Following type ultra-large carrying vehicle positioning method and system based on laser radar

Abstract

The invention discloses a laser radar-based following type ultra-large carrying vehicle positioning method and system, and relates to the technical field of multi-vehicle intelligent cooperative positioning. The method comprises the following steps: establishing a coordinate system where a front vehicle, a rear vehicle, goods and a positioning module are located, calibrating the coordinate system where the front vehicle, the rear vehicle, the goods and the positioning module are located, and storing a calibration result; point cloud information of surrounding environments of the front vehicle and the rear vehicle is obtained based on the laser radar, respective point cloud maps of the front vehicle and the rear vehicle are constructed, and self-vehicle positioning of the front vehicle or the rear vehicle is completed based on a laser radar inertial odometer in the positioning module. According to the invention, a single-vehicle autonomous positioning method, a multi-vehicle cooperative positioning method and a cargo attitude estimation method are integrated, a factor graph optimization method is adopted to fuse workshop perception and in-vehicle prediction results, and accurate and reliable following type carrying vehicle cooperative positioning covering multi-source observation information and a vehicle-cargo kinematics action relationship is realized. And the state estimation real-time performance and the positioning precision of the following type super-large carrying vehicle are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of multi-vehicle intelligent collaborative positioning, and in particular to a laser radar-based following type oversized transport vehicle positioning method. Background Art

[0002] Currently, a general-purpose transport vehicle cannot carry oversized items such as large bridge structures. The common solution is to manufacture special transport vehicles for oversized items, which is costly and has low versatility. In contrast, multi-vehicle collaborative transport is a more universal solution. Due to the large mass and high center of mass of oversized goods, the positioning deviation of multiple workshops can easily cause damage to the goods, which may cause significant loss of life and property in serious cases. In addition, the current inefficient manual collaborative transport cannot meet the rapidly growing demand for oversized transport.

[0003] To this end, the present invention proposes a laser radar-based tracking oversized transport vehicle positioning method and system. Summary of the invention

[0004] The purpose of the present invention is to provide a laser radar-based following oversized load transport vehicle positioning method and system, which can integrate multi-source information and vehicle-cargo kinematic characteristics, reduce the uncertainty factors caused by the self-positioning process of the leading or trailing vehicle, improve the positioning accuracy, and realize the collaborative positioning of multiple vehicles and cargo posture estimation when following heavy-duty vehicles cooperate to carry oversized loads, and has strong environmental adaptability.

[0005] According to the first aspect of the present invention, in order to achieve the above-mentioned purpose, the present invention provides the following technical solutions: a laser radar-based follow-up oversized load transport vehicle positioning method, which is used to estimate the relative posture of the front vehicle and the rear vehicle that cooperate to carry oversized loads, and is implemented by using a front vehicle positioning module arranged at the front end of the front vehicle, a rear vehicle positioning module arranged at the rear end of the rear vehicle, and a feature identifier arranged on the front end surface of the support device, comprising the following steps: Establish the coordinate system of the front vehicle, the rear vehicle, the cargo, and the positioning module, calibrate the coordinate system of the front vehicle, the rear vehicle, the cargo, and the positioning module, and store the calibration results; Based on the laser radar, the point cloud information of the surrounding environment of the front and rear vehicles is obtained, and the point cloud maps of the front and rear vehicles are constructed. The self-positioning of the front or rear vehicle is completed based on the laser radar inertial odometer in the positioning module, and then the rear vehicle sends the IMU measurement value and posture to the front vehicle; The relative posture between the front vehicle and the cargo is obtained based on the front vehicle positioning module and the feature markers on the support device. The Kalman filter is used to estimate the cargo posture in real time based on the kinematic equation between the vehicle and the cargo. Multi-vehicle collaborative positioning is achieved based on real-time estimation of cargo posture. When the posture change of the rear vehicle reaches the Nth key frame, the point cloud and the posture of the rear vehicle at this moment are transmitted to the front vehicle. The positioning of the rear vehicle and the cargo is corrected through the factor graph optimization method, and the optimized rear vehicle state is transmitted to the point cloud map of the rear vehicle to correct the map information.

[0006] Furthermore, the coordinate systems of the front vehicle, the rear vehicle, the cargo, and the positioning module are established, and the coordinate systems of the front vehicle, the rear vehicle, the cargo, and the positioning module are calibrated, as follows: (21) Positioning module calibration Get IMU coordinate system For the positioning module coordinate system, the mobile positioning module, the laser radar uses the ICP point cloud registration method to estimate the laser radar coordinate system The IMU uses the pre-integration method to estimate the posture change, and the hand-eye calibration method is used to determine the coordinates of each lidar. Coordinate calibration extrinsic parameters , where the calibration external parameter By rotation matrix and displacement vector composition: ; (22) Calibration of positioning module and vehicle coordinate system Take the support position of the front or rear vehicle as the origin and establish the vehicle coordinate system of the front or rear vehicle , take the forward direction of the front or rear vehicle as the positive direction of the X axis, the vertical upward direction of the front or rear vehicle as the positive direction of the Z axis, temporarily arrange RTK at the support position of the front or rear vehicle, move the front or rear vehicle, obtain the position change of the positioning module and the front or rear vehicle respectively, and obtain the calibration external parameters from the positioning module coordinates to the front or rear vehicle coordinates through the hand-eye calibration method ; (23) Calibration of the front vehicle coordinate system and the rear vehicle coordinate system After the front vehicle arrives at the loading location, a point cloud environment map around the loading location is constructed based on the LiDAR inertial odometer, and the accuracy of the point cloud map is improved through loop correction. While the front vehicle and the rear vehicle are stationary waiting to load the oversized cargo, the rear vehicle will The laser radar point cloud under the The established point cloud map is used for point cloud registration to obtain the initial calibration external parameters from the rear vehicle coordinates to the front vehicle coordinates , and take the position of the front vehicle coordinate system at this time as the world coordinate system origin; (24) Calibration of vehicle coordinate system and cargo coordinate system The direction of the cargo from the rear vehicle support point to the front vehicle support point is taken as the cargo posture in the world coordinate system. During transportation, the vehicle and the cargo are articulated. According to the calibrated external parameters in step (23), The distance between the two supporting points of the corresponding goods can be directly obtained and the initial heading angle in the world coordinate system ; (25) Feature identification and cargo coordinate system calibration The laser radar in the front vehicle positioning module recognizes the feature mark on the support device, filters the point cloud based on the laser echo intensity, obtains the feature mark and calculates the plane normal vector , based on the calibration external parameters of the positioning module coordinates to the world coordinate system , observe the laser radar coordinates of the goods to Coordinate calibration extrinsic parameters Get the normal vector Representation in world coordinate system ,calculate Projection on the XOY plane of the world coordinate system and cargo heading angle Angle .

