A kind of intelligent robot and system for road geometry parameter detection acceptance

Abstract

The application relates to the technical field of roadbed and pavement detection equipment, and discloses an intelligent robot and system for detecting and accepting road geometric parameters, which comprises a holder measurement module, an RTK positioning module, a chassis, an RTK base station and a total station instrument erected on a to-be-detected area of a to-be-detected road; the holder measurement module and the RTK positioning module are both installed on the chassis; the chassis moves in an intermittent mode of walking for a period and then stopping on the to-be-detected road; the RTK positioning module is electrically connected with the RTK base station and is used for providing global coordinates of the chassis in real time; the holder measurement module is used for detecting the roadbed and pavement of the to-be-detected road when the chassis stops moving, and is electrically connected with the total station instrument; and the total station instrument is used for positioning the coordinates of the holder measurement module. The robot and system realize accurate and efficient collection and analysis of road surface data, can accurately evaluate road quality, and provide reliable technical support for road construction, maintenance and management.

Description

Technical Field

[0001] This invention relates to the field of roadbed and pavement testing equipment technology, and in particular to an intelligent robot and system for road geometric parameter testing and acceptance in foundation pit engineering, and its usage method. Background Technology

[0002] Quality monitoring of road infrastructure is crucial for traffic safety and transportation efficiency. Traditional pavement inspection methods, such as manual measurement and simple vehicle-mounted equipment, are generally inefficient, lack accuracy, and heavily reliant on human judgment, making it difficult to meet the urgent needs of modern road management for high-precision, three-dimensional geometric data.

[0003] In recent years, mobile LiDAR measurement systems have become the mainstream technology in road inspection. These systems rapidly acquire high-density road surface point cloud data via vehicle-mounted systems to extract road surface defects, smoothness, and geometric parameters. However, existing vehicle-mounted mobile measurement systems mainly employ continuous dynamic scanning modes, heavily relying on Global Navigation Satellite Systems (GNSS, such as RTK) and Inertial Measurement Units (IMUs) for positioning and attitude calculation. In practical applications, especially in areas with limited GNSS signals, such as urban canyons, under viaducts, or tunnels, IMU errors accumulate over time, leading to yaw drift and decreased attitude calculation accuracy, directly affecting the absolute coordinate accuracy of the point cloud data and the reliability of the final road surface geometric parameters. Furthermore, for road structure measurements requiring centimeter-level or even millimeter-level absolute accuracy, the error models of existing dynamic systems are complex and difficult to completely eliminate.

[0004] To overcome positioning and attitude errors in dynamic measurements, some high-precision measurement solutions have begun to explore the use of external high-precision measuring equipment (such as total stations) for assistance. However, in existing technologies, total stations are mostly used for auxiliary calibration or as a means of inspection, lacking an efficient and systematic method to utilize the extremely high precision of total stations to directly perform high-precision calibration and real-time correction of the geometric center and key attitude parameters of lidar in static or quasi-static modes. Specifically, existing systems struggle to quickly and accurately determine the rotation center of the lidar and the precise geometric relationship (lever value) between the prism and the lidar center on-site, and cannot use external measurement results to eliminate the inherent zero-drift error of the IMU before measurement. This results in systematic biases in the subsequent attitude calculation of the IMU. Furthermore, although static measurements are highly accurate, the lack of a reliable and high-precision point cloud stitching mechanism after segmented static measurements makes it difficult to guarantee data continuity and overall coordinate consistency. Therefore, a solution is urgently needed. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide an intelligent robot and system for road geometric parameter detection and acceptance. The robot and system can combine high-precision hardware (such as high-precision gimbal and 2D radar) with innovative working mode (i.e. segmented static measurement), and introduce high-precision centering and heading angle calibration technology of total station bidirectional measurement prism coordinates to completely eliminate the cumulative error and IMU zero drift in dynamic measurement, achieve absolute high precision and reliability of road surface data acquisition, and ensure seamless high-precision splicing between segmented data.

[0006] To achieve the aforementioned objective, the technical solution of the present invention is implemented as follows: an intelligent robot and system for road geometric parameter detection and acceptance, comprising an RTK positioning module, a chassis, an RTK base station and a total station mounted on the area to be measured on the road to be measured; both the gimbal measurement module and the RTK positioning module are mounted on the chassis; the chassis moves on the road to be inspected using an intermittent movement method of moving a section and then stopping; the RTK positioning module is electrically connected to the RTK base station (i.e., connected via a wireless protocol) to provide the global coordinates of the chassis on the road to be inspected in real time; the gimbal measurement module is used to detect the roadbed and pavement of the road to be inspected when the chassis stops moving, and is electrically connected to the total station (i.e., connected via a 4G signal); the total station is used to lock the real-time global coordinates of the gimbal measurement module on the road to be inspected.

