, etc. There is no restriction on the method of obtaining the curve model corresponding to the first vehicle, and it is set according to actual needs.

[0097] As mentioned above, the acquisition of the curve model belongs to the early calibration process, and the results obtained in the calibration process are not limited to the curve model, but can also include normal vectors of key points in the curve model and the like.

[0098] The current state is the driving state of the first vehicle and the second vehicle, that is, during the driving process of the first vehicle and the second vehicle, the acquisition device is used to collect data from the first vehicle to obtain the collected data. As mentioned above, due to the different types of acquisition devices, the acquisition device can be image data or point cloud data. Therefore, in order to facilitate subsequent processing, it is uniformly converted into point cloud data. The point cloud data used in the subsequent processing process can be the point cloud data obtained by conversion, or it can be obtained by filtering on the basis of the point cloud data, and so on. For the specific form of the acquisition device, please refer to the above description and will not be repeated here.

[0099] The method for obtaining the curve model will be described in detail below.

[0100] S12, rotating the curve model to fit the target point cloud to determine the target curve model.

[0101] As mentioned above, the curve model is obtained in the initial state, so the initial relative rotation angle between the first vehicle and the second vehicle in the initial state can also be obtained. Therefore, during the operation of the vehicle, although the relative rotation angle may change, the structure of the first vehicle remains unchanged, that is, the curve model of the first vehicle remains unchanged, and what changes is the position and posture of the curve model in the coordinate system of the acquisition device. Based on this, the electronic device rotates the curve model and fits the target point cloud, and determines the curve model closest to the target point cloud as the target curve model.

[0102] For the curve model closest to the target point cloud, each pair of curve models is rotated once to obtain an updated curve model. The electronic device calculates the distance between each point in the target point cloud and the updated curve model, and finally determines the updated curve model with the smallest distance as the target curve model.

[0103] The curve model may be rotated at preset angles, or may be rotated at a large angle first and then gradually reduced, that is, the target curve model may be determined by varying the rotation angle.

[0104] This step will be described in detail below.

[0105] S13, determining a relative rotation angle based on a relationship between a target normal vector of a key point in the target curve model and a normal vector of the key point in the curve model.

[0106] The key point is a point of a window, door, etc. in the first vehicle in the coordinate system of the acquisition device; the key point may also be a point of a marker set on the first vehicle in the coordinate system of the acquisition device, such as a reflective strip, etc.

[0107] For the target curve model, the electronic device first determines the position of the key point in the target curve model, and then calculates the target normal vector of the key point in the target curve model; for the curve model, the normal vector of its key point can be obtained in the initial calibration. The electronic device uses the angle between these two normal vectors to determine the relative rotation angle. Specifically, the target curve model is obtained by rotating the curve model. Therefore, the mathematical expression of the target curve model is calculated by combining the curve model with the rotation angle. Accordingly, the target normal vector of the target curve model can reflect the rotation angle. Based on this principle, the electronic device can determine the relative rotation angle using the angle between the target normal vector and the normal vector of the key point in the curve model.

[0108] like Figure 5 As shown, the dotted line in the first vehicle (front vehicle) represents the normal vector of the key point of the curve model, and the solid line with an arrow represents the target normal vector of the key point of the target curve model. The electronic device calculates the angle θ between the normal vector of the key point of the curve model and the target normal vector, that is, the relative rotation angle can be obtained.

[0109] This step will be described in detail below.

[0110] The turning angle measurement method provided in this embodiment uses a curve model and a target point cloud fitting method to determine the relative turning angle. Since the curve model can characterize various forms of first vehicles, the specific processing process is not affected by the vehicle form and acquisition conditions, thereby improving the accuracy of the measurement results.

[0111] In this embodiment, a rotation angle measurement method is provided, which can be used for the above-mentioned electronic device. Figure 6 is a flow chart of a method for measuring a rotation angle according to an embodiment of the present invention. Figure 6 As shown, the process includes the following steps:

[0112] S21, obtaining a curve model corresponding to the first vehicle in an initial state and a target point cloud of the first vehicle in a current state.

