Odometer calibration method and related apparatus

Abstract

The application provides an odometer calibration method and related device. The odometer calibration method is applied to a three-wheel mobile device, which comprises one driving wheel and two following wheels. The method comprises: determining odometer pose transformation data of the three-wheel mobile device during calibration driving by using a forward kinematics model of the three-wheel mobile device and coupling zero offsets of the driving wheel and the following wheels; and compensating the zero offsets of the rudder wheel in the three-wheel mobile device by using actual pose transformation data and odometer pose transformation data of the three-wheel mobile device during calibration driving, so as to align the odometer trajectory of the adjusted and compensated three-wheel mobile device with the actual motion trajectory. The application can improve the efficiency of odometer calibration.

Description

Technical Field

[0001] This application relates to the field of odometer calibration technology, and in particular to odometer calibration methods and related apparatus. Background Technology

[0002] Odometer calibration technology is a crucial step in improving the positioning accuracy of various intelligent platforms such as mobile robots and autonomous vehicles, playing an important role in multiple industries. During their long-term research and development process, the inventors of this application discovered that current odometer calibration methods still have certain limitations, which also affect the efficiency of odometer calibration to some extent. Summary of the Invention

[0003] This application provides an odometer calibration method and related apparatus to improve the efficiency of odometer calibration.

[0004] To address the above objectives, this application provides an odometer calibration method applied to a three-wheeled mobile device, the three-wheeled mobile device comprising one drive wheel and two follow wheels; the method includes: Using the positive kinematics model of the three-wheeled mobile device and coupling the zero bias of the drive wheel and the following wheel, the odometer pose transformation data of the three-wheeled mobile device during calibration driving is determined. The positive kinematics model is a quasi-vehicle kinematics model constructed by merging the two following wheels into a virtual wheel. The positive kinematics model of the three-wheeled mobile device is used to calculate the pose of the three-wheeled mobile device in space using the angles of all the steering wheels of the three-wheeled mobile device and the speed of the drive wheel. Using the actual posture transformation data and odometer posture transformation data of the three-wheeled mobile device during calibration driving, the zero bias of the steering wheel in the three-wheeled mobile device is compensated so that the odometer trajectory of the three-wheeled mobile device after adjustment and compensation is aligned with the actual motion trajectory.

[0005] To address the aforementioned problems, this application provides an electronic device, which includes a processor; the processor is used to execute instructions to implement the steps of the above-described method.

[0006] To address the aforementioned problems, this application provides a computer storage medium storing instruction / program data, which, when executed, implements the steps of the method described above.

[0007] The method of this application is as follows: During the calibration of the odometer of a three-wheeled mobile device, the three-wheeled mobile device is driven, and the forward kinematics model of the three-wheeled mobile device is used, coupled with the zero bias of the drive wheel and the following wheel, to determine the odometer pose transformation data of the three-wheeled mobile device during the calibration driving. Then, using the actual pose transformation data of the three-wheeled mobile device during the calibration driving and the odometer pose transformation data, the zero bias of the steering wheel in the three-wheeled mobile device is compensated, so that the odometer trajectory of the three-wheeled mobile device after zero bias compensation is aligned with the actual motion trajectory. The forward kinematics model of the three-wheeled mobile device is a single-vehicle kinematics model constructed by merging the two following wheels into a virtual wheel. This degenerates the three-steering-wheel chassis into a two-wheeled single-vehicle model, simplifying the odometer calibration process. Furthermore, only the drive wheel speed data is needed, which avoids the parameter solution failure problem caused by the lack of rear wheel speed observation. Odometer calibration can be performed even without the drive speed of the following steering wheel, improving the odometer calibration efficiency. Attached Figure Description

[0008] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings that can be used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 This is a flowchart illustrating the implementation method of the odometer calibration method of this application; Figure 2 This is a schematic diagram of the motion model of the three steering wheels in the odometer calibration method of this application; Figure 3 This is a schematic diagram of an implementation method of the three-wheeled mobile device in the odometer calibration method of this application; Figure 4 This is a schematic diagram showing the zero-off angle of the steering wheel of the three-wheeled moving device in the odometer calibration method of this application; Figure 5 This is a schematic diagram of the hand-eye calibration constraint in the odometer calibration method of this application; Figure 6 This is a schematic diagram of the structure of one embodiment of the electronic device of this application; Figure 7 This is a schematic diagram of one embodiment of the computer storage medium of this application. Detailed Implementation

[0010] To enable those skilled in the art to better understand the technical solution of this application, the odometer calibration method and related apparatus provided in this application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0011] The terms "first," "second," and "third" used in this application are for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first," "second," or "third" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0012] In this document, the term "implementation" means that a specific feature, structure, or characteristic described in connection with an implementation may be included in at least one implementation of this application. The appearance of this phrase in various places in the specification does not necessarily refer to the same implementation, nor is it a separate or alternative implementation mutually exclusive with other implementations. It will be explicitly and implicitly understood by those skilled in the art that, without conflict, the implementations described herein may be combined with other implementations.

