Mobile chassis, navigation device and method, online registration method and storage medium

Abstract

A mobile chassis, a navigation device and method, an online registration method, and a storage medium. A navigation device is installed on the mobile chassis. The navigation device includes a camera; an inertial measurement unit disposed within the camera; a lidar; a housing, the camera is disposed in a slot on the front side of the top of the housing, the lidar is installed within the housing and is located below the camera; an embedded processor installed within the housing and located below the lidar, the camera, the inertial measurement unit, and the lidar are respectively connected to the embedded processor; and a rear cover connected to the housing, an interaction interface and a GNSS antenna interface are provided on the rear cover, and the interaction interface and the GNSS antenna interface are respectively connected to the embedded processor. The present invention also discloses an online registration method for the navigation device, a navigation method for the mobile chassis, and a computer-readable storage medium storing a computer program for implementing the navigation method.

Description

Technical Field

[0001] The present invention relates to navigation and positioning technologies, and particularly to a mobile chassis that realizes autonomous navigation by means of multi-sensor fusion, its navigation device, autonomous navigation method, online registration method, and storage medium. Background Art

[0002] The intelligentization of agricultural machinery is a new trend in agricultural development, a product of economic development, and meets the new demands of agricultural development. The use of intelligent agricultural machinery can improve agricultural production efficiency, reduce the workload of farmers, and improve the quality of agricultural products. In recent years, agricultural mobile chassis have developed rapidly and shown great superiority in improving agricultural productivity, changing agricultural production models, solving labor shortages, and achieving large-scale, diversified, and precise agriculture, and have gradually become one of the important directions in the field of agricultural engineering.

[0003] There are various types and functions of agricultural mobile chassis, among which navigation and planning are the core of agricultural mobile chassis technology. To achieve the autonomous navigation function for a general agricultural mobile chassis, a positioning sensor and a decision-making and planning control system need to be installed. Since achieving the autonomous navigation function requires various types of sensors to be equipped and strict positional relationships between the sensors to be ensured, and most agricultural mobile chassis do not consider the autonomous navigation function, the types of sensors carried by the chassis are few, the accuracy is low, and the degree of intelligence is low. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a mobile chassis, a navigation device and method, an online registration method, and a storage medium that realize autonomous navigation by means of multi-sensor fusion in view of the above-mentioned defects of the prior art.

[0005] To achieve the above object, the present invention provides a navigation device installed on a mobile chassis, which includes:

[0006] A camera for collecting front image information for image-based positioning;

[0007] An inertial measurement unit disposed in the camera for collecting acceleration and angle information for multi-sensor fusion-based positioning;

[0008] A lidar for collecting surrounding point cloud information for point cloud-based positioning;

[0009] A housing for determining the positional relationship between the camera and the lidar and realizing spatial synchronization of the data of the camera, the inertial measurement unit, and the lidar; the camera is disposed in a machine slot on the front side of the top of the housing; the lidar is installed in the housing and is located below the camera;

[0010] An embedded processor, installed inside the housing and located below the lidar, is used to process information from the camera, inertial measurement unit, lidar, and the mobile chassis, perform high-precision positioning and path planning, and issue control instructions; the camera, inertial measurement unit, and lidar are respectively connected to the embedded processor; and

[0011] A rear cover, connected to the housing and determining the relative coordinate relationship between the camera and the lidar together with the housing, is provided with an interaction interface and a GNSS antenna interface for connecting to the agricultural mobile chassis, and the interaction interface and the GNSS antenna interface are respectively connected to the embedded processor.

[0012] For the above navigation device of the mobile chassis, GNSS unified timekeeping is used for simultaneous triggering to ensure the alignment of timestamps during the fusion of the camera, inertial measurement unit, and lidar, and to achieve the time synchronization of the camera, inertial measurement unit, and lidar data.