[0113] Wherein, the acquisition device is fixed on a second vehicle, and the second vehicle is used to be connected to the first vehicle.

[0114] For details, please see Figure 4 S11 of the illustrated embodiment will not be described in detail here.

[0115] S22, rotating the curve model to fit the target point cloud to determine the target curve model.

[0116] Specifically, the above S22 includes:

[0117] S221, performing key point matching between the curve model and the target point cloud.

[0118] The key points are points of the markers set on the first vehicle in the curve model and the target point cloud respectively.

[0119] This key point is used as the origin of the subsequent rotation of the curve model, that is, the curve model is rotated with this key point as the origin to obtain an updated curve model. The key point matching is to determine the position of the key point in the curve model and the target point cloud respectively, and match the two key points. Figure 7 As shown, key points are determined in the curve model and the target point cloud respectively.

[0120] S222: Based on the matching results of the key points, the curve model is rotated along the key points to obtain an updated curve model.

[0121] After determining the key point, the curve model is rotated along the key point, and an updated curve model is obtained each time it is rotated by a preset angle. For example, the accuracy of angle measurement is 0.1°. A circle of 360° is divided into 3600 parts, and 3600 updated curve models can be obtained each time it is rotated by 0.1°. Subsequently, by comparing the distance between these 3600 updated curve models and the target point cloud, it is possible to determine which angle is the best angle, thereby determining the target curve model.

[0122] In some optional implementations, the above S222 includes:

[0123] (1) Taking the key point as the starting point, the rotatable interval of the curve model is divided into intervals based on the first angle to obtain the first branch boundary of each first rotation interval.

[0124] (2) Rotating the curve model along the key point at a first angle to obtain a first updated curve model, wherein the normal vector of the first updated curve model is located at the first branch boundary.

[0125] Specifically, the rotatable interval can be pre-set or a full-angle interval, that is, a 360° rotation interval. The electronic device takes the key point as the starting point and first divides the rotatable angle into intervals based on the first angle. For example, the first angle is 180°, and the 360° rotation interval is divided into two first rotation intervals. Each first rotation interval has a corresponding first branch boundary. The electronic device rotates the first curve model 180° along the key point to obtain the corresponding first updated curve model.

[0126] In a specific example, if Figure 8As shown, a rectangular coordinate system is established at the key point of the curve model B, and the curve model BN is updated. The coordinate system is divided into branches, and a circle of 360° is divided into 4 branches, each branch contains a 90° interval. The curve model BN is rotated counterclockwise around the key point by 90°, 180°, 270° and 360° in sequence, so that the normal vector at the key point of the updated curve model is located at the branch boundary. Among them, after the rotation β, the curve model satisfies:

[0127] x=x0cosβ-y0sinβ

[0128] y=x0sinβ+y0cosβ

[0129] Among them, the curve model corresponding to x0 and y0 is the curve model on the newly established coordinate system. Specifically, the curve model in S21 is on the coordinate system of the acquisition device, while x0 and y0 here are on the newly established coordinate system. The transformation of the two coordinate systems is equivalent to translating the origin of the coordinate system of the acquisition device from the position origin of the acquisition device to the midpoint (key point) of the S21 curve model. Ideally, the two origins will coincide. However, due to the installation error of the acquisition device during installation, its x-axis direction is not ideally forward, and the angle between the normal vector of the key point and the x-axis of the acquisition device is θ0. This angle is defined as zero deflection, that is: in the coordinate system of the acquisition device, when the normal vector of the key point has an angle θ0 with the x-axis of the acquisition device, the vehicle is in a straight-line driving state. Based on this, after the translation, it is also necessary to rotate the XOY plane counterclockwise around the origin by an angle of θ0, which is the newly established coordinate system.

[0130] By dividing the rotatable interval into intervals, the curve model is updated by rotating, and the rotatable interval is gradually narrowed, thereby improving the measurement efficiency.