[0007] Furthermore, (31) a laser radar is used to obtain point cloud information of the surrounding environment, a Gaussian filter is used to smooth the point cloud data and remove noise, and an incremental KD tree is used as the data structure of the point cloud space to establish a point cloud map; Gaussian filtering: multiply the k nearest neighbors of each point by the Gaussian weight G, and then perform weighted average to obtain the filtered point cloud coordinate value : In the formula, is the distance from the point cloud to the k nearest neighbor points, is the standard deviation of the Gaussian function, is the current point coordinate, For the neighborhood The coordinates of the points; (32) Combined with the calibrated external parameters, the IMU uses linear interpolation displacement and spherical interpolation attitude to track the movement of the lidar coordinate system, align all point clouds to the IMU coordinate system at the same time, and eliminate point cloud motion distortion; (33) Based on the point cloud map established in real time, the point-to-surface distance is used as the observation error, and the positioning information of the leading and following vehicles is updated in real time through the error state Kalman filter. and ; The discrete update equation for the positioning of the leading or following vehicle is: In the formula, express State quantity at time , express State quantity at time , express Observable quantity at time , express Noise of the moment ; State quantity , Observation ,noise and function f They are: Represents the rotation matrix, translation vector, and velocity vector of the leading or trailing vehicle in the world coordinate system. Represent the zero bias of the accelerometer and gyroscope respectively, Represents the sampling time interval of IMU, Represents acceleration and angular velocity, "~" represents the IMU observation, represents the matrix transpose, assuming zero bias , Satisfies the Wiener process, noise The components of are all independent zero-mean Gaussian white noise, symbol Used to parameterize the state error on an n-dimensional manifold The specific operation rules are as follows: is the exponential map in the three-dimensional rotation group SO(3); The observation error of the point-surface distance is defined as: In the formula, Represents the rigid transformation matrix from the front or rear vehicle's ego coordinate system to the world coordinate system, Represents the point cloud coordinates obtained in the vehicle coordinate system of the front or rear vehicle. Including the front car Or the rear car , Point Cloud The nearest neighbor point in the point cloud map, for The normal vector of the corresponding plane; (34) The key frame change threshold is set to 0.4 m displacement or 10° heading angle or 0.5 s. The rear vehicle sends the positioning result and IMU measurement value to the front vehicle at the point cloud key frame.

[0008] Furthermore, the relative posture between the front vehicle and the cargo is observed based on the front vehicle positioning module and the characteristic identification on the support device. Based on the kinematic equation between the vehicle and the cargo, the Kalman filter is used to estimate the cargo posture in real time, as follows: (41) According to the kinematic relationship between vehicle and cargo, the cargo posture prediction equation (1) and the corresponding discretization equation (2) are obtained: (1) (2) In the formula, , , , , , , , Represent the lateral velocity, longitudinal velocity, yaw rate and heading angle of the vehicle coordinate system respectively. , , , The subscripts 1 and 3 represent the front and rear vehicles respectively. and Respectively represent the distances from the front and rear vehicle coordinate systems to their respective support points, is a discrete time interval, is the process noise, and its variance is assumed to be Zero-mean Gaussian white noise, a and b are the representations of the formula derivation results; (42) Control the laser radar of the front vehicle positioning module to obtain the normal vector of the feature identifier of the support device in the front vehicle coordinate system , combined with the positioning results of the positioning module to obtain the heading angle measurement value of the cargo in the world coordinate system , so we can get the observation equation (3): (3) In the formula, is the measurement noise, and is assumed to be the variance Zero-mean Gaussian white noise; (43) Predict the discretized equation (2) and the observation equation (3), and use Kalman filtering to update the cargo heading angle in real time .

[0009] Furthermore, multi-vehicle collaborative positioning is achieved based on the real-time estimation of cargo posture. When the posture change of the rear vehicle reaches the Nth key frame, the point cloud and the posture of the rear vehicle at this moment are transmitted to the front vehicle, and the positioning of the rear vehicle and the cargo is corrected by the factor graph optimization method, as follows: (51) When the estimated posture of the rear vehicle reaches When a key frame changes the threshold, the point cloud information at this moment is sent to the preceding vehicle; (52) Definition State variables to be optimized for multi-vehicle transportation system with time-to-time tracking for: (4) In the formula , represent The position of the front and rear vehicles in the world coordinate system at the moment, represent The state of the front vehicle IMU in the front vehicle coordinate system at the moment, including the acquired acceleration, angular velocity, and zero bias of the accelerometer and angular velocity meter. represent The state of the rear vehicle IMU in the rear vehicle coordinate system at time t, including the obtained acceleration , angular velocity , Accelerometer bias and the zero bias of the angular velocity sensor ; M is the number of key frames of the rear vehicle posture involved in a factor graph optimization; (53) IMU is realized by pre-integration method Time to The estimation of vehicle posture change at the moment is as follows: (5) In the formula and are the discretized zero-mean Gaussian white noise, and are the zero bias of the accelerometer and gyroscope at time k, Represents a state quantity from i Time has come j The amount of change at a moment; Construct the state residual based on the pre-integrated estimation result of the IMU measurement value: (6) In the formula, is gravity, for The inverse mapping of (54) The point cloud map of the rear vehicle and the point cloud map of the front vehicle is constructed with the point-to-surface distance as the error. : (7) In the formula Point cloud for the rear vehicle The nearest neighbor point in the point cloud map, for The normal vector of the corresponding plane; (55) Based on cargo heading angle The prediction and observation equations are constructed to construct the residuals and: (8) (9) (56) Based on the construction residual, the sliding window algorithm and factor graph optimization are used to solve the optimal , to achieve multi-vehicle collaborative positioning, as follows: (10) In the formula, represents the Huber norm, represents the Mahalanobis distance, Represents the IMU measurement residual, represents the residual of the rear vehicle point cloud, Represents the residual error of cargo prediction posture, Represents the residual of cargo observation posture; (57) The corrected position of the rear vehicle is sent to the front vehicle, and the positioning status and point cloud map of the rear vehicle are updated.