[0013] For the above navigation device of the mobile chassis, coordinate transformation is performed on the camera, inertial measurement unit, lidar, and GNSS, and the conversion relationships for normalizing the camera coordinate system a M , lidar coordinate system a L , IMU coordinate system a I , and GNSS coordinate system a G to the device coordinate system a E are as follows:

[0014] a E = R EM a M + t M ;

[0015] a E = R EL a L + t L ;

[0016] a E = R EG a G + t G ;

[0017] a E = R EI a I + t I ;

[0018] wherein, R EM , R EL , R EG , R EIThey are the rotation matrices of the camera, lidar, GNSS, and IMU coordinate systems respectively; t M , t L , t G , t I They are the translation matrices of the camera, lidar, GNSS, and IMU coordinate systems respectively.

[0019] To better achieve the above object, the present invention also provides an online registration method for a navigation device, wherein, for aligning the coordinate systems between the navigation device and the mobile chassis, obtaining the rotation matrix and translation matrix between the navigation device and the mobile chassis, includes the following steps:

[0020] S100. Obtain the pose P of the wheel speed odometer installed on the mobile chassis O ;

[0021] S200. Obtain the pose P of the IMU odometer set on the navigation device I ;

[0022] S300. Synchronize the wheel speed odometer and the IMU odometer, rotate and translate the IMU odometer to align it with the wheel speed odometer, and the conversion relationship between the wheel speed odometer and the IMU odometer is:

[0023]

[0024] Wherein, is the pose conversion matrix for converting the IMU odometer to the wheel speed odometer; R is the rotation matrix, and t is the translation matrix; and

[0025] S400. Solve the pose conversion matrix between the wheel speed odometer and the IMU odometer, and obtain the least squares solution: x * = argmin x ||Ax - y|| 2 , the A matrix is composed of multiple sets of continuous IMU odometer data P I , the y matrix is composed of multiple sets of wheel speed odometer data P corresponding to the same moment as the IMU odometer data O , and x * is the one to be solved

[0026] To better achieve the above object, the present invention also provides a mobile chassis, which includes the above navigation device and uses the above online registration method to align the coordinate systems between the navigation device and the mobile chassis.

[0027] To better achieve the above object, the present invention also provides an autonomous navigation method for a mobile chassis, which includes the following steps:

[0028] S100. Install the above-mentioned navigation device on a mobile chassis;

[0029] S200. Align the coordinate systems between the navigation device and the mobile chassis, and solve the pose transformation matrix between the navigation device and the mobile chassis;

[0030] S300. Perform global and local positioning on the sensors of the navigation device and the mobile chassis, construct a surrounding environment map, and at the same time map the surrounding environment map in the global coordinate system;

[0031] S400. The navigation device performs global and local path planning according to the current positioning and the surrounding environment map based on the task type;

[0032] S500. The navigation device sends control signals including vehicle speed and steering angle to the mobile chassis; and

[0033] S600. The mobile chassis formulates corresponding control strategies according to the control signals and the corresponding odometer model.

[0034] For the above-mentioned autonomous navigation method of the mobile chassis, further comprising the following steps:

[0035] S700. During the movement of the mobile chassis, speed feedback is performed in real time, and the navigation device adjusts and optimizes the control signal according to the speed feedback signal.

[0036] For the above-mentioned autonomous navigation method of the mobile chassis, further comprising the following steps:

[0037] S800. During the movement of the mobile chassis, the navigation device performs position feedback in real time, and the mobile chassis plans and adjusts the control strategy in real time according to the position transformation information.

[0038] For the above-mentioned autonomous navigation method of the mobile chassis, in step S200, the above-mentioned online registration method is used to align the coordinate systems between the navigation device and the mobile chassis.

[0039] To better achieve the above object, the present invention also provides a computer-readable storage medium, on which a computer program is stored, wherein the computer program, when executed by a processor, implements the above-mentioned autonomous navigation method of the mobile chassis.