[0131] S223, fitting the updated curve model with the target point cloud to determine the target curve model.

[0132] As described above, when fitting, the electronic device may calculate the distance between each point cloud in the target point cloud and the updated curve model. For example, after the above-mentioned processing of S222, two updated curve models are obtained, then the electronic device calculates the distance between each point cloud in the target point cloud and the two updated curve models respectively. In order to characterize the distances corresponding to all points in the target point cloud, the electronic device may calculate the statistical values ​​of each distance. For example, the sum of the squares of the distances from each point in the point cloud data to the updated curve model may be calculated, that is:

[0133]

[0134] Among them, t nis the nth point cloud in the target point cloud, y n is the corresponding point in the updated curve model. Specifically, the distance can be solved using the binary interval method or other methods. Calculate the distance of each position of the curve in the four rotations respectively, and retain the position with the minimum loss function for the next rotation.

[0135] In some optional implementations, the above S223 includes:

[0136] (1) For each first rotation interval, a first distance between the target point cloud and the first updated curve model is calculated.

[0137] (2) The first rotation interval corresponding to the minimum first distance is determined as the second rotation interval.

[0138] (3) Rotating the first updated curve model based on a second rotation interval and a second angle to fit the target point cloud to determine a target curve model, wherein the second angle is smaller than the first angle.

[0139] As described above, the electronic device compares the sizes of each first distance, and the first distance is used to represent the distance between all point clouds in the target point cloud and the updated curve model. Since the updated curve model corresponds to the first rotation interval, one first distance corresponds to one first rotation interval. Therefore, the electronic device compares the sizes of each first distance and determines the first rotation interval corresponding to the smallest first distance as the second rotation interval, which is used for the next rotation.

[0140] After obtaining the second rotation interval, the second angle is combined to perform rotation and distance calculation again, and so on, until the Nth angle is less than the rotation angle accuracy value, then the updated curve model obtained at this time is the target curve model.

[0141] Continuing with the above example, within the boundaries on both sides of the second rotation interval (180° interval), continue to divide into 4 sub-branches, each sub-branch contains 45°. Repeat the rotation of the curve model 4 times so that the normal vectors of its key points fall on the boundaries of the new 4 sub-branches, and calculate the distances corresponding to these 4 positions. By comparing the size of the distances, retain the position of the new minimum distance, and continue to divide into 4 new branches within the boundaries on both sides, each sub-branch contains 22.5°, and continue the same calculation process. Repeat the above process continuously, such as Figure 8As shown in the figure, until the Nth angle of the last division is less than the preset trailer angle precision value, the branch division is stopped, and the position of the curve rotation at this time is set as the determined target curve model. Among them, each cycle will continue to reduce this angle (the size of the angle is divided by 2 each cycle). Because each angle is determined according to the rules, not randomly, it can ensure that the normal vector of the key point obtained each time falls on the branch boundary.

[0142] The target curve model is determined by continuously reducing the rotation range and angle, that is, gradually reducing the fitting interval, reducing the amount of data processing, and improving measurement efficiency.

[0143] S23, determining a relative rotation angle based on a relationship between a target normal vector of a key point in the target curve model and a normal vector of the key point in the curve model.

[0144] For details, please see Figure 4 S13 of the illustrated embodiment will not be described in detail here.

[0145] In the turning angle measurement method provided in this embodiment, since the relative position of the marker in the first vehicle is fixed, the marker is used as a key point, which can ensure the fixation of the reference object and the reliability of the measurement result.

[0146] In this embodiment, a rotation angle measurement method is provided, which can be used for the above-mentioned electronic device. Fig. 9 is a flow chart of a method for measuring a rotation angle according to an embodiment of the present invention. Fig. 9 As shown, the process includes the following steps:

[0147] S31, obtaining a curve model corresponding to the first vehicle in an initial state and a target point cloud of the first vehicle in a current state.

[0148] Wherein, the acquisition device is fixed on a second vehicle, and the second vehicle is used to be connected to the first vehicle.