[0010] According to a second aspect of the present invention, the present invention provides a laser radar-based follow-up oversized transport vehicle positioning system, which is used to implement the above-mentioned laser radar-based follow-up oversized transport vehicle positioning method, comprising: A calibration unit is used to establish a coordinate system where the front vehicle, the rear vehicle, the cargo, and the positioning module are located, calibrate the coordinate system where the front vehicle, the rear vehicle, the cargo, and the positioning module are located, and store the calibration results; A construction unit is used to obtain point cloud information of the surrounding environment of the front vehicle and the rear vehicle based on the laser radar, construct point cloud maps of the front vehicle and the rear vehicle respectively, and complete the self-vehicle positioning of the front vehicle or the rear vehicle based on the laser radar inertial odometer in the positioning module, and then enable the rear vehicle to send the IMU measurement value and posture to the front vehicle; The cargo posture estimation unit is used to obtain the relative posture between the leading vehicle and the cargo based on the leading vehicle positioning module and the characteristic identification on the supporting device, and to estimate the cargo posture in real time using Kalman filtering based on the kinematic equation between the vehicle and the cargo; The multi-vehicle collaborative positioning unit is used to realize multi-vehicle collaborative positioning based on the real-time estimation of the cargo posture. When the posture change of the rear vehicle reaches the Nth key frame, the point cloud and the posture of the rear vehicle at this moment are transmitted to the front vehicle. The positioning of the rear vehicle and the cargo is corrected through the factor graph optimization method, and the optimized rear vehicle state is transmitted to the point cloud map of the rear vehicle to correct the map information.

[0011] Further, the calibration unit includes a front vehicle positioning module arranged at the front end of the front vehicle and a rear vehicle positioning module arranged at the rear end of the rear vehicle; The front vehicle positioning module includes three laser radars that perceive surrounding space information, a laser radar that observes the cargo posture, a front vehicle positioning calculation module, an IMU and a communication module. The laser wavelength of the laser radar that observes the cargo posture is 905nm; The rear vehicle positioning module includes two laser radars for sensing surrounding space information, a rear vehicle positioning calculation module, and an IMU and communication module.

[0012] Furthermore, the cargo posture estimation unit includes a supporting device and a characteristic mark arranged on the front end surface of the supporting device, and the characteristic mark is set as a reflective tape.

[0013] According to a third aspect of the present invention, the present invention provides a terminal device, comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the memory stores a computer program capable of running on the processor, and when the processor loads and executes the computer program, the above-mentioned laser radar-based following oversized transport vehicle positioning method is adopted.

[0014] According to a fourth aspect of the present invention, the present invention provides a storage medium comprising computer executable instructions, which, when executed by a computer processor, are used to execute the above-mentioned laser radar-based following oversized transport vehicle positioning method.

[0015] The present invention has at least the following beneficial effects: 1. The positioning method proposed in the present invention integrates single-vehicle autonomous positioning, multi-vehicle collaborative positioning and cargo posture estimation methods, and adopts a factor graph optimization method to integrate workshop perception and in-vehicle prediction results, thereby realizing accurate and reliable collaborative positioning of follow-up transport vehicles covering multi-source observation information and vehicle-cargo kinematic interaction relationships, thereby improving the real-time state estimation and positioning accuracy of follow-up oversized transport vehicles.

[0016] 2. The modular arrangement of sensors for the positioning module proposed in the present invention is not only convenient for the calibration and debugging of the sensors and the coordinate system of the leading or trailing vehicle, but also improves the flexibility of sensor arrangement when faced with spatial constraints of irregular and oversized cargo, and leaves a calibrated sensor arrangement platform for other environmental perception needs, for example, providing a good hardware platform for vehicle holographic environmental perception.

[0017] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a flow chart of the positioning method according to the first embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of the positioning system according to the second embodiment of the present invention; Figure 3 This is a schematic diagram of calibrating external parameters between a front vehicle, a rear vehicle, cargo, and a positioning module in the first embodiment of the present invention; Figure 4 Schematic diagram of the principle of multi-vehicle collaborative positioning based on factor graph optimization and sliding window algorithm in Embodiment 1 of the present invention. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present disclosure to clearly and completely describe the technical solutions in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present disclosure.

[0020] Related terms explanation: Vehicle-cargo refers to a vehicle and the cargo carried on the vehicle, where the vehicle includes the front vehicle and the rear vehicle. It can refer to the front vehicle or the rear vehicle in general or specifically. When it comes to describing the relationship between a specific vehicle and cargo, it refers to the specific vehicle. For example, the kinematic equation between vehicle-cargo refers to the kinematic equation between the front vehicle or the rear vehicle and the cargo. Vehicle: includes the front vehicle and the rear vehicle. For example, the vehicle coordinate system includes the front vehicle coordinate system and the rear vehicle coordinate system.