[0013] like Figure 1 As shown, Figure 1 This is a flowchart illustrating one embodiment of the odometer calibration method of this application. The odometer calibration method of this application is applied to the odometer calibration of a three-wheeled mobile device, wherein, as... Figure 2 As shown, the three-wheeled mobility device includes a drive wheel ( Figure 2 Wheel 0 in the middle and two following wheels ( Figure 2 (Wheel 1 and Wheel 2 in the present application), the subject executing the odometer calibration method of this application is not limited, for example, it can be a robot or a server.

[0014] Figure 2 The green part represents the three steering wheels, the red arrow represents the vehicle body (Odom) coordinate system, and the yellow arrow represents the laser (Laser) coordinate system.

[0015] In one embodiment, the drive wheel (wheel 0) can be a conventional steering wheel, possessing both driving and steering capabilities. The left steering wheel (following wheel 1) and the right steering wheel (following wheel 2) are unconventional steering wheels, possessing only steering capabilities, and are also referred to as follow-up steering wheels.

[0016] In another embodiment, the left steering wheel (following wheel 1) and / or the right steering wheel (following wheel 2) may also have driving capability, but in actual movement, as following wheels, they do not activate their driving capability, but only activate their steering capability, that is, they only act as following wheels and not as driving wheels.

[0017] The odometer calibration method of this application may include the following steps.

[0018] S11: Using the positive kinematic model of the three-wheeled mobile device and coupling the zero bias of the drive wheel and the following wheel, the odometer pose transformation data of the three-wheeled mobile device during calibration driving is determined.

[0019] During the calibration of the odometer of the three-wheeled mobile device, the three-wheeled mobile device can be driven, and the forward kinematic model of the three-wheeled mobile device can be used, coupled with the zero bias of the drive wheel and the following wheel, to determine the odometer pose transformation data of the three-wheeled mobile device during the calibration driving. In order to subsequently use the actual pose transformation data of the three-wheeled mobile device during the calibration driving and the odometer pose transformation data, the zero bias of the steering wheel in the three-wheeled mobile device can be compensated, so that the odometer trajectory of the three-wheeled mobile device after zero bias compensation is aligned with the actual motion trajectory.

[0020] The forward kinematic model of the three-wheeled mobile device is a quasi-vehicle kinematic model constructed by merging the two following wheels into a virtual wheel. The core idea is to merge wheel 1 and wheel 2 into a virtual wheel 3 (i.e., a virtual wheel), thus modeling the chassis as a two-wheeled model (single-vehicle model). The drive wheel (wheel 0) has both driving and steering capabilities, while the following wheel (virtual wheel 3) only has steering capabilities. This degenerates the three-wheeled chassis into a two-wheeled single-vehicle model, simplifying the odometer calibration process. Furthermore, it only requires drive wheel speed data, avoiding the parameter solution failure problem caused by the lack of rear wheel speed observation. Odometer model parameters can be solved even without the drive speed of the following steering wheel. To improve the odometer calibration effect, the angles of wheel 1 and wheel 2 can be kept consistent, and it is assumed that the angle deviation (steering wheel zero offset) of wheel 1 and wheel 2 is basically consistent at the factory.

[0021] The forward kinematic model of a three-wheeled mobile device can be used to calculate the position and attitude of the three-wheeled mobile device in space by using the angles of all the steering wheels and the speed of the drive wheels.

[0022] The forward kinematic model of a three-wheeled mobile device can include a linear velocity calculation model and an angular velocity calculation model for the three-wheeled mobile device.

[0023] The linear velocity calculation model can be expressed as follows:

[0024]

[0025]

[0026] in, Let be the linear velocity of the three-wheeled moving device in the x-direction; Let be the linear velocity of the three-wheeled vehicle in the y-direction; n be the rotational speed of the drive wheel; and r be the radius of the drive wheel. It indicates the angle of the drive wheels, and can also be understood as the angle between the drive wheels and the vehicle body of the three-wheeled mobility device; It represents the sideslip angle, which can be understood as the angle between the direction of the vehicle's linear velocity and the direction of the vehicle's nose.