[0149] Specifically, the above S31 includes:

[0150] S311, obtaining initial point cloud data of the first vehicle in an initial state.

[0151] The vehicle is parked in a straight-line driving state, and the acquisition device acquires the initial point cloud data of the first vehicle in the initial state. After the initial point cloud data is obtained, it can also be filtered to remove interference before subsequent curve meshing.

[0152] S312, performing curve fitting on the initial point cloud data to determine the curve model and the normal vectors of key points in the curve model.

[0153] The electronic device performs polynomial curve fitting on the initial point cloud data. Due to the limitations of the machining process, the vertical projection of the car body surface of the first vehicle is mostly a combination of different arcs. The parameters of the curve model are recorded, and the points corresponding to the markers are set as the key points of the curve. Accordingly, the normal vectors of the key points in the curve model are obtained.

[0154] The recorded curve model is as follows:

[0155]

[0156] S313, performing zero bias correction on the acquisition device based on the normal vector of the key point in the curve model to determine the zero bias angle.

[0157] like Fig.10 As shown, the vehicle is parked in a straight-line driving state, and the normal vector direction of the key point is set to the 0° turning direction of the trailer, that is, Fig.10 The arrow direction of the front vehicle is shown in the figure. At this time, in the coordinate system of the acquisition device, the normal vector direction and the x-axis of the laser radar will form an angle θ0. This angle is caused by the installation error of the acquisition device. Therefore, it is necessary to determine this angle during the initial calibration, which is called the zero deflection angle.

[0158] S314, acquiring the collection data of the first vehicle collected by the collection device in the current state to obtain the point cloud data in the current state.

[0159] The driving process of the vehicle is the current state, during which the acquisition device collects the acquisition data of the first vehicle and converts it into point cloud data. The point cloud data may contain point cloud data of interference objects. In order to ensure the accuracy of subsequent data processing, the point cloud data of the interference objects needs to be removed. For the removal of the interference point cloud data, a point cloud threshold can be set and the collected point cloud data can be compared with the point cloud threshold.

[0160] In some optional embodiments, such as Figure 3a As shown, since the acquisition device has a certain acquisition range, a region of interest can be set, and only the point cloud data within the region of interest is subsequently processed, thereby reducing the amount of data processing.

[0161] After filtering the point cloud data, the electronic device obtains a target point cloud, which is subsequently used to fit the curve model to determine the target curve model.

[0162] S315: Obtain a point cloud threshold corresponding to the first vehicle.

[0163] The point cloud threshold may be the reflection intensity of the first vehicle and the marker, or may be a pre-set threshold.

[0164] S316, filtering the point cloud data based on the point cloud threshold.

[0165] The point cloud collected by the acquisition device is preprocessed. First, filtering and noise reduction are performed to remove noise points and outliers in the point cloud. Then, point cloud clustering is performed to divide the point cloud into clusters. The point cloud clusters with reflection intensity higher than the point cloud threshold δ in the point cloud data are judged as interference objects, and the point cloud clusters higher than the point cloud threshold δ are used for subsequent calculation and processing.

[0166] S317, extracting the region of interest from the filtered point cloud to determine the target point cloud.

[0167] Due to the structural characteristics of the trailer, the front and rear vehicles only rotate around the connection point. At this time, the point cloud scanned by the first vehicle will only appear within a certain range in the coordinate system of the acquisition device. Therefore, according to this principle, the point cloud area of ​​interest is set, and only the point cloud in the area of ​​interest is retained to obtain the target point cloud to improve the efficiency of subsequent algorithms.

[0168] S32, rotating the curve model to fit the target point cloud to determine the target curve model.

[0169] For details, please see Figure 6 S23 of the illustrated embodiment will not be described in detail here.

[0170] S33, determining a relative rotation angle based on a relationship between a target normal vector of a key point in the target curve model and a normal vector of the key point in the curve model.

[0171] Specifically, the above S33 includes:

[0172] S331, obtaining a target distance between a target curve model and a target point cloud.