[0040] The technical effect of the present invention is as follows:

[0041] The navigation device of the present invention can be applied to various mobile chassis and achieve autonomous navigation. It integrates various sensors and decision-making systems required for autonomous navigation, and determines the positional relationship of each sensor through the structure, avoiding unpredictable errors generated during the process of separately installing sensors. It jointly optimizes the IMU (Inertial Measurement Unit) and the angular velocity sensor carried by the chassis to obtain the pose transformation relationship between the two. It solves the adaptation problem caused by different mobile chassis, can adapt to most current mobile chassis, especially agricultural mobile chassis, and does not require other accessories and specific scenario assistance during the registration process, reducing costs and realizing the autonomous navigation of mobile chassis, especially agricultural mobile chassis.

[0042] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments, but it is not intended to limit the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 Schematic diagram of the device structure according to an embodiment of the present invention;

[0044] Figure 2 Schematic diagram of the online registration principle according to an embodiment of the present invention;

[0045] Figure 3 Schematic diagram of the two-wheel differential odometer model according to an embodiment of the present invention;

[0046] Figure 4 Schematic diagram of the odometer pose relationship according to an embodiment of the present invention;

[0047] Figure 5 Schematic diagram of the autonomous navigation principle of the mobile chassis according to an embodiment of the present invention.

[0048] Among them, the reference numerals

[0049] 1 Housing

[0050] 2 Rear cover

[0051] 3 Camera

[0052] 4 Camera cover

[0053] 5 Embedded processor

[0054] 6 Lidar

[0055] 7 GNSS antenna interface

[0056] 8 Interaction interface

[0057] 9 Machine slot

[0058] 10 Wheel speed odometer path

[0059] 11 IMU odometry path

[0060] 12 Pose transformation relationship between two odometers Specific implementation manner

[0061] The structural principle and working principle of the present invention will be specifically described below with reference to the accompanying drawings:

[0062] See Figure 1 , Figure 1 , which is a schematic diagram of the device structure of an embodiment of the present invention. The navigation device of the present invention integrates various sensors and processors necessary for autonomous navigation, and calibrates the external parameters of each sensor, and can be quickly installed on various mobile chassis to achieve autonomous navigation. The navigation device includes: a camera 3, the front camera 3 is used to collect front image information and transmit it to the embedded processor 5 for image-based positioning. A binocular depth camera 3 can be used, and it is also equipped with an IMU (inertial measurement unit); an inertial measurement unit (IMU), which is arranged in the camera 3, and the gyroscope is used to collect acceleration and angle information and transmit it to the embedded processor 5 for multi-sensor fusion-based positioning; a lidar 6, which is used to collect point cloud information around the mobile chassis and transmit it to the embedded processor 5 for point cloud-based positioning; a GNSS receiver, which is installed in the housing 1; the housing 1 is used to determine the positional relationship between the camera 3 and the lidar 6 to achieve spatial synchronization of the data of the camera 3, the inertial measurement unit and the lidar 6; the camera 3 is arranged in the machine slot 9 at the front side of the top of the housing 1, and the camera cover 4 is buckled on the machine slot 9 to encapsulate the camera 3; the lidar 6 is installed in the housing 1 and is located below the camera 3; an embedded processor 5, which is installed in the housing 1 and is located below the lidar 6, is used to process information from the camera 3, the inertial measurement unit, the lidar 6 and the mobile chassis, perform high-precision positioning and path planning, and issue control instructions; the camera 3, the inertial measurement unit and the lidar 6 are respectively connected to the embedded processor 5; and a rear cover 2, which is connected to the housing 1 and together with the housing 1 determines the relative coordinate relationship between the camera 3 and the lidar 6. An interaction interface 8 for connecting to the agricultural mobile chassis and a GNSS antenna interface 7 are arranged on the rear cover 2, and the interaction interface 8 and the GNSS antenna interface 7 are respectively connected to the embedded processor 5. The interaction interface 8 includes a power supply interface, a sensor input interface, a control signal output interface, a device debugging interface, etc.