[0027] As shown above, the linear velocity of the three-wheeled mobile device in the x-direction is calculated based on the rotational speed, radius, and angle of the drive wheels. The linear velocity of the three-wheeled mobile device in the x-direction is positively correlated with the rotational speed and radius of the drive wheels, and can also be positively correlated with the cosine of the angle of the drive wheels. The linear velocity of the three-wheeled mobile device in the x-direction can be equal to the product of the rotational speed, radius, and cosine of the angle of the drive wheels.

[0028] The linear velocity in the y-direction of a three-wheeled mobile device is calculated based on the sideslip angle of the center of mass, as well as the rotational speed, radius, and angle of the drive wheels. The linear velocity in the y-direction of the three-wheeled mobile device is positively correlated with the rotational speed and radius of the drive wheels, and can also be positively correlated with the cosine of the angle of the drive wheels and the sine of the sideslip angle. The linear velocity in the y-direction of the three-wheeled mobile device can be equal to the product of the rotational speed, radius, cosine of the angle of the drive wheels, and the sine of the sideslip angle.

[0029] Among them, the centroid side slip angle The calculation formula is as follows:

[0030] in, Indicates the distance from the drive wheels to the center of the vehicle body; This represents the distance from the virtual wheel to the center of the vehicle body; This represents the angle of the virtual wheel, which can be the angle between the virtual wheel and the vehicle body.

[0031] The sideslip angle of the three-wheeled mobile device can be calculated using the distance from the drive wheel to the center of the vehicle body, the distance from the virtual wheel to the center of the vehicle body, the angle of the virtual wheel, and the angle of the drive wheel.

[0032] The distance from the drive wheel to the center of the vehicle body can be determined based on the known wheelbase parameters. For each three-wheeled mobile device, this distance can be a fixed parameter determined during the chassis design, without the need for additional calculation.

[0033] The distance from the virtual wheel to the center of the vehicle body can be calculated using the geometric positional relationship between the two following wheels, which is a fixed parameter that can be determined during the chassis design.

[0034] The angle of the virtual wheel can be equal to the average of the angles of the two following wheels.

[0035] The angular velocity calculation model can be expressed as follows:

[0036] in, ω is the angular velocity of the three-wheeled vehicle; n is the rotational speed of the drive wheel; r is the radius of the drive wheel; Indicates the angle of the drive wheel; Indicates the angle of the virtual wheel; Indicates the distance from the drive wheels to the center of the vehicle body; This represents the distance from the virtual wheel to the center of the vehicle body.

[0037] Thus, the angular velocity of the three-wheeled mobile device can be calculated using the rotational speed of the drive wheel, the radius of the drive wheel, the angle of the drive wheel, the angle of the virtual wheel, the distance from the drive wheel to the center of the vehicle body, and the distance from the virtual wheel to the center of the vehicle body.

[0038] Among them, such as Figure 3 As shown, Figure 3 In the diagram, point A represents the center position of the drive wheel, point B represents the center position of the virtual wheel, point C represents the center of the vehicle body, and point O represents the instantaneous center of rotation. R represents the distance from the center of the vehicle body to the instantaneous center of rotation. The derivation process of the forward kinematic model of the three-wheeled mobile device is shown below: Applying the law of sines to triangle ACO, we have:

[0039] Applying the law of sines to triangle BCO, we have:

[0040] Expanding and combining the above formulas, we get:

[0041] At this moment, the angular velocity of the vehicle body for:

[0042]

[0043] Since the chassis is a rigid body, the angular velocities of points C and A are the same at any given time, therefore:

[0044] The linear velocity can be obtained by applying the sine theorem to the triangle ACO. Size :

[0045]

[0046]

[0047] By rearranging the above formulas, we can obtain... :

[0048]

[0049]

[0050]

[0051] in By combining formulas (1) and (2), we obtain: .

[0052] In the ideal model, the steering wheel's installation angle is consistent with the device's forward direction. However, in practical applications, the steering wheel's installation angle may deviate. The angle between the steering wheel's installation angle and the vehicle's forward direction can be called the steering wheel angle zero deviation, with counterclockwise being positive. Figure 4 As shown, the zero bias of the steering wheels of wheels 0 and 2 is negative, while the zero bias of wheels 1 and 3 is positive.

[0053] It can be used or To indicate zero bias of the drive wheel, use This indicates the zero bias of the virtual wheel.

[0054] Assuming the zero-bias of the steering wheels of wheels 1 and 2 are not significantly different, the solution shifts to calculating the zero-bias of the virtual steering wheel 3 (i.e., the average zero-bias of wheels 1 and 2), and using this as the predicted zero-bias value for the steering wheels of wheels 1 and 2:

[0055]

[0056] Furthermore, the method for calculating the angle of wheel #3 is to take the average of the angles of wheels #1 and #2:

[0057] It is understandable that this application can be applied to a three-wheeled mobile device with two following wheels. That is, the three-wheeled mobile device lacks speed observation of wheel 1 and wheel 2, so it is impossible to accurately calculate the zero bias of the steering wheel of wheel 1 and wheel 2. Instead, the average zero bias of the two wheels is calculated, that is, the zero bias of wheel 3.