[0007] Preferably, the gimbal measurement module includes a prism, a lidar, a laser pointer, a cam mechanism, an IMU, a gimbal, a servo motor, and a base;

[0008] The base is mounted on the chassis; the gimbal is mounted above the base; the lidar, cam mechanism, and IMU are all mounted above the gimbal, and the cam mechanism is connected to the laser pointer; the prism is mounted above the lidar; the servo motor is mounted below the gimbal and is used to drive the prism, lidar, laser pointer, cam mechanism, IMU, and gimbal to rotate synchronously; and the encoder of the servo motor records the rotation angle of the gimbal.

[0009] The lidar is used to perform static scanning of the road surface and roadbed of the road to be inspected; the cam mechanism is used to adjust the height of the laser pointer; the IMU is used to measure the yaw angle and roll angle of the gimbal; and the total station is used to measure the real-time coordinates of the prism.

[0010] Preferably, the real-time coordinates of the prism and the rotation angle of the gimbal are used to calculate the coordinates of the lidar, and the coordinates of the prism are used to calculate the initial heading angle of the robot chassis. The initial heading angle Used to correct and eliminate yaw angle zero drift of the IMU integrated on the gimbal, to obtain the robot's pose information on the road to be inspected.

[0011] Preferably, the control system of the intelligent robot for road geometric parameter detection and acceptance described above is characterized by including a cloud-based digital platform and an APP control terminal; the cloud-based digital platform is used to interact with the robot and the APP control terminal, and after processing the data, to perform three-dimensional reconstruction, geometric parameter calculation and visualization management of the road to be inspected.

[0012] Preferably, the cloud-based digital platform includes an IoT communication module and a cleaning module; the IoT communication module is used to monitor and receive in real time the operation data, robot task information, and robot raw data uploaded by the robot end and the APP control end using the MQTT protocol, and transmit the received data to the cleaning module for preprocessing to form multimodal data.

[0013] Preferably, the robot's raw data includes the robot's global pose (pose information) in a static position on the road to be inspected; the operational data includes point cloud data containing roadbed and pavement detection information obtained by the lidar after segmenting the road to be inspected; and the angle data measured by the IMU and servo motor when measuring each point cloud data.

[0014] Preferably, the cloud-based digital platform includes a data processing service module; the data processing service module receives multimodal data, and splices the point cloud data collected by the robot in segments from the multimodal data with the pose information corrected by the total station, and performs three-dimensional reconstruction and geometric parameter calculation to form a three-dimensional model of the road surface of the road to be inspected.

[0015] Preferably, the cloud-based digital platform includes a map management service module; the map management service module is used to display map data of the road to be measured, process CAD drawings of the road to be measured, import line element tables, and render models.

[0016] Preferably, the cloud-based digital platform includes a job monitoring service module; the job monitoring service module is used to track and record the robot's job tasks, measurement results, and current status in real time.

[0017] Preferably, the cloud-based digital platform includes a system management service module; the management service module is installed on the APP control terminal and is used to manage user permissions, information, and robots.

[0018] The beneficial effects of this invention are:

[0019] (1) The robot and system provided by this invention achieve extremely high absolute accuracy of road point cloud data when detecting the road surface, eliminating the dynamic cumulative error of the robot during the scanning and detection process of the road surface. Existing vehicle-mounted systems adopt a continuous dynamic measurement mode, which inevitably produces cumulative errors and instantaneous drift in GNSS / IMU fusion. However, the robot of this invention adopts a segmented static measurement mode of "walking a section and stopping for static scanning". At each acquisition station, data is acquired through a static high-precision 2D lidar and a high-precision gimbal, which fundamentally eliminates the influence of the robot's dynamic movement and attitude changes on the point cloud accuracy and ensures the absolute accuracy of the acquired data. High-precision stitching ensures overall accuracy: After static measurement, multi-sensor fusion is performed using the global coordinates provided by the RTK positioning module (i.e., high-precision GNSS receiver) and the precise prism coordinates measured by the total station. This ensures that the pose of each measurement station of the robot reaches extremely high accuracy, thereby ensuring the overall coordinate accuracy and geometric continuity of all segmented point cloud data after stitching.