[0063] The basis of sensor fusion for navigation and positioning is the need to clarify the positional relationship between each sensor. The navigation device can improve the accuracy of the spatial position of multiple sensors. The installation method of the prior art is to manually arrange the mobile chassis body, and it is impossible to obtain accurate position installation parameters. The use of this navigation device can avoid installation errors. The position arrangement of each sensor does not affect the use of the algorithm. A more accurate sensor installation relationship can be obtained through the structure of the shell 1. The shell 1 and the back cover 2 fix the relative position of each sensor, determine the relative coordinate relationship between each sensor, reduce the complexity of the coordinate conversion between the sensor and the mobile chassis, and calibrate the spatial position of the sensor on this basis.

[0064] See also Figure 2 , Figure 2 The figure is a schematic diagram of the online registration principle of an embodiment of the present invention. The realization of multi-sensor fusion positioning needs to involve the fusion of data from each sensor, and the most basic problem is the time and space synchronization of data from each sensor. Among them, the spatial synchronization adopts a fixed device shell 1 to determine the positional relationship between each sensor in advance and shorten the development cycle. Time synchronization uses GNSS unified timing to trigger at the same time to ensure that the timestamps of the camera 3, the inertial measurement unit and the laser radar 6 are aligned when they are fused, so as to achieve the time synchronization of the data of the camera 3, the inertial measurement unit and the laser radar 6. The front fusion positioning algorithm is adopted, and the raw data of each sensor is processed by the embedded processor 5, and feature extraction is performed. The advantages and disadvantages of each sensor are complemented at the feature and data level to solve the degradation problem of a single sensor in a challenging outdoor scene. The position information solved by each sensor is fused through the post-fusion decision layer fusion algorithm, and the confidence interval of each sensor is adjusted according to the current environmental characteristics to provide a decision basis for planning and control to cope with the complex environmental changes in the outdoor scene of the agricultural mobile chassis. Front fusion and post-fusion are steps of multi-sensor fusion. Regardless of front fusion or post-fusion, the posture relationship between each sensor needs to be obtained. When installing sensors required for autonomous navigation on agricultural mobile chassis that do not have navigation functions, the appearance of the agricultural mobile chassis has often been designed, and there is no fixed installation position for these sensors. Therefore, in the process of installing sensors in the later stage, they will be arranged according to local conditions, so that the accurate installation position cannot be obtained. Without an accurate installation position, sensor fusion cannot be performed. The present invention can avoid the interference of these adverse factors and can provide subsequent developers with the sensor posture relationship required for front fusion and back fusion.

[0065] When each sensor is positioned, it has an independent coordinate system. Before fusion positioning, the coordinate systems of the sensors should be aligned first, and the coordinate transformation matrices between the sensors and between the sensors and the mobile chassis should be obtained. In the prior art, the positions of the sensors mounted on different mobile chassis are also different, which brings trouble to the alignment of the coordinate systems. This navigation device adopts standardized customization, and the coordinate transformation relationships between the sensors are known and fixed. Therefore, coordinate transformations are performed on each camera 3, inertial measurement unit, lidar 6, and GNSS, and the camera coordinate system a M 、the lidar coordinate system a L 、the IMU coordinate system a I 、and the GNSS coordinate system a G are normalized to the device coordinate system a E , and the conversion relationship is:

[0066] a E =R EM a M +t M ;

[0067] a E =R EL a L +t L ;

[0068] a E =R EG a G +t G ;

[0069] a E =R EI a I +t I ;

[0070] Among them, R EM , R EL , R EG , R EI are the rotation matrices of the camera, lidar, GNSS, and IMU coordinate systems respectively; t M , t L , t G , t I are the translation matrices of the camera, lidar, GNSS, and IMU coordinate systems respectively.