[0021] Embodiment 1: See also Figure 1 The present invention provides a technical solution: a laser radar-based tracking oversized load transport vehicle positioning method, which is used to estimate the relative posture of the front vehicle and the rear vehicle that cooperate to transport oversized loads, and is implemented by using a front vehicle positioning module set at the front end of the front vehicle, a rear vehicle positioning module set at the rear end of the rear vehicle, and a feature identifier set at the front end surface of the support device, including the following steps: S1. Establish the coordinate system of the front vehicle, the rear vehicle, the cargo, and the positioning module, calibrate the coordinate system of the front vehicle, the rear vehicle, the cargo, and the positioning module, and store the calibration results, such as Figure 3 As shown, the details are as follows: (S11) Positioning module calibration Get IMU coordinate system For the positioning module coordinate system, the mobile positioning module, the laser radar uses the ICP point cloud registration method to estimate the laser radar coordinate system The IMU uses the pre-integration method to estimate the posture change, and the hand-eye calibration method is used between each sensor to determine the coordinates of each lidar Coordinate calibration extrinsic parameters , where the calibration external parameter By rotation matrix and displacement vector composition: ; (S12) Calibration of positioning module and vehicle coordinate system Establish the vehicle coordinate system with the vehicle (front vehicle and / or rear vehicle) support position as the origin , taking the vehicle's forward direction as the positive direction of the X axis and the vehicle's vertical upward direction as the positive direction of the Z axis, temporarily arrange RTK at the vehicle's support position, move the vehicle, obtain the position changes of the positioning module and the vehicle respectively, and obtain the calibration external parameters from the positioning module coordinates to the vehicle coordinates through the hand-eye calibration method ; (S13) Calibration of the front vehicle coordinate system and the rear vehicle coordinate system After the front vehicle arrives at the loading location, a point cloud environment map around the loading location is constructed based on the LiDAR inertial odometer, and the accuracy of the point cloud map is improved through loop correction. While the front vehicle and the rear vehicle are stationary waiting to load the oversized cargo, the rear vehicle will The laser radar point cloud under the The established point cloud map is used for point cloud registration to obtain the initial calibration external parameters from the rear vehicle coordinates to the front vehicle coordinates , and take the position of the front vehicle coordinate system at this time as the world coordinate system origin; (S14) Calibration of vehicle coordinate system and cargo coordinate system The direction of the cargo from the rear vehicle support point to the front vehicle support point is taken as the posture of the cargo in the world coordinate system. During transportation, the vehicle and the cargo are articulated. According to the calibrated external parameters in step (S13), The distance between the two supporting points of the corresponding goods can be directly obtained and the initial heading angle in the world coordinate system ; (S15) Feature identification and cargo coordinate system calibration The laser radar in the front vehicle positioning module recognizes the feature mark on the support device, filters the point cloud based on the laser echo intensity, obtains the feature mark and calculates the plane normal vector , based on the above calibration external parameters, the normal vector is obtained Representation in world coordinate system ,calculate Projection on the XOY plane of the world coordinate system and cargo heading angle Angle ; It should be noted that the aforementioned calibration external parameters include: calibration external parameters from the positioning module coordinates to the world coordinate system (It can also be expressed as , that is, V is the calibration external parameter from the positioning module coordinates obtained by the front vehicle to the vehicle coordinates , since the vehicle is not moving, the world coordinate system is considered to be the same as the front vehicle coordinate system, is the unit matrix), the laser radar coordinates of the observed goods to Coordinate calibration extrinsic parameters (i.e. i=1, corresponding to the laser radar serial number of the front vehicle observing the goods, the laser radar coordinates to Coordinate calibration extrinsic parameters ).

[0022] ; S2. Based on the laser radar, the point cloud information of the surrounding environment of the front and rear vehicles is obtained, and the point cloud maps of the front and rear vehicles are constructed. The self-positioning of the front or rear vehicle is completed based on the laser radar inertial odometer in the positioning module, and then the rear vehicle sends the IMU measurement value and posture to the front vehicle, as follows: (S21) Using laser radar to obtain point cloud information of the surrounding environment, using Gaussian filtering to smooth the point cloud data and remove noise, using incremental KD tree as the data structure of the point cloud space to reduce the depth imbalance caused by a large amount of data addition and deletion, and establishing a point cloud map; Gaussian filtering: multiply the k nearest neighbors of each point by the Gaussian weight G, and then perform weighted average to obtain the filtered point cloud coordinate value : In the formula, is the distance from the point cloud to the k nearest neighbor points, is the standard deviation of the Gaussian function, is the current point coordinate, For the neighborhood The coordinates of the points; (S22) Combined with the calibrated external parameters, the IMU uses linear interpolation displacement and spherical interpolation attitude to track the movement of the laser radar coordinate system, align all point clouds to the IMU coordinate system at the same time, and eliminate point cloud motion distortion; (S23) Based on the point cloud map established in real time, the point-to-surface distance is used as the observation error, and the positioning information of the leading and following vehicles is updated in real time through the error state Kalman filter. and ; The discrete update equation for the positioning of the leading or following vehicle is: In the formula, express State quantity at time , express State quantity at time , express Observable quantity at time , express Noise of the moment ; State quantity , Observation ,noise and function f They are: Represents the rotation matrix, translation vector, and velocity vector of the leading or trailing vehicle in the world coordinate system. Represent the zero bias of the accelerometer and gyroscope respectively, Represents the sampling time interval of IMU, Represents acceleration and angular velocity, "~" represents the IMU observation, represents the matrix transpose, assuming zero bias , Satisfies the Wiener process, noise The components of are all independent zero-mean Gaussian white noise, symbol Used to parameterize the state error on an n-dimensional manifold The specific operation rules are as follows: is the exponential map in the three-dimensional rotation group SO(3); The observation error of the point-surface distance is defined as In the formula, Represents the rigid transformation matrix from the front or rear vehicle's ego coordinate system to the world coordinate system, Represents the point cloud coordinates obtained in the vehicle coordinate system of the front or rear vehicle. Including the front car Or the rear car , Point Cloud The nearest neighbor point in the point cloud map, for The normal vector of the corresponding plane; (S24) setting the key frame change threshold to 0.4 m displacement or 10° heading angle or 0.5 s, and the rear vehicle sends the positioning result and IMU measurement value to the front vehicle when the point cloud key frame is reached; S3. Based on the observation of the front vehicle positioning module and the feature identification on the support device, the relative posture between the front vehicle and the cargo is obtained. Based on the kinematic equation between the vehicle and the cargo, the Kalman filter is used to estimate the cargo posture in real time, as follows: (S31) According to the kinematic relationship between the vehicle and the cargo, the cargo posture prediction equation (1) and the corresponding discretization equation (2) are obtained: (1) (2) In the formula, , , , ; , , , Represent the lateral velocity, longitudinal velocity, yaw rate and heading angle of the vehicle coordinate system respectively. , , , The subscripts 1 and 3 represent the front and rear vehicles respectively. and Respectively represent the distances from the front and rear vehicle coordinate systems to their respective support points, is a discrete time interval, is the process noise, and its variance is assumed to be Zero-mean Gaussian white noise, a and b are the representations of the formula derivation results; (S32) Controlling the laser radar of the front vehicle positioning module to obtain the normal vector of the feature identifier of the support device in the front vehicle coordinate system , combined with the positioning results of the positioning module to obtain the heading angle of the cargo in the world coordinate system , so we can get the observation equation (3): (3) In the formula, is the measurement noise, and is assumed to be the variance Zero-mean Gaussian white noise; (S33) Predict the discretized equation (2) and the observation equation (3), and use Kalman filtering to update the cargo heading angle in real time ; S4. Realize multi-vehicle collaborative positioning based on real-time estimation of cargo posture. When the rear vehicle posture changes to the Nth key frame, the point cloud and the rear vehicle posture at this moment are transmitted to the front vehicle. The positioning of the rear vehicle and the cargo (rear vehicle posture) is corrected by the factor graph optimization method, and the optimized rear vehicle state is transmitted to the point cloud map of the rear vehicle to correct the map information, such as Figure 4 As shown, the details are as follows: (S41) When the estimated posture of the rear vehicle (the position of the rear vehicle) reaches the first When a key frame changes to a threshold, the point cloud information at that moment and the estimated position and posture of the rear vehicle are sent to the front vehicle; (S42) Definition State variables to be optimized for multi-vehicle transportation system with time-to-time tracking for: (4) In the formula , represent The position of the front and rear vehicles in the world coordinate system at the moment, represent The state of the IMU factor of the front vehicle in the front vehicle coordinate system at the moment, including the obtained acceleration, angular velocity, and zero bias of the accelerometer and angular velocity meter. represent The state of the rear vehicle IMU in the rear vehicle coordinate system at time t, including the obtained acceleration , angular velocity , Accelerometer bias and the zero bias of the angular velocity sensor ; M is the number of key frames of the rear vehicle posture involved in a factor graph optimization; (S43) IMU is realized by pre-integration method Time to The estimation of vehicle posture change at the moment is as follows: IMU is realized by pre-integration method Time to The estimation of vehicle posture change at the moment is as follows: (5) In the formula and are the discretized zero-mean Gaussian white noise, and are the zero bias of the accelerometer and gyroscope at time k, It is a composite function of the mapping from vector to antisymmetric matrix and the exponential mapping from Lie algebra to Lie group, which converts the rotation vector into a rotation matrix. , Represents a state quantity from i Time has come j The amount of change at a moment; Construct the state residual based on the pre-integrated estimation result of the IMU measurement value: (6) In the formula, is gravity, for The inverse mapping of (S44) The point cloud of the rear vehicle and the point cloud map of the front vehicle are constructed with the point-surface distance as the error. : (7) In the formula Point cloud for the rear vehicle The nearest neighbor point in the point cloud map, for The normal vector of the corresponding plane; (S45) Based on cargo heading angle The prediction and observation equations are constructed to construct the residuals and: (8) (9) (S46) Based on the above constructed residual, the sliding window algorithm and factor graph optimization are used to solve the optimal , to achieve multi-vehicle collaborative positioning, as follows: (10) In the formula, represents the Huber norm, represents the Mahalanobis distance, Represents the IMU measurement residual, represents the residual of the rear vehicle point cloud, Represents the residual error of cargo prediction posture, Represents the residual of cargo observation posture; (S47) The corrected position of the rear vehicle is sent to the front vehicle, and the positioning status and point cloud map of the rear vehicle are updated.