[0173] The target distance between the target curve model and the target point cloud is calculated in the target curve model determined in S32 above. The electronic device compares the target distance with the preset value. If the target distance is greater than or equal to the preset value, it means that the target point cloud is an interference object, that is, the interference object that has not been filtered out in S316 above, and no subsequent calculation is performed. The purpose of comparing the target distance with the preset value after fitting is to determine whether the object is a missed interference object by comparing the target distance with the preset value, and remove it again. If it is an interference object, the target distance obtained in S33 will be significantly larger.

[0174] S332, when the target distance is less than a preset value, calculating the angle between the target normal vector of the key point in the target curve model and the normal vector of the key point in the curve model.

[0175] like Figure 5As shown, when the target distance is less than a preset value, the angle θ between the target normal vector of the key point in the target curve model and the normal vector of the key point in the curve model is calculated.

[0176] S333, determining the relative rotation angle based on the zero deflection angle and the included angle.

[0177] The relative rotation angle Heading is expressed by the following formula:

[0178] Heading=θ-θ0

[0179] Among them, θ is the included angle and θ0 is the zero deflection angle.

[0180] The angle measurement method provided in this embodiment not only obtains the curve model of the first vehicle in the initial state, but also performs zero bias correction to correct the installation error of the acquisition device, thereby further improving the accuracy of the measurement result. After the point cloud threshold is screened, since the point cloud data scanned to the first vehicle will only appear within a certain range under the acquisition device coordinate system, the region of interest is set according to this principle, and only the point cloud of the region of interest is retained to improve the measurement efficiency. After determining the target curve model, the target curve model is also screened using the target distance, which can further screen interference objects and improve the accuracy of the determined relative rotation angle.

[0181] The turning angle measurement method provided by the embodiment of the present invention is a method for measuring the turning angle by using an improved polynomial curve fitting point cloud algorithm. The method is applicable to vehicles with different curved surface shapes, can eliminate interference from other vehicles in the field of view of the acquisition device, and has an efficient and robust algorithm.

[0182] In this embodiment, an automatic driving method is provided, which can be used for the above-mentioned electronic device. Fig.11 is a flow chart of a method for measuring a rotation angle according to an embodiment of the present invention. Fig.11 As shown, the process includes the following steps:

[0183] S41, obtaining a relative rotation angle between a first vehicle and a second vehicle in the target vehicles.

[0184] Wherein, the relative rotation angle is determined according to the rotation angle measurement method described in any of the above embodiments.

[0185] Regarding the specific structure of the target vehicle and the specific processing process of the turning angle measurement method, please refer to the corresponding embodiments described above and will not be repeated here.

[0186] S42, controlling the target vehicle based on the relative rotation angle.

[0187] After the electronic device determines the relative rotation angle, when controlling the target vehicle, if it is automatic driving, the control strategy is determined based on the relative rotation angle; if it is non-automatic driving, the relative rotation angle can be displayed on the driving screen in the form of a graphic interface so that the driver can know the relative rotation angle between two adjacent vehicles in the target vehicle in real time.

[0188] The automatic driving method provided in this embodiment ensures the reliability of automatic driving and improves the safety of the target vehicle because the control of the target vehicle is performed based on an accurate relative rotation angle.

[0189] In this embodiment, a turning angle measuring device and an automatic driving device are also provided, which are used to implement the above-mentioned embodiments and preferred implementation modes, and will not be repeated hereafter. As used below, the term "module" may be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0190] This embodiment provides a rotation angle measuring device, such as Fig.12 As shown, including:

[0191] A first acquisition module 51 is used to acquire a curve model corresponding to a first vehicle in an initial state and a target point cloud of the first vehicle in a current state, wherein the acquisition device is fixed on a second vehicle, and the second vehicle is used to be connected to the first vehicle;

[0192] A fitting module 52, configured to rotate the curve model to fit the target point cloud to determine a target curve model;

[0193] The determination module 53 is used to determine the relative rotation angle based on the relationship between the target normal vector of the key point in the target curve model and the normal vector of the key point in the curve model.