[0071] The key to the mobile chassis to achieve autonomous navigation is to install the positioning sensors required for autonomous navigation. Since the sensors are not rigidly connected to the chassis, there will be errors in the pose estimation obtained by the sensors when converted to the chassis, resulting in inaccurate pose estimation of the chassis. Most mobile chassis are equipped with angular velocity sensors, which can perform pose estimation of the chassis. The pose transformation relationship between the sensors and the chassis can be obtained by online registration with the sensors.

[0072] See Figure 3 and Figure 4 , Figure 3 which is a schematic diagram of a two-wheel differential odometer model according to an embodiment of the present invention, Figure 4 and which is a schematic diagram of the pose relationship of the odometer according to an embodiment of the present invention. Figure 4 It shows the wheel speed odometer path 10, the IMU odometer path 11 and the pose transformation relationship 12 between the two odometers. The mobile chassis mainly includes three models: Ackermann, two-wheel differential and four-wheel drive models. In order to make the navigation device compatible with all vehicle models, the Ackermann model and the four-wheel drive model are uniformly simplified into a two-wheel differential model for online registration. The online registration method of the navigation device of the present invention is used for aligning the coordinate systems between the navigation device and the mobile chassis, and obtaining the rotation matrix and translation matrix between the navigation device and the mobile chassis, and includes the following steps:

[0073] Step S100, obtaining the pose P of the wheel speed odometer installed on the mobile chassis O ;

[0074] The steering angular velocity of the mobile chassis:

[0075]

[0076] where ω is the steering angular velocity of the mobile chassis, V r is the right wheel speed of the mobile chassis, V1 is the left wheel speed of the mobile chassis, and l is the wheelbase of the mobile chassis.

[0077] The linear velocity of the mobile chassis:

[0078]

[0079] V is the linear velocity of the mobile chassis.

[0080] The position and heading of the mobile chassis:

[0081]

[0082] where the coordinates of the mobile chassis are (X0, Y0), t is the running time of the mobile chassis, and θ is the orientation of the mobile chassis.

[0083] Step S200, obtaining the pose P of the IMU odometer set on the navigation device I ;

[0084] Using the average angular velocity of the current moment t and the next moment t + 1 as the average angular velocity within the Δt time:

[0085]

[0086] where ω′ t is the average angular velocity, ω t is the angular velocity at time t, ω t+1 is the angular velocity at time t + 1, and b g is the angular velocity offset noise.

[0087] The pose at time t + 1 is approximately calculated using the angular velocity:

[0088] Q t+1 = Q t (ω′ t * Δ t );

[0089] where Q t+1 is the pose at time t + 1, Q t is the pose at time t, and Δt is the time interval.

[0090] The IMU acceleration is transformed from the body coordinate system to the world coordinate system and zero-bias compensation is performed:

[0091] a t+1，w = Q t+1 (a t+1，b - B a ) - g;

[0092] where a t+1，w is the acceleration in the world coordinate system, a t+1，b is the acceleration in the body coordinate system, and g is the acceleration due to gravity.

[0093] The acceleration within Δt is approximately represented by the average velocity at time t and t + 1:

[0094]

[0095] a′ t，w is the average acceleration in the world coordinate system, a t，w is the acceleration in the world coordinate system at time t, and a t+1，w is the acceleration in the world coordinate system at time t + 1.

[0096] The pose of the IMU can be expressed as:

[0097]

[0098] where P t+1 is the pose of the IMU at time t + 1, P t is the pose of the IMU at time t, and V t is the velocity of the IMU at time t.

[0099] Step S300: Sensor synchronization. Synchronize the wheel speed odometer and the IMU odometer. By performing a certain rotation and translation on the IMU odometer, it can be aligned with the wheel speed odometer. The expression is: where P O is the pose of the wheel speed odometer, and P I is the pose of the IMU odometer, is the pose transformation matrix for converting the IMU odometer to the wheel speed odometer;

[0100] The conversion relationship between the wheel speed odometer and the IMU odometer is:

[0101]

[0102] where R is the rotation matrix and t is the translation matrix;

[0103] Step S400: Joint optimization of external attitude parameters. Solve the pose transformation matrix between the wheel speed odometer and the IMU odometer to obtain the least squares solution: x * = arg min x ||Ax - y|| 2 . Where A is a matrix composed of multiple sets of consecutive IMU odometer data P I , x * is the quantity to be solved y is a matrix composed of multiple sets of wheel speed odometer data P O corresponding to the same moment as the IMU odometer data.