[0023] It should be noted that in factor graph optimization and sliding window algorithm, , , N is the key frame threshold for the rear vehicle to send perception results to the front vehicle. Whenever the estimated posture of the rear vehicle reaches the Nth key frame, the perception result of the rear vehicle at that moment (the point cloud at that moment, the positioning result corresponding to the unsent rear vehicle key frame and the pre-integration result of the IMU observation during the period) is sent to the front vehicle; M is the number of rear vehicle posture key frames involved in one factor graph optimization.

[0024] Embodiment 2: This embodiment provides a laser radar-based tracking oversized cargo transport vehicle positioning system, which is used to implement the laser radar-based tracking oversized cargo transport vehicle positioning method described in the first embodiment, including: A calibration unit is used to establish a coordinate system where the front vehicle, the rear vehicle, the cargo, and the positioning module are located, calibrate the coordinate system where the front vehicle, the rear vehicle, the cargo, and the positioning module are located, and store the calibration results; A construction unit is used to obtain point cloud information of the surrounding environment of the front vehicle and the rear vehicle based on the laser radar, construct point cloud maps of the front vehicle and the rear vehicle respectively, and complete the self-vehicle positioning of the front vehicle or the rear vehicle based on the laser radar inertial odometer in the positioning module, and then enable the rear vehicle to send the IMU measurement value and posture to the front vehicle; The cargo posture estimation unit is used to obtain the relative posture between the leading vehicle and the cargo based on the leading vehicle positioning module and the characteristic identification on the supporting device, and to estimate the cargo posture in real time using Kalman filtering based on the kinematic equation between the vehicle and the cargo; The multi-vehicle collaborative positioning unit is used to realize multi-vehicle collaborative positioning based on the real-time estimation of the cargo posture. When the posture change of the rear vehicle reaches the Nth key frame, the point cloud and the posture of the rear vehicle at this moment are transmitted to the front vehicle. The positioning of the rear vehicle and the cargo is corrected through the factor graph optimization method, and the optimized rear vehicle state is transmitted to the point cloud map of the rear vehicle to correct the map information.

[0025] Further, the calibration unit includes a front vehicle positioning module arranged at the front end of the front vehicle and a rear vehicle positioning module arranged at the rear end of the rear vehicle; The front vehicle positioning module includes three laser radars that perceive the surrounding space information, a laser radar that observes the cargo posture, a front vehicle positioning calculation module, an IMU and a communication module. The laser wavelength of the laser radar that observes the cargo posture is 905nm. The rear vehicle positioning module includes two laser radars for sensing surrounding space information, a rear vehicle positioning calculation module, and an IMU and communication module.