[0058] Based on this, the step of "using the forward kinematics model of the three-wheeled mobile device and coupling the zero offsets of the drive wheel and the following wheel to determine the odometer pose transformation data of the three-wheeled mobile device during calibration driving" can include: substituting the zero offsets of the drive wheel and the following wheel into the forward kinematics model of the three-wheeled mobile device. Specifically, the sum of the odometer observation angle of the drive wheel and its zero offset is used as the angle of the drive wheel and substituted into the forward kinematics model. Similarly, the sum of the zero offset of the virtual wheel and its odometer observation angle is used as the angle of the virtual wheel and substituted into the forward kinematics model to obtain the angles of the drive wheel and the following wheel. The relationship between the zero bias and the linear velocity of the three-wheeled mobile device is obtained, and / or the relationship between the zero bias of the drive wheel and the following wheel and the angular velocity of the three-wheeled mobile device is obtained; by substituting the zero bias into the forward kinematic model, the odometer pose transformation data of the three-wheeled mobile device during calibration driving is calculated. The odometer pose transformation data of the three-wheeled mobile device can be obtained by integrating the forward kinematic model. The odometer pose transformation data can be solved by closed integration. For example, the odometer pose transformation of each frame can be calculated frame by frame to obtain the odometer pose transformation data.

[0059] Assuming that, within the calibrated travel time Inner vehicle linear velocity and angular velocity .

[0060] Changes in vehicle body angle over time .

[0061] Changes in vehicle position over time for:

[0062] Odometer pose change data of the vehicle body within a time period for:

[0063] in .

[0064] The odometer angle of the virtual wheel mentioned above can be equal to the average of the odometer angles of the two following wheels.

[0065] Substituting zero bias (where, To achieve zero bias following the wheel, The forward kinematics model after zero bias of the drive wheel can be shown below:

[0066]

[0067]

[0068] in, ; in, It is the odometer viewing angle of the drive wheel. It is zero bias of the drive wheel. and These are the odometer observation angles of the two following wheels. It follows the zero bias of the wheel.

[0069] S12: Using the actual posture transformation data and odometer posture transformation data of the three-wheeled mobile device during calibration driving, the zero bias of the steering wheel of the three-wheeled mobile device is compensated so that the odometer trajectory of the three-wheeled mobile device after zero bias compensation is aligned with the actual motion trajectory.

[0070] By using the forward kinematics model of the three-wheeled mobile device and coupling the zero bias of the drive wheel and the follower wheel, the odometer pose transformation data of the three-wheeled mobile device during calibration driving is determined. Then, the zero bias of the steering wheel in the three-wheeled mobile device can be compensated using the actual pose transformation data and odometer pose transformation data of the three-wheeled mobile device during calibration driving, so that the odometer trajectory of the three-wheeled mobile device after zero bias compensation is aligned with the actual motion trajectory.

[0071] In one implementation, in step S12, the zero-bias of the steering wheel of the three-wheeled mobile device can be compensated using a calibration method based on sensor data fusion. This implementation can involve using a high-precision external sensor (such as a lidar) to calculate a more accurate pose transformation of the three-wheeled mobile device using a scanning matching algorithm, which serves as the actual pose transformation data. Then, algorithms such as least squares are used to solve for the optimal parameters of the motion model (i.e., the zero-bias data of the steering wheel) in one step. Finally, the zero-bias data of the steering wheel is used to compensate for the zero-bias of the steering wheel in the three-wheeled mobile device.

[0072] The aforementioned sensor data can be lidar data, inertial measurement unit data, navigation satellite system data, or camera data.

[0073] In one scenario, when the zero offset of the drive wheel and the zero offset of the following wheel are inconsistent, the car will travel in an arc when a straight line is launched (at this time, the actual angle of the drive wheel and the actual angle of the virtual wheel are inconsistent). In this case, the zero offset of the following wheel (i.e., the virtual wheel) can be kept constant, and the relative zero offset of the drive wheel relative to the following wheel (i.e., the virtual wheel) can be solved by algorithms such as the least squares method. Then, the relative zero offset can be used to compensate for the zero offset of the drive wheel.