[0104] The essence of online calibration is to solve the pose transformation matrix between two sensors, is the quantity to be solved, and P O and P I are known. When a set of data is collected, a non - homogeneous over - determined system of equations in the form of Ax = y can be obtained, and the problem is then transformed into solving the least squares solution of this over - determined equation.

[0105] The least squares solution of the system of equations is: x * = argmin x ||Ax - y|| 2 .

[0106] First, perform singular value decomposition on matrix A: A = UDV T , and then let calculate b, where d i is the i - th element of the diagonal matrix D. Finally, obtain x * = Vb. UDT is a common mathematical method suitable for computers, and this least squares solution is obtained from the operations after the singular value decomposition of matrix A.

[0107] The present invention also provides a mobile chassis including the above navigation device, and the coordinate system alignment between the navigation device and the mobile chassis is performed by using the above online registration method.

[0108] See Figure 5 , Figure 5 which is the schematic diagram of the autonomous navigation of the mobile chassis according to an embodiment of the present invention. The autonomous navigation method of the mobile chassis of the present invention includes the following steps:

[0109] Step S100: Install the above navigation device on the mobile chassis; according to different scenarios, mobile chassis characteristics, navigation requirements, etc., different types of sensors can be selected for installation in the initial stage, and the coordinate transformation between the sensors is known. The sensors and processors carried by this navigation device can be selected and matched as needed. According to the scenario, a camera 3 can be selected indoors, while the navigation effect of the camera 3 is not good outdoors. GNSS can be selected for navigation outdoors, and there is no GNSS signal indoors. The lidar 6 can be used as an auxiliary for vision or GNSS for real-time obstacle avoidance. According to the working content of the mobile chassis, vision or radar can be used for line detection to achieve automatic navigation. All the sensors installed on this navigation device have fixed installation positions, so the coordinate transformation between the selected sensors is known;

[0110] Step S200: Connect the aviation plug of the control signal line and power supply line of the mobile chassis, and supply power to the navigation device through the mobile chassis. After the navigation device is powered on, each sensor starts to work. First, the coordinate systems between the navigation device and the mobile chassis are aligned, and the pose transformation matrix between the navigation device and the mobile chassis is solved, that is, the rotation matrix and translation matrix between the navigation device and the chassis; the above online registration method can be used to align the coordinate systems between the navigation device and the mobile chassis; measure the rotation and translation coordinate transformation between the navigation device coordinate system and the mobile chassis coordinate system, map the device coordinate system to the chassis coordinate system, and perform positioning and navigation. Through the above online registration method, the pose transformation matrix between the navigation device and the mobile chassis, that is, the rotation and translation relationship between the two, can be obtained. Since the mobile chassis navigation uses the chassis coordinate system, and the mobile chassis positioning uses the sensor coordinate system of the navigation device. The pose transformation matrix obtained through the online registration method can convert the sensors for positioning into the chassis coordinate system required for the mobile chassis navigation;

[0111] Step S300: Perform global and local positioning on the sensors of the navigation device and the mobile chassis, construct a surrounding environment map, and at the same time map the surrounding environment map in the global coordinate system;

[0112] Step S400: According to the current positioning and the surrounding environment map, the navigation device performs global and local path planning based on different task types. At this time, the mobile chassis can obtain the surrounding environment map. Global and local planning are part of autonomous navigation. When the mobile chassis obtains its work task, it starts path planning according to the work task. This planning includes local planning and global planning.