[0026] It should be noted that the front vehicle positioning calculation module is used to control the laser radar in the positioning module, estimate the cargo posture, establish the point cloud map obtained by the front vehicle observation, integrate the point cloud information of the rear vehicle and the point cloud map of the front vehicle, and realize collaborative positioning; the IMU is used to observe the motion state of the front vehicle in real time and estimate the posture of the front vehicle; the communication module is used to receive the perception results from the rear vehicle and send the corrected rear vehicle positioning results; The rear vehicle positioning calculation module is used to control the laser radar in the positioning module, establish the point cloud map obtained by the rear vehicle observation, and estimate the absolute position and posture of the rear vehicle; the IMU is used to observe the motion state of the rear vehicle in real time and estimate the position and posture of the rear vehicle; the communication module is used to send the perception results of the rear vehicle to the front vehicle and receive the rear vehicle correction positioning from the front vehicle.

[0027] Furthermore, the cargo posture estimation unit includes a supporting device and a characteristic mark arranged on the front end surface of the supporting device, and the characteristic mark is set as a reflective tape.

[0028] Specifically, the above-mentioned calibration unit, construction unit, cargo posture estimation unit and multi-vehicle collaborative positioning unit can be embedded in a computer processing system. The computer calls the above-mentioned units to complete the task of real-time positioning of the front vehicle and the rear vehicle according to the above-mentioned laser radar-based following oversized transport vehicle positioning method; the above-mentioned calibration unit, construction unit, cargo posture estimation unit and multi-vehicle collaborative positioning unit can perform operations according to the specific steps given in the laser radar-based following oversized transport vehicle positioning method.

[0029] It should be noted that it should be understood that the division of the various units of the above system is only a division of logical functions. In actual implementation, they can be fully or partially integrated into one physical entity, or they can be physically separated, and these units can all be implemented in the form of software calling through processing elements; they can also be all implemented in the form of hardware; some units can also be implemented in the form of software calling through processing elements, and some units can be implemented in the form of hardware. For example, the calibration unit can be a separately established processing element, or it can be integrated in a chip of the above-mentioned device. In addition, it can also be stored in the memory of the above-mentioned device in the form of program code, and called and executed by a processing element of the above-mentioned device. The implementation of other units is similar. In addition, all or part of these units can be integrated together, or they can be implemented independently. The processing element described here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each of the above units can be completed by the hardware integrated logic circuit in the processor element or software instructions.

[0030] For example, the above units may be one or more integrated circuits configured to implement the above methods, such as one or more application specific integrated circuits (ASICs), or one or more digital singnal processors (DSPs), or one or more field programmable gate arrays (FPGAs). For another example, when a certain unit above is implemented in the form of a processing element scheduling program code, the processing element may be a general-purpose processor, such as a central processing unit (CPU) or other processors that can call program code. For another example, these units may be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0031] In summary, the present invention integrates single-vehicle autonomous positioning, multi-vehicle collaborative positioning and cargo posture estimation methods, and adopts factor graph optimization method to fuse workshop perception and in-vehicle prediction results, so as to realize accurate and reliable follow-up transport vehicle collaborative positioning covering multi-source observation information and vehicle-cargo kinematic interaction relationship. In addition, the modular layout scheme of sensors in the proposed positioning module is not only convenient for the calibration and debugging of sensors and vehicle coordinate systems, but also improves the flexibility of sensor layout when facing spatial constraints of irregular and oversized cargo, and also leaves a calibrated sensor layout platform for other environmental perception needs, for example, it can provide a good hardware platform for vehicle holographic environmental perception.

[0032] Embodiment three: The present invention provides a terminal device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor. The memory stores a computer program capable of running on the processor. When the processor loads and executes the computer program, the above-mentioned laser radar-based following oversized transport vehicle positioning method is adopted.

[0033] It should be noted that the terminal device can be a computer device such as a desktop computer, a laptop computer or a cloud server, and the terminal device includes but is not limited to a processor and a memory. For example, the terminal device can also include input and output devices, network access devices and buses.

[0034] Furthermore, the processor may adopt a central processing unit (CPU). Of course, according to actual usage, other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. may also be adopted. The general-purpose processor may adopt a microprocessor or any conventional processor, etc., and this application does not impose any restrictions on this.

[0035] Embodiment 4: The present invention provides a storage medium containing computer executable instructions, which are used to execute the above-mentioned laser radar-based following oversized transport vehicle positioning method when executed by a computer processor.

[0036] Among them, the computer program can be stored in a computer-readable medium, the computer program includes computer program code, the computer program code can be in the form of source code, object code, executable file or certain middleware, etc. The computer-readable medium includes any entity or device that can carry the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the computer-readable medium includes but is not limited to the above-mentioned components.

[0037] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0038] For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances. When an element is referred to as being "assembled on", "installed on", "fixed on" or "set on" another element, it can be directly on the other element or there can also be a centered element. When an element is considered to be "connected" to another element, it can be directly connected to the other element or there may be a centered element at the same time. The terms "vertical", "horizontal", "up", "down", "left", "right" and similar expressions used herein are only for illustrative purposes and are not intended to be the only implementation method.

[0039] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

[0040] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

Claims

1. A laser radar-based tracking oversized cargo transport vehicle positioning method is used to estimate the relative posture of the leading vehicle and the trailing vehicle that are collaboratively carrying oversized cargo. The method is implemented by using a leading vehicle positioning module disposed at the front end of the leading vehicle, a trailing vehicle positioning module disposed at the rear end of the trailing vehicle, and a feature marker disposed at the front end of the supporting device. The method is characterized in that: The following steps are involved: Establish the coordinate system of the front vehicle, the rear vehicle, the cargo, and the positioning module, calibrate the coordinate system of the front vehicle, the rear vehicle, the cargo, and the positioning module, and store the calibration results; Based on the laser radar, the point cloud information of the surrounding environment of the front and rear vehicles is obtained, and the point cloud maps of the front and rear vehicles are constructed. The self-positioning of the front or rear vehicle is completed based on the laser radar inertial odometer in the positioning module, and then the rear vehicle sends the IMU measurement value and posture to the front vehicle; The relative posture between the front vehicle and the cargo is obtained based on the front vehicle positioning module and the feature markers on the support device. The Kalman filter is used to estimate the cargo posture in real time based on the kinematic equation between the vehicle and the cargo. Multi-vehicle collaborative positioning is achieved based on real-time estimation of cargo posture. When the posture change of the rear vehicle reaches the Nth key frame, the point cloud and the posture of the rear vehicle at this moment are transmitted to the front vehicle. The positioning of the rear vehicle and the cargo is corrected through the factor graph optimization method, and the optimized rear vehicle state is transmitted to the point cloud map of the rear vehicle to correct the map information.