[0074] In practice, the odometer pose transformation data of the three-wheeled mobile device during calibration should be equal to the actual pose transformation data during calibration. This can be used to establish a relationship between the actual pose transformation data and the zero bias of the drive wheel and the following wheel (this can be called the relative zero bias relationship). The relative zero bias relationship can be solved using algorithms such as the least squares method to obtain the relative zero bias of the drive wheel relative to the following wheel. In one embodiment, when solving for the relative zero bias, the zero bias of the virtual wheel can be set to 0, and then the relative zero bias relationship can be solved using algorithms such as the nonlinear least squares method to obtain the relative zero bias of the drive wheel relative to the following wheel. .

[0075] The relative zero-bias relation can be expressed as follows: .

[0076] in, It is the actual angle change during calibrated driving.

[0077] The above relative zero bias relationship is determined by the fact that "the change in the odometer angle of the three-wheeled mobile device during the calibrated driving is equal to the actual change in angle." This is the rotational constraint of hand-eye calibration.

[0078] like Figure 5 As shown, the derivation process of the rotation constraint formula for hand-eye calibration is as follows: As shown in the diagram above, let O be the vehicle coordinate system, L be the laser coordinate system, and the subscripts indicate the time corresponding to the coordinate system. Let X be the transformation matrix from the vehicle coordinate system to the laser coordinate system, A be the transformation matrix from the vehicle coordinate system at time t to the vehicle coordinate system at time t+1, and B be the transformation matrix from the laser coordinate system at time t to the laser coordinate system at time t+1. Then, the transformation matrix from the vehicle coordinate system at time t to the laser coordinate system at time t+1 can be represented in two ways, and they are equal:

[0079] Expanding, we get the following form:

[0080] Where R is the rotation matrix component corresponding to the transformation matrix, t is the translation vector component corresponding to the transformation matrix, and the subscript corresponds to the transformation matrix.

[0081] Expanding further, we get the following form:

[0082]

[0083] These two formulas are respectively called the rotational and translational constraints of hand-eye calibration.

[0084] Since the two-dimensional rotation matrix is ​​commutative with respect to matrix multiplication, the rotation constraint degenerates into...

[0085] In other words, the change in pose in the laser coordinate system from time t to time t+1 is equal to the change in pose in the vehicle coordinate system.

[0086] in The change in the odometer angle from time t to time t+1. This represents the actual change in angle from time t to time t+1.

[0087] In practice, steps S11 and S12 can be repeated to iteratively adjust the relative zero bias of the drive wheel relative to the following wheel in the three-wheeled moving device until the final calculated relative zero bias converges.

[0088] At this point, it is assumed that the relative zero bias of the drive wheel with respect to the two rear steering wheels is approximately 0.

[0089] During implementation, the currently calculated relative zero bias can be set through the zero bias setting interface to compensate the angle of the drive wheel in real time, that is, to compensate for the zero bias of the drive wheel using the relative zero bias.

[0090] The core compensation operation can occur during the stage where the controller generates and sends the angle command to the steering wheel driver. Assume the upper-level navigation system commands the robot's steering wheel to turn 30 degrees (target_angle = 30°), while the calibrated zero offset_angle is -2 degrees (i.e., the actual zero position of the steering wheel is 2 degrees to the left of the theoretical zero position). Then, the controller will not directly send a 30-degree command, but rather a command of 30 - (-2) = 32 degrees. Thus, under the influence of its own zero offset of -2 degrees, the actual steering angle of the steering wheel will be close to 32 + (-2) = 30 degrees, thereby aligning with the target.

[0091] Furthermore, after adjusting the relative zero bias, a straight line with a lateral displacement angle of 0 is sent to the three-wheeled moving device. At this point, according to the definition of zero bias of the steering wheel, it can be known that the actual angles of the three steering wheels are consistent, and the vehicle as a whole will move in a straight line, although the lateral displacement angle may not be 0. At this time, the vehicle motion trajectory calculated by the odometer and the vehicle motion trajectory observed by the laser differ only by a rotation transformation. The absolute zero bias of the three steering wheels in the three-wheeled moving device can be solved by the difference between the odometer angle and the actual angle. Then, the absolute zero bias of the steering wheels is used to compensate for the zero bias of the three steering wheels in the three-wheeled moving device. The actual pose data calculated by high-precision external sensors (such as lidar) is used as the reference data, and the overall zero bias of the three steering wheels is adjusted to align the odometer trajectory with the actual trajectory.