[0113] Step S500: Through the signal lines included in the aviation plug, the navigation device sends control signals including vehicle speed and steering angle to the mobile chassis; and

[0114] Step S600: The mobile chassis formulates corresponding control strategies according to the control signals and the corresponding odometer models. Among them, the odometer models include Ackermann mode, four-wheel four-rotation mode, and differential mode. The commands sent by the navigation device to the mobile chassis are only linear velocity and angular velocity, and the mobile chassis completes actions in different ways according to their respective characteristics when obtaining these commands. The essence of autonomous driving is to patrol along a pre-planned route, so it is necessary to know its own position and speed at all times to optimize the control strategy. For example: if there is a large bend ahead on the path, the vehicle speed at this time should be added to the control strategy in the underlying control instructions.

[0115] This embodiment may further include the following steps:

[0116] Step S700: During the movement of the mobile chassis, speed feedback is performed in real time, and the navigation device adjusts and optimizes the control signal according to the speed feedback signal.

[0117] And it includes the following steps:

[0118] Step S800: During the movement of the mobile chassis, the navigation device performs position feedback in real time, and the mobile chassis plans and adjusts the control strategy in real time according to the position change information.

[0119] Based on the same inventive concept, the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the above-mentioned autonomous navigation method for the mobile chassis is implemented.

[0120] The present invention is a multi-sensor fusion autonomous navigation device for agricultural mobile chassis positioning. It realizes absolute positioning through GNSS, relative positioning with the environment through Camera 3 and LiDAR 6, processes the information of each sensor through the embedded processor 5, and makes path planning and control decisions. It has a bus interface for communicating with the drive-by-wire chassis and can issue action commands to the chassis. It comes with an embedded processor 5, which is connected to each sensor and can obtain the data of each sensor, meeting the computational requirements for autonomous driving. It does not require a calibration board needed in traditional calibration schemes. Only through the online registration of the IMU configured in the navigation device and the wheel speedometer of the agricultural mobile chassis, high-precision pose estimation between the chassis and the navigation device can be achieved. By acquiring the raw data of the IMU and the wheel speedometer and based on the pose transformation relationship between the two, the poses of each are calculated. Since the IMU and the wheel speedometer measure the pose transformation of the same chassis, the pose relationship between the IMU and the wheel speedometer can be obtained by solving the pose transformation equation constructed by the odometer between the IMU and the wheel speedometer. By determining the relative positions of each sensor in advance through a specific structure, the complex coordinate transformation caused by the installation of various different types of sensors is avoided, thus reducing the installation cumulative error and enabling the rapid development of autonomous navigation functions for agricultural mobile chassis. During the process of installing sensors on the chassis, they are often arranged according to requirements. Therefore, the positions of sensors installed on different chassis or different designs are different. This navigation device is a fixed unified device that integrates the required sensors together and has fixed installation positions within the navigation device. The positions between each sensor are determined, so the installation cumulative error can be avoided.

[0121] Of course, the present invention can also have many other embodiments. Without departing from the spirit and essence of the present invention, those skilled in the art can make various corresponding changes and deformations according to the present invention. However, these corresponding changes and deformations should all fall within the protection scope of the appended claims of the present invention.

Claims

1. An online registration method for a navigation device, characterized in that, For aligning the coordinate systems between a navigation device and a mobile chassis, and obtaining the rotation matrix and translation matrix between the navigation device and the mobile chassis, the following steps are included: S100, Obtain the pose P of the wheel speed odometer installed on the mobile chassis O ; S200. Obtain the pose P of the IMU odometer set in the navigation device I ; S300. Synchronize the wheel speed odometer and the IMU odometer, rotate and translate the IMU odometer to align it with the wheel speed odometer. The conversion relationship between the wheel speed odometer and the IMU odometer is: wherein, is the pose transformation matrix for converting the IMU odometer to the wheel speed odometer; R is the rotation matrix and t is the translation matrix; and S400. Solve the pose transformation matrix between the wheel speed odometer and the IMU odometer to obtain the least-squares solution: x * = arg min x ‖Ax - y‖ 2 , where A is a matrix composed of multiple sets of consecutive IMU odometer data P I ; x * is the one to be solved ; and y is a matrix composed of multiple sets of wheel speed odometer data P corresponding to the same moment as the IMU odometer data O .