2. The laser radar-based tracking oversized transport vehicle positioning method according to claim 1 is characterized in that: Establish the coordinate system of the front vehicle, rear vehicle, cargo, and positioning module, and calibrate the coordinate system of the front vehicle, rear vehicle, cargo, and positioning module, as follows: (21) Positioning module calibration Get IMU coordinate system For the positioning module coordinate system, the mobile positioning module, the laser radar uses the ICP point cloud registration method to estimate the laser radar coordinate system The IMU uses the pre-integration method to estimate the posture change, and the hand-eye calibration method is used to determine the coordinates of each lidar. Coordinate calibration extrinsic parameters , where the calibration external parameter By rotation matrix and displacement vector composition: ; (22) Calibration of positioning module and vehicle coordinate system Take the support position of the front or rear vehicle as the origin and establish the vehicle coordinate system of the front or rear vehicle , take the forward direction of the front or rear vehicle as the positive direction of the X axis, the vertical upward direction of the front or rear vehicle as the positive direction of the Z axis, temporarily arrange RTK at the support position of the front or rear vehicle, move the front or rear vehicle, obtain the position change of the positioning module and the front or rear vehicle respectively, and obtain the calibration external parameters from the positioning module coordinates to the front or rear vehicle coordinates through the hand-eye calibration method ; (23) Calibration of the front vehicle coordinate system and the rear vehicle coordinate system After the front vehicle arrives at the loading location, a point cloud environment map around the loading location is constructed based on the LiDAR inertial odometer, and the accuracy of the point cloud map is improved through loop correction. While the front vehicle and the rear vehicle are stationary waiting to load the oversized cargo, the rear vehicle will The laser radar point cloud under the The established point cloud map is used for point cloud registration to obtain the initial calibration external parameters from the rear vehicle coordinates to the front vehicle coordinates , and take the position of the front vehicle coordinate system at this time as the world coordinate system origin; (24) Calibration of vehicle coordinate system and cargo coordinate system The direction of the cargo from the rear vehicle support point to the front vehicle support point is taken as the cargo posture in the world coordinate system. During transportation, the vehicle and the cargo are articulated. According to the calibrated external parameters in step (23), The distance between the two supporting points of the corresponding goods can be directly obtained and the initial heading angle in the world coordinate system ; (25) Feature identification and cargo coordinate system calibration The laser radar in the front vehicle positioning module recognizes the feature mark on the support device, filters the point cloud based on the laser echo intensity, obtains the feature mark and calculates the plane normal vector , based on the calibration external parameters of the positioning module coordinates to the world coordinate system , observe the laser radar coordinates of the goods to Coordinate calibration extrinsic parameters Get the normal vector Representation in world coordinate system ,calculate Projection on the XOY plane of the world coordinate system and cargo heading angle Angle .

3. The laser radar-based tracking oversized transport vehicle positioning method according to claim 1 is characterized in that: Based on the laser radar, the point cloud information of the surrounding environment of the front and rear vehicles is obtained, and the point cloud maps of the front and rear vehicles are constructed. The self-positioning of the front or rear vehicle is completed based on the laser radar inertial odometer in the positioning module, as follows: (31) Use laser radar to obtain point cloud information of the surrounding environment, use Gaussian filtering to smooth the point cloud data, use incremental KD tree as the data structure of the point cloud space, and establish a point cloud map; Gaussian filtering: multiply the k nearest neighbors of each point by the Gaussian weight G, and then perform weighted average to obtain the filtered point cloud coordinate value : In the formula, is the distance from the point cloud to the k nearest neighbor points, is the standard deviation of the Gaussian function, is the current point coordinate, For the neighborhood The coordinates of the points; (32) Combined with the calibrated external parameters, the IMU uses linear interpolation displacement and spherical interpolation attitude to track the movement of the lidar coordinate system, align all point clouds to the IMU coordinate system at the same time, and eliminate point cloud motion distortion; (33) Based on the point cloud map established in real time, the point-to-surface distance is used as the observation error, and the positioning information of the leading and trailing vehicles is updated in real time through the error state Kalman filter. and ; The discrete update equation for the positioning of the leading or following vehicle is: In the formula, express State quantity at time , express State quantity at time , express Observable quantity at time , express Noise of the moment ; State quantity , Observation ,noise and function f They are: Represents the rotation matrix, translation vector, and velocity vector of the leading or trailing vehicle in the world coordinate system. Represent the zero bias of the accelerometer and gyroscope respectively, Represents the sampling time interval of IMU, Represents acceleration and angular velocity, "~" represents the IMU observation, represents the matrix transpose, assuming zero bias , Satisfies the Wiener process, noise The components of are all independent zero-mean Gaussian white noise, symbol Used to parameterize the state error on an n-dimensional manifold The specific operation rules are as follows: is the exponential map in the three-dimensional rotation group SO(3); Defining the observation error of point-to-surface distance for: In the formula, Represents the rigid transformation matrix from the front or rear vehicle's ego coordinate system to the world coordinate system, Represents the point cloud coordinates obtained in the vehicle coordinate system of the front or rear vehicle. Including the front car Or the rear car , Point Cloud The nearest neighbor point in the point cloud map, for The normal vector of the corresponding plane; (34) The key frame change threshold is set to 0.4 m displacement or 10° heading angle or 0.5 s. The rear vehicle sends the positioning result and IMU measurement value to the front vehicle at the point cloud key frame.