[0194] In some optional implementations, the fitting module 52 includes:

[0195] A matching unit, used for matching the curve model with the target point cloud for key points, wherein the key points are points of a marker disposed on the first vehicle in the curve model and the target point cloud respectively;

[0196] A rotation unit, configured to rotate the curve model along the key point based on the matching result of the key point to obtain an updated curve model;

[0197] The first fitting unit is used to fit the updated curve model with the target point cloud to determine the target curve model.

[0198] In some optional embodiments, the rotating unit includes:

[0199] a division subunit, configured to divide the rotatable interval of the curve model based on the first angle and taking the key point as a starting point, so as to obtain a first branch boundary of each first rotation interval;

[0200] The first rotation subunit is used to rotate the curve model along the key point at the first angle to obtain a first updated curve model, wherein the normal vector of the first updated curve model is located at the first branch boundary.

[0201] In some optional embodiments, the fitting unit includes:

[0202] a calculation subunit, configured to calculate, for each of the first rotation intervals, a first distance between the target point cloud and the first updated curve model;

[0203] a determination subunit, configured to determine a first rotation interval corresponding to the smallest first distance as a second rotation interval;

[0204] The second rotation subunit is used to rotate the first updated curve model based on the second rotation interval and a second angle to fit the target point cloud to determine the target curve model, and the second angle is smaller than the first angle.

[0205] In some optional implementations, the first acquisition module 51 includes:

[0206] A first acquisition unit, configured to acquire initial point cloud data of the first vehicle in the initial state;

[0207] A second fitting unit, used for performing curve fitting on the initial point cloud data to determine the curve model and normal vectors of key points in the curve model;

[0208] The correction unit is used to perform zero bias correction on the acquisition device based on the normal vector of the key point in the curve model to determine the zero bias angle.

[0209] In some optional implementations, the determination module 54 includes:

[0210] A second acquisition unit, used to acquire a target distance between a target curve model and the target point cloud;

[0211] a calculation unit, configured to calculate an angle between a target normal vector of a key point in the target curve model and a normal vector of the key point in the curve model when the target distance is less than a preset value;

[0212] A determination unit is used to determine the relative rotation angle based on the zero deflection angle and the included angle.

[0213] In some optional implementations, the first acquisition module 51 includes:

[0214] A third acquisition unit is used to acquire the acquisition data of the first vehicle acquired by the acquisition device in the current state to obtain the point cloud data in the current state;

[0215] A fourth acquisition unit, configured to acquire a point cloud threshold corresponding to the first vehicle;

[0216] A filtering unit, configured to filter the point cloud data based on the point cloud threshold;

[0217] The extraction unit is used to extract the region of interest from the filtered point cloud to determine the target point cloud.

[0218] This embodiment also provides an automatic driving device, such as Fig.13 As shown, including:

[0219] A second acquisition module 61 is used to acquire a relative rotation angle between a first vehicle and a second vehicle in the target vehicle, wherein the relative rotation angle is determined according to the above-mentioned rotation angle measurement method;

[0220] The control module 62 is used to control the target vehicle based on the relative rotation angle.

[0221] The angle measurement device and the automatic driving device in this embodiment are presented in the form of functional units, where the units refer to ASIC circuits, processors and memories that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0222] The further functional description of each of the above modules is the same as that of the above corresponding embodiments and will not be repeated here.

[0223] The embodiment of the present invention further provides a non-transitory computer storage medium, wherein the computer storage medium stores computer executable instructions, and the computer executable instructions can execute the angle measurement method or the automatic driving method in any of the above method embodiments. The storage medium may be a disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), a flash memory, a hard disk (HDD) or a solid-state drive (SSD), etc.; the storage medium may also include a combination of the above types of memories.