2. A navigation device is installed on a mobile chassis, characterized in that, Implement the coordinate system alignment between the navigation device and the mobile chassis by using the online registration method described in claim 1. The navigation device includes: A camera for collecting front image information for image-based positioning; An inertial measurement unit disposed in the camera for collecting acceleration and angle information for multi-sensor fusion-based positioning; A lidar for collecting surrounding point cloud information for point cloud-based positioning; A housing for determining the positional relationship between the camera and the lidar, and achieving spatial synchronization of the data of the camera, inertial measurement unit, and lidar; the camera is disposed in a slot on the front side of the top of the housing; the lidar is installed in the housing and is located below the camera; An embedded processor installed in the housing and located below the lidar for processing information from the camera, inertial measurement unit, lidar, and the mobile chassis, performing high-precision positioning and path planning, and issuing control instructions; the camera, inertial measurement unit, and lidar are respectively connected to the embedded processor; and A rear cover connected to the housing and jointly determining the relative coordinate relationship between the camera and the lidar with the housing. An interaction interface for connecting to an agricultural mobile chassis and a GNSS antenna interface are provided on the rear cover, and the interaction interface and the GNSS antenna interface are respectively connected to the embedded processor.

3. The navigation device of the mobile chassis according to claim 2, characterized in that, Adopt GNSS unified timekeeping and trigger simultaneously to ensure the alignment of timestamps when the camera, inertial measurement unit, and lidar are fused, and achieve the time synchronization of the data of the camera, inertial measurement unit, and lidar.

4. The navigation device of the mobile chassis according to claim 3, characterized in that, The coordinate transformation is performed on the camera, inertial measurement unit, lidar, and GNSS. The camera coordinate system a M , the lidar coordinate system a L , the IMU coordinate system a I , and the GNSS coordinate system a G are normalized to the device coordinate system a E . The conversion relationship is as follows: a E = R EM a M + t M ; a E = R EL a L + t L ; a E = R EG a G + t G ; a E = R EI a I + t I ; Among them, R EM , R EL , R EG , R EI are the rotation matrices of the camera, lidar, GNSS, and IMU coordinate systems respectively; t M , t L , t G , t I are the translation matrices of the camera, lidar, GNSS, and IMU coordinate systems respectively.

5. A mobile chassis, characterized in that, Include the navigation device described in any one of claims 2-4.

6. An autonomous navigation method for a mobile chassis, characterized in that, Include the following steps: S100. Install the navigation device described in any one of claims 2-4 on a mobile chassis; S200. Use the online registration method described in claim 1 to perform coordinate system alignment between the navigation device and the mobile chassis, and solve the pose transformation matrix between the navigation device and the mobile chassis; S300. Perform global and local positioning on each sensor of the navigation device and the mobile chassis, and construct a surrounding environment map, and at the same time map the surrounding environment map in the global coordinate system; S400. The navigation device performs global and local path planning according to the current positioning and the surrounding environment map based on the task type; S500. The navigation device sends control signals including vehicle speed and turning angle to the mobile chassis; and S600. The mobile chassis formulates corresponding control strategies according to the control signals and the corresponding odometer model.

7. The autonomous navigation method of the mobile chassis according to claim 6, characterized in that, Also include the following steps: During the movement of the mobile chassis S700, speed feedback is performed in real time, and the navigation device adjusts and optimizes the control signal according to the speed feedback signal.

8. The autonomous navigation method of the mobile chassis according to claim 6 or 7, characterized in that The following steps are further included: S800, during the movement of the mobile chassis, the navigation device performs position feedback in real time, and the mobile chassis plans and adjusts the control strategy in real time according to the position change information.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the autonomous navigation method of the mobile chassis according to any one of claims 6 to 8.

Images