[0020] (2) This invention innovatively solves the IMU yaw zero-drift problem and provides accurate yaw calibration: by measuring the prism coordinates twice consecutively with a total station (e.g., rotating the gimbal forward and backward), the local coordinates of the radar center are accurately calculated, and the robot's initial yaw angle is accurately calculated accordingly. This method provides an external, high-precision yaw reference independent of the IMU itself. Eliminating IMU zero drift: by fusing the accurate yaw angle measured by the total station with the yaw angle provided by the vehicle-mounted IMU, the systematic zero-drift error of the IMU is effectively corrected and eliminated. This allows the IMU to provide more reliable high-frequency attitude data for the robot in subsequent attitude fusion, greatly improving the reliability of the final pose calculation.

[0021] (3) The present invention improves the efficiency of geometric center calibration and data acquisition, and can quickly and accurately calibrate: By using a total station to measure the prism in two directions, the local positioning of the radar center and the determination of the prism arm value are completed quickly and accurately on site, eliminating the traditional complex target field calibration process.

[0022] (4) The robot and system provided by the present invention can efficiently collect data with a wide field of view: the combination of high-precision gimbal and 2D lidar can quickly scan a wide road area during static parking, and realize rapid segmented collection of road data while ensuring accuracy.

[0023] (5) The present invention provides high-quality data and reliable calculation results for road detection: The point cloud data obtained through static measurement and high-precision pose calculation has extremely high purity and density. The accuracy and reliability of the road surface plane geometry, vertical geometry and surface characteristic indicators calculated using this data are far higher than those obtained using traditional dynamic measurement systems, providing a more scientific and reliable basis for road maintenance and management decisions. Attached Figure Description

[0024] Figure 1 This is a three-dimensional structural diagram of the present invention;

[0025] Figure 2 This is a schematic diagram of the measurement system structure of the present invention;

[0026] Figure 3 This is a hardware control architecture diagram of the present invention;

[0027] Figure 4 This is a flowchart of the process of the present invention;

[0028] Figure 5 This is a flowchart illustrating the measurement mechanism of the present invention.

[0029] Figure 6 This is a diagram showing the solution for the heading angle in this invention;

[0030] Figure 7 This is a scanning measurement working diagram of the present invention;

[0031] Figure 8 This is the measurement point cloud diagram of the present invention;

[0032] Figure 9 This is an enlarged view of the measurement point cloud of the present invention;

[0033] Figure 10 This is a system workflow diagram of the present invention;

[0034] Figure 11 This is a cloud deployment architecture diagram of the present invention;

[0035] Figure 12 This is the framework of the robot control system of the present invention;

[0036] Figure 13 This invention provides a robot application software framework.

[0037] Figure 14 This is the interface of the robot application software of the present invention.

[0038] Figure labels and descriptions:

[0039] 1. Gimbal Measurement Module; 101. Prism; 102. LiDAR; 103. Laser Pointer; 104. Cam Mechanism; 105. IMU; 106. Gimbal; 107. Servo Motor; 108. Base; 2. RTK Positioning Module; 3. Housing; 4. Chassis. Detailed Implementation

[0040] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0041] Example 1

[0042] See Figure 1-14 As shown: This invention provides an intelligent robot and system for road geometric parameter detection and acceptance, aiming to overcome the technical bottlenecks of existing vehicle-mounted lidar systems, such as error accumulation in dynamic attitude correction, insufficient wide field-of-view data acquisition efficiency, and low purity of point cloud data, thereby achieving high-precision, high-efficiency measurement and accurate calculation of road geometric indicators.

[0043] The intelligent robot provided by this invention mainly consists of a total station, an RTK base station, a gimbal measurement module 1, an RTK positioning module 2, a shell 3, and a chassis 4.

[0044] The gimbal measurement module 1 and the RTK positioning module 2 are mounted on the robot chassis 4. The RTK positioning module 2 is a high-precision GNSS receiver.

[0045] The RTK positioning module 2 is connected to the RTK base station via a wireless protocol to provide the global coordinates of the chassis 4 on the road to be inspected in real time. These coordinates are used to guide the robot to move on the road to be inspected, that is, to guide the robot to move from one static inspection point (static station) on the road to be inspected to the next static inspection point.

[0046] The gimbal measurement module 1 is the core integrated unit of the robot, internally integrating key components such as a prism 101, a lidar 102, a laser pointer 103, a cam mechanism 104, an IMU 105, a gimbal 106, a servo motor 107, and a base 108. All measurement components are integrated and fixed on the chassis 4 and protected by the outer shell 3. The IMU 105 is an inertial measurement sensor.

[0047] The aforementioned intelligent robot system includes a cloud-based digital platform and an APP control terminal.