[0224] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A method for measuring a rotation angle, characterized in that: include: Acquire a curve model corresponding to the first vehicle in an initial state and a target point cloud of the first vehicle in a current state, wherein the acquisition device is fixed on a second vehicle, and the second vehicle is used to connect with the first vehicle; the initial state is a state in which the first vehicle and the second vehicle are on the same straight line, and the current state is a driving state of the first vehicle and the second vehicle; Rotating the curve model to fit the target point cloud to determine a target curve model; Determining a relative rotation angle between the first vehicle and the second vehicle based on a relationship between a target normal vector of a key point in the target curve model and a normal vector of a key point in the curve model; The step of rotating the curve model to fit the target point cloud to determine the target curve model includes: Matching the curve model with the target point cloud at key points, wherein the key points are points of a marker disposed on the first vehicle in the curve model and the target point cloud respectively; Based on the matching result of the key point, the curve model is rotated along the key point to obtain an updated curve model; The updated curve model is fitted with the target point cloud to determine the target curve model.

2. The method according to claim 1, characterized in that The step of rotating the curve model along the key point based on the matching result of the key point to obtain an updated curve model includes: Taking the key point as a starting point, dividing the rotatable interval of the curve model into intervals based on a first angle, and obtaining first branch boundaries of each first rotation interval; The curve model is rotated at the first angle along the key point to obtain a first updated curve model, wherein the normal vector of the first updated curve model is located at the first branch boundary.

3. The method according to claim 2, characterized in that The step of fitting the updated curve model with the target point cloud to determine the target curve model includes: For each of the first rotation intervals, calculating a first distance between the target point cloud and the first updated curve model; determining a first rotation interval corresponding to the smallest first distance as a second rotation interval; The first updated curve model is rotated based on the second rotation interval and the second angle to fit the target point cloud to determine the target curve model, and the second angle is smaller than the first angle.

4. The method according to claim 1, characterized in that The obtaining of the curve model corresponding to the first vehicle in the initial state includes: Acquire initial point cloud data of the first vehicle in the initial state; Performing curve fitting on the initial point cloud data to determine the curve model and normal vectors of key points in the curve model; The acquisition device is corrected for zero bias based on the normal vector of the key point in the curve model to determine the zero bias angle.

5. The method according to claim 4, characterized in that The determining the relative rotation angle between the first vehicle and the second vehicle based on the relationship between the target normal vector of the key point in the target curve model and the normal vector of the key point in the curve model includes: Acquire a target distance between a target curve model and the target point cloud; When the target distance is less than a preset value, calculating the angle between the target normal vector of the key point in the target curve model and the normal vector of the key point in the curve model; The relative rotation angle is determined based on the zero deflection angle and the included angle.

6. The method according to claim 1, characterized in that Acquiring a target point cloud of the first vehicle in the current state includes: Acquire the collected data of the first vehicle collected by the collection device in the current state to obtain the point cloud data in the current state; Obtaining a point cloud threshold corresponding to the first vehicle; Filtering the point cloud data based on the point cloud threshold; The region of interest is extracted from the filtered point cloud to determine the target point cloud.

7. An automatic driving method, characterized in that: include: Acquire a relative rotation angle between a first vehicle and a second vehicle in the target vehicle, wherein the relative rotation angle is determined according to the rotation angle measurement method according to any one of claims 1 to 6; The target vehicle is controlled based on the relative rotation angle.

8. An electronic device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the angle measurement method described in any one of claims 1 to 6, or executes the automatic driving method described in claim 7 by executing the computer instructions.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the angle measurement method described in any one of claims 1 to 6, or to execute the automatic driving method described in claim 7.

10. A rotation angle measurement system, characterized in that: include: A collection device, arranged on a second vehicle, the collection device is used to collect point cloud data corresponding to the first vehicle, and the second vehicle is used to be connected to the first vehicle; The electronic device as described in claim 8 is connected to the acquisition device.

11. The system according to claim 10, characterized in that Also includes: A marker is arranged on the first vehicle, and the marker is used to determine key points of the curve model and the target point cloud.

12. A vehicle, characterized in that: include: a first vehicle and a second vehicle, the second vehicle being adapted to be connected to the first vehicle; The rotation angle measurement system according to claim 10 or 11.

Images