[0092] In a specific application scenario, step 1 involves sending a straight line to the three-wheeled mobile device. Using the forward kinematics model of the three-wheeled mobile device and coupling the zero bias of the drive wheel and the following wheel, the odometer pose transformation data of the three-wheeled mobile device in the latest travel segment is determined. Step 2 involves using the actual pose transformation data of the three-wheeled mobile device in the latest travel segment and the odometer pose transformation data, fixing the zero bias of the following wheel at 0, and solving for the relative zero bias of the drive wheel relative to the following wheel using algorithms such as least squares. This relative zero bias is then used to compensate for the zero bias of the drive wheel. Steps 1 and 2 are repeated sequentially to iteratively adjust the relative zero bias of the drive wheel relative to the following wheel in the three-wheeled mobile device. Step 4: Next, a straight line is sent to the three-wheeled mobile device. Using the forward kinematics model of the three-wheeled mobile device and coupling the zero bias of the drive wheel and the following wheel, the odometer pose transformation data of the three-wheeled mobile device in the latest segment of travel is determined. Step 5: Using the actual pose transformation data and odometer pose transformation data of the three-wheeled mobile device in the latest segment of travel, the absolute zero bias of the three steering wheels in the three-wheeled mobile device is solved, and the zero bias of the three steering wheels is compensated using the absolute zero bias. Steps 4 and 5 are repeated sequentially to iteratively adjust the absolute zero bias of the three steering wheels in the three-wheeled mobile device until the final absolute zero bias converges.

[0093] At this point, it is assumed that the absolute zero bias of the three steering wheels as a whole is approximately 0, and the zero bias of each of the three steering wheels has been adjusted.

[0094] Optionally, the absolute zero bias of the three steering wheels can be calculated using the relative zero bias relationship. When calculating the absolute zero bias, considering that the zero bias of the drive wheel is equal to the zero bias of the follow wheel, the zero bias of the follow wheel or drive wheel can be calculated, and the calculated zero bias of the follow wheel or drive wheel can be used as the absolute zero bias.

[0095] Alternatively, the difference between the odometer steering angle and the actual steering angle can be calculated to obtain absolute zero bias. The odometer steering angle is calculated from the positional change offset of the three-wheeled vehicle derived from the odometer, while the actual steering angle is calculated from the actual positional change offset of the three-wheeled vehicle. In this example, the formula for calculating absolute zero bias can be:

[0096] in It is the offset of the pose transformation from the laser coordinate system to the vehicle coordinate system. It is the pose change offset calculated by the odometer.

[0097] Corresponding to relative zero bias, during implementation, the currently calculated absolute zero bias can also be set through the zero bias setting interface to compensate the angle of the drive wheel in real time, that is, to compensate for the zero bias of the drive wheel using the absolute zero bias.

[0098] This application can use zero-bias setting interface technology to compensate for the zero bias of the steering wheel through microcontroller compensation, thus eliminating the need to manually compensate for the zero bias of the steering wheel during each odometer calculation.

[0099] In another implementation, in step S12, the zero-offset of the steering wheel of the three-wheeled mobile device can be adjusted using a geometric calculation method based on the reference trajectory. This implementation allows the three-wheeled mobile device to travel a distance along a straight line, and its lateral offset relative to an ideal straight line is measured; then, the zero-offset angle of the front wheel is calculated back through geometric relationships and compensated for.

[0100] In another implementation, in step S12, the zero bias of the steering wheel of the three-wheeled mobile device can be adjusted using a trajectory alignment-based accuracy assessment and parameter back-calculation method. This implementation allows for the recording of an estimated odometer trajectory and a true trajectory (e.g., from GPS), followed by the calculation of the transformation relationship between the two coordinate systems using an algorithm (e.g., SVD) to assess the error. This error assessment result can provide direction for parameter adjustment.

[0101] Please see Figure 6 , Figure 6 This is a schematic diagram of one embodiment of the electronic device of this application. The electronic device 10 includes a processor 12, which executes instructions to implement the above-described odometer calibration method. For details of the implementation process, please refer to the description of the above embodiment; it will not be repeated here.

[0102] Processor 12 can also be referred to as a CPU (Central Processing Unit). Processor 12 may be an integrated circuit chip with signal processing capabilities. Processor 12 may also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. A general-purpose processor may be a microprocessor, or processor 12 may be any conventional processor.

[0103] The electronic device 10 may further include a memory 11 for storing instructions and data required for the processor 12 to run.

[0104] The processor 12 is used to execute instructions to implement the method provided by any embodiment and any non-conflicting combination of the odometer calibration method of this application described above.

[0105] Please see Figure 7 , Figure 7 This is a schematic diagram of the structure of a computer-readable storage medium in an embodiment of this application. The computer-readable storage medium 20 in this embodiment stores instruction / program data 21. When executed, this instruction / program data 21 implements the method provided by any embodiment of the odometer calibration method of this application and any non-conflicting combination thereof. The instruction / program data 21 can be formed into a program file and stored in the storage medium 20 in the form of a software product, so that a computer device (which may be a personal computer, server, or network device, etc.) or processor can execute all or part of the steps of the methods in various embodiments of this application. The aforementioned storage medium 20 includes various media capable of storing program code, such as a USB flash drive, portable hard drive, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk, or terminal devices such as computers, servers, mobile phones, and tablets.