[0048] The cloud-based digital platform is built on high-performance servers and adopts a microservice architecture design, aiming to achieve full lifecycle management of the data measured by the robot on the road to be inspected.

[0049] In practical applications, the cloud-based digital platform serves as the data processing, management, and display center for the system of this invention. It is used to receive high-precision segmented point cloud and pose data collected by the robot, and to perform 3D reconstruction, geometric parameter calculation, and visualization management to achieve digital acceptance of road engineering projects.

[0050] The cloud-based digital platform is deployed on a high-performance x86_64 architecture server, managed using a Docker containerized environment, and based on the CentOS operating system. The system adopts an advanced microservice architecture and interacts with external devices (such as robots and APP control terminals) through an IoT communication module and a Kafka message queue.

[0051] The cloud-based digital platform includes: an IoT communication module, a cleaning module, a data processing service module, a map management service module, an operation monitoring service module, and a system management service module.

[0052] In practical applications, the cloud-based digital platform, through the IoT communication module, can monitor and receive in real time the operation data and task information uploaded by the robot and the APP control terminal using the MQTT protocol, as well as the raw data from high-precision sensors (including LiDAR 102, total station, and RTK positioning module 2). All received data is then distributed to the cleaning module for preprocessing to form multimodal data through the message middleware.

[0053] The data processing service module is responsible for receiving the cleaned multimodal data and performing complex fusion and storage tasks. Crucially, this module combines the high-precision road point cloud data collected by the robot in segments with the accurate pose information corrected by the total station to perform 3D reconstruction and geometric parameter calculation, ultimately forming a unified, high-precision digital road model.

[0054] In addition, the map management service module undertakes key functions such as map data display, processing CAD drawings, importing line element tables, and model rendering; the operation monitoring service module tracks and records the robot's operation tasks, measurement results, and current status in real time; and the system management service module ensures the stable operation of basic functions such as user permissions, information, and robot management.

[0055] As the central hub of the robot system in this application, the cloud-based digital platform has achieved several key innovations, greatly improving the level of automation and digitalization.

[0056] During the preparation phase, the cloud-based digital platform has powerful automated operation path planning capabilities. By importing CAD drawings and parsing line element tables, it can automatically obtain road geometric elements and intelligently plan the robot's precise driving route and operation tasks, reducing manual intervention.

[0057] At the data application level, the cloud-based digital platform enables the construction of a digital twin platform model: by performing in-depth processing and 3D rendering of high-precision point cloud data, highly realistic digital and virtual roads are constructed, thereby realizing a digital twin model of the roadbed and pavement geometric parameters. This digital twin model can reflect the road's geometric state in real time and provide powerful decision support for the entire lifecycle management of roads, including design, construction, and maintenance.

[0058] Furthermore, the cloud-based digital platform significantly improves the efficiency and accuracy of the robot's presentation of road inspection results, achieving high-precision data visualization and automated reporting; the calculated measurement results are intuitively displayed through 3D rendering. Simultaneously, the platform integrates powerful data reporting functions, automatically outputting seven key geometric parameters of the road under inspection, including flatness, thickness, and centerline deviation, in report form, greatly improving the efficiency of road acceptance inspection and the accuracy of data processing.

[0059] Finally, the robot system of this application constructs a distributed data storage system by using various types of databases such as Redis, MySQL, ClickHouse, and TDengine, thereby effectively addressing the high-performance challenges of storage, querying, and analysis brought about by massive multimodal data.

[0060] The workflow of the aforementioned intelligent robot system strictly follows three closely connected stages: preparation and high-precision geometric calibration, segmented static acquisition, and multi-sensor fusion calculation, to ensure the absolute measurement accuracy of the final roadbed and pavement point cloud data.

[0061] 1) In the preparation and high-precision static calibration stage:

[0062] 11) First, install and initialize the system, firmly fix the gimbal measurement module 1 and RTK positioning module 2 to the robot chassis 4, and protect them with the outer shell 3. At the same time, set up the RTK base station and total station in the measurement area, and accurately establish the global geodetic coordinate system of the robot on the road to be inspected through control point measurement.

[0063] 12) Perform core geometric center calibration: that is, start the servo motor 107 to drive the gimbal 106 to accurately return to the zero position.

[0064] This step includes geometric center alignment detection: The laser pointer 103 is activated, and its height is controlled by the cam mechanism 104. The gimbal 106 is then slowly rotated 360 degrees around its rotation axis. Utilizing the high-precision ranging function of the lidar 102, the diameter of the circle formed by the laser pointer 103's projection point is checked to see if it is within the allowable tolerance range. The point cloud image obtained by the lidar 102 is used to further verify the diameter of the