4. The laser radar-based tracking oversized transport vehicle positioning method according to claim 1 is characterized in that: Based on the observation of the front vehicle positioning module and the feature identification on the support device, the relative posture between the front vehicle and the cargo is obtained. Based on the kinematic equation between the vehicle and the cargo, the Kalman filter is used to estimate the cargo posture in real time, as follows: (41) According to the kinematic relationship between vehicle and cargo, the cargo posture prediction equation (1) and the corresponding discretization equation (2) are obtained: （1） （2） In the formula, , , , , , , , Represent the lateral velocity, longitudinal velocity, yaw rate and heading angle of the vehicle coordinate system respectively. , , , The subscripts 1 and 3 represent the front and rear vehicles respectively. and Respectively represent the distances from the front and rear vehicle coordinate systems to their respective support points, is a discrete time interval, is the process noise, and its variance is assumed to be Zero-mean Gaussian white noise, a and b are the representations of the formula derivation results; (42) Control the laser radar of the front vehicle positioning module to obtain the normal vector of the feature identifier of the support device in the front vehicle coordinate system , combined with the positioning results of the positioning module to obtain the heading angle measurement value of the cargo in the world coordinate system , so we can get the observation equation (3): （3） In the formula, is the measurement noise, and is assumed to be the variance Zero-mean Gaussian white noise; (43) Predict the discretized equation (2) and the observation equation (3), and use Kalman filtering to update the cargo heading angle in real time .

5. The laser radar-based tracking oversized cargo transport vehicle positioning method according to claim 1, characterized in that: The multi-vehicle collaborative positioning is realized based on the real-time estimation of the cargo posture. When the posture change of the rear vehicle reaches the Nth key frame, the point cloud and the posture of the rear vehicle at this moment are transmitted to the front vehicle. The positioning of the rear vehicle and the cargo is corrected by the factor graph optimization method, as follows: (51) When the estimated posture of the rear vehicle reaches When a key frame changes the threshold, the point cloud information at this moment is sent to the preceding vehicle; (52) Definition State variables to be optimized for multi-vehicle transportation system with time-to-time tracking for: （4） In the formula , represent The position of the front and rear vehicles in the world coordinate system at the moment, represent The state of the front vehicle IMU in the front vehicle coordinate system at the moment, including the obtained acceleration , angular velocity , accelerometer and angular velocity meter Zero bias, represent The state of the rear vehicle IMU in the rear vehicle coordinate system at time t, including the obtained acceleration , angular velocity , Accelerometer bias and the zero bias of the angular velocity sensor ; M is the number of key frames of the rear vehicle posture involved in a factor graph optimization; (53) IMU is realized by pre-integration method Time to The estimation of vehicle posture change at the moment is as follows: （5） In the formula and are the discretized zero-mean Gaussian white noise, and are the zero bias of the accelerometer and gyroscope at time k, Represents a state quantity from i Time has come j The change at each moment is based on the pre-integrated estimation result of the IMU measurement value to construct the state residual: （6） In the formula, is gravity, for The inverse mapping of (54) The point cloud map of the rear vehicle and the point cloud map of the front vehicle is constructed with the point-to-surface distance as the error. : （7） In the formula Point cloud for the rear vehicle The nearest neighbor point in the point cloud map, for The normal vector of the corresponding plane; (55) Based on cargo heading angle The prediction and observation equations are constructed to construct the residuals and: （8） （9） (56) Based on the construction residual, the sliding window algorithm and factor graph optimization are used to solve the optimal , to achieve multi-vehicle collaborative positioning, as follows: （10） In the formula, represents the Huber norm, represents the Mahalanobis distance, Represents the IMU measurement residual, Represents the residual of the rear vehicle point cloud, Represents the residual error of cargo prediction posture, Represents the residual of cargo observation posture; (57) The corrected position of the rear vehicle is sent to the front vehicle, and the positioning status and point cloud map of the rear vehicle are updated.

6. A laser radar-based follow-up oversized cargo transport vehicle positioning system, used to implement the laser radar-based follow-up oversized cargo transport vehicle positioning method described in any one of claims 1 to 5, characterized in that: include: A calibration unit is used to establish a coordinate system where the front vehicle, the rear vehicle, the cargo, and the positioning module are located, calibrate the coordinate system where the front vehicle, the rear vehicle, the cargo, and the positioning module are located, and store the calibration results; A construction unit is used to obtain point cloud information of the surrounding environment of the front vehicle and the rear vehicle based on the laser radar, construct point cloud maps of the front vehicle and the rear vehicle respectively, and complete the self-vehicle positioning of the front vehicle or the rear vehicle based on the laser radar inertial odometer in the positioning module, and then enable the rear vehicle to send the IMU measurement value and posture to the front vehicle; The cargo posture estimation unit is used to obtain the relative posture between the leading vehicle and the cargo based on the leading vehicle positioning module and the characteristic identification on the supporting device, and to estimate the cargo posture in real time using Kalman filtering based on the kinematic equation between the vehicle and the cargo; The multi-vehicle collaborative positioning unit is used to realize multi-vehicle collaborative positioning based on the real-time estimation of the cargo posture. When the posture change of the rear vehicle reaches the Nth key frame, the point cloud and the posture of the rear vehicle at this moment are transmitted to the front vehicle. The positioning of the rear vehicle and the cargo is corrected through the factor graph optimization method, and the optimized rear vehicle state is transmitted to the point cloud map of the rear vehicle to correct the map information.

7. The laser radar-based tracking oversized cargo transport vehicle positioning system according to claim 6 is characterized in that: The calibration unit includes a front vehicle positioning module arranged at the front end of the front vehicle and a rear vehicle positioning module arranged at the rear end of the rear vehicle; The front vehicle positioning module includes three laser radars that perceive surrounding space information, a laser radar that observes the cargo posture, a front vehicle positioning calculation module, an IMU and a communication module. The laser wavelength of the laser radar that observes the cargo posture is 905nm; The rear vehicle positioning module includes two laser radars for sensing surrounding space information, a rear vehicle positioning calculation module, and an IMU and communication module.

8. The laser radar-based tracking oversized cargo transport vehicle positioning system according to claim 7 is characterized in that: The cargo posture estimation unit comprises a supporting device and a characteristic mark arranged on the front end surface of the supporting device, wherein the characteristic mark is arranged as a reflective tape.

9. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that: The memory stores a computer program that can be run on the processor. When the processor loads and executes the computer program, the laser radar-based following oversized transport vehicle positioning method described in any one of claims 1 to 5 is adopted.

10. A storage medium containing computer executable instructions, characterized in that: The computer executable instructions are used to execute the laser radar-based following oversized transport vehicle positioning method as described in any one of claims 1 to 5 when executed by a computer processor.

Images