[0106] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, or indirect coupling or communication connection between apparatuses or units, and may be electrical, mechanical, or other forms.

[0107] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0108] The above are merely embodiments of this application and do not limit the scope of this patent application. Any equivalent structural or procedural changes made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of this application.

Claims

1. A method for calibrating an odometer, characterized in that, The odometer calibration method is applied to a three-wheeled mobile device, which includes one drive wheel and two follow wheels; the method includes: Using the positive kinematics model of the three-wheeled mobile device and coupling the zero bias of the drive wheel and the following wheel, the odometer pose transformation data of the three-wheeled mobile device during calibration driving is determined. The positive kinematics model is a quasi-vehicle kinematics model constructed by merging the two following wheels into a virtual wheel. The positive kinematics model of the three-wheeled mobile device is used to calculate the pose of the three-wheeled mobile device in space using the angles of all the steering wheels of the three-wheeled mobile device and the speed of the drive wheel. Using the actual posture transformation data and odometer posture transformation data of the three-wheeled mobile device during calibration driving, the zero bias of the steering wheel in the three-wheeled mobile device is compensated to align the odometer trajectory of the three-wheeled mobile device with the actual motion trajectory after adjustment and compensation. The zero bias of the steering wheel in the three-wheeled mobile device includes relative zero bias and absolute zero bias. A relative zero bias relationship is established, and the relative zero bias relationship is solved to obtain the relative zero bias of the drive wheel relative to the virtual wheel, and the absolute zero bias of the three steering wheels in the three-wheeled mobile device is also solved.

2. The odometer calibration method according to claim 1, characterized in that, The positive kinematics model includes a linear velocity calculation model and an angular velocity calculation model for the three-wheeled mobile device; the step of using the positive kinematics model of the three-wheeled mobile device and coupling it with the zero offset of the drive wheel and the following wheel to determine the odometer pose transformation data of the three-wheeled mobile device during calibration driving includes: The sum of the odometer observation angle and the zero bias of the drive wheel is substituted into the forward kinematics model as the angle of the drive wheel, and the sum of the zero bias of the virtual wheel and the odometer observation angle is substituted into the forward kinematics model as the angle of the virtual wheel, to obtain the relationship between the zero bias of the drive wheel and the following wheel and the linear velocity of the three-wheeled moving device, and the relationship between the zero bias of the drive wheel and the following wheel and the angular velocity of the three-wheeled moving device. The odometer pose transformation data of the three-wheeled mobile device is obtained by integrating the angular velocity and linear velocity of the three-wheeled mobile device. The odometer observation angle of the virtual wheel is equal to the average of the odometer observation angles of the two following wheels.

3. The odometer calibration method according to claim 1, characterized in that, The positive kinematics model includes a linear velocity calculation model and an angular velocity calculation model for the three-wheeled moving device; The linear velocity of the three-wheeled mobile device in the x-direction is calculated based on the rotational speed, radius, and angle of the drive wheel. The linear velocity of the three-wheeled mobile device in the y-direction is calculated based on the centroid sideslip angle, the rotational speed of the drive wheel, the radius of the drive wheel, and the angle of the drive wheel. The sideslip angle of the three-wheeled mobile device is calculated using the distance from the drive wheel to the center of the vehicle body, the distance from the virtual wheel to the center of the vehicle body, the angle of the virtual wheel, and the angle of the drive wheel. The angular velocity of the three-wheeled mobile device is calculated using the rotational speed of the drive wheel, the radius of the drive wheel, the angle of the drive wheel, the angle of the virtual wheel, the distance from the drive wheel to the center of the vehicle body, and the distance from the virtual wheel to the center of the vehicle body.

4. The odometer calibration method according to claim 3, characterized in that, The formula for the forward kinematics model is as follows: in, Let x be the linear velocity of the three-wheeled mobile device in the x-direction. Let be the linear velocity of the three-wheeled mobile device in the y-direction; n be the rotational speed of the drive wheel; and r be the radius of the drive wheel. Indicates the angle of the drive wheel; Indicates the centroid side slip angle The angular velocity of the three-wheeled moving device; Indicates the angle of the virtual wheel; Indicates the distance from the drive wheels to the center of the vehicle body; This represents the distance from the virtual wheel to the center of the vehicle body.

5. The odometer calibration method according to claim 1, characterized in that, The step of using the actual posture transformation data and odometer posture transformation data of the three-wheeled mobile device during calibration driving to compensate for the zero deviation of the steering wheel in the three-wheeled mobile device, so as to align the odometer trajectory of the three-wheeled mobile device after adjustment and compensation with the actual motion trajectory, includes: The actual pose transformation data of the three-wheeled mobile device is calculated using a high-precision external sensor. Using the actual posture transformation data and odometer posture transformation data of the three-wheeled mobile device during calibration driving, the zero-bias data of the steering wheel of the three-wheeled mobile device is solved. The zero-bias data of the steering wheel is used to compensate for the zero-bias of the steering wheel in the three-wheeled moving device.

6. The odometer calibration method according to claim 5, characterized in that, The step of using the actual posture transformation data and odometer posture transformation data of the three-wheeled mobile device during calibration driving to solve for the zero-bias data of the steering wheel of the three-wheeled mobile device includes: Based on the principle that the odometer pose transformation data of the three-wheeled mobile device during calibration should be equal to the actual pose transformation data during calibration, a relative zero bias relationship is established. The relative zero bias relationship is the relationship between the actual pose transformation data, the zero bias of the drive wheel, and the zero bias of the virtual wheel. Set the zero bias of the virtual wheel to 0, solve the relative zero bias relationship to obtain the relative zero bias of the drive wheel relative to the virtual wheel; The method of compensating for the zero-bias of the steering wheel in the three-wheeled moving device using the zero-bias data of the steering wheel includes: The relative zero bias is used to compensate for the zero bias of the drive wheel.

7. The odometer calibration method according to claim 6, characterized in that, The process of compensating for the zero bias of the drive wheel using the relative zero bias includes: A straight line with a lateral displacement angle of 0 is sent to the three-wheeled mobile device. Using the positive kinematic model of the three-wheeled mobile device and coupling the zero bias of the drive wheel and the following wheel, the odometer pose transformation data of the three-wheeled mobile device in the latest segment of travel is determined. Using the actual posture transformation data and odometer posture transformation data of the three-wheeled mobile device during the latest segment of travel, the absolute zero bias of the three steering wheels in the three-wheeled mobile device is solved, and the zero bias of the three steering wheels is compensated using the absolute zero bias.

8. The odometer calibration method according to claim 7, characterized in that, The odometer pose transformation data includes the odometer pose transformation offset, and the actual pose transformation data includes the actual pose transformation offset. The step of using the actual pose transformation data and odometer pose transformation data of the three-wheeled mobile device in the latest segment of travel to determine the absolute zero offset of the three steering wheels in the three-wheeled mobile device includes: The difference between the odometer steering angle and the actual steering angle is calculated to obtain the absolute zero bias; The odometer steering angle is calculated from the odometer pose transformation offset, and the actual steering angle is calculated from the actual pose transformation offset of the three-wheeled moving device.

9. The odometer calibration method according to claim 7, characterized in that, The step of using the actual posture transformation data and odometer posture transformation data of the three-wheeled mobile device during calibration driving to compensate for the zero deviation of the steering wheel in the three-wheeled mobile device, so as to align the odometer trajectory of the three-wheeled mobile device after adjustment and compensation with the actual motion trajectory, includes: The steps of setting the zero bias of the virtual wheel to 0, solving the relative zero bias relationship to obtain the relative zero bias of the drive wheel relative to the virtual wheel, and compensating for the zero bias of the drive wheel using the relative zero bias are repeated sequentially to continuously iterate and adjust the relative zero bias of the drive wheel relative to the virtual wheel in the three-wheeled moving device until the final solved relative zero bias converges. The steps of issuing a straight line with a lateral displacement angle of 0 to the three-wheeled mobile device, using the positive kinematic model of the three-wheeled mobile device and coupling the zero bias of the drive wheel and the following wheel to determine the odometer pose transformation data of the three-wheeled mobile device in the latest segment of travel, and using the actual pose transformation data and odometer pose transformation data of the three-wheeled mobile device in the latest segment of travel to solve for the absolute zero bias of the three steering wheels in the three-wheeled mobile device, and using the absolute zero bias to compensate for the zero bias of the three steering wheels, are repeated in sequence to continuously iterate and adjust the absolute zero bias of the three steering wheels in the three-wheeled mobile device until the final solved absolute zero bias converges.

10. An electronic device, characterized in that, The electronic device includes a processor; the processor is configured to execute instructions to implement the steps of the method as described in any one of claims 1-9.

11. A computer-readable storage medium having a program stored thereon, characterized in that, When the program is executed by a processor, it implements the steps of the method according to any one of claims 1-9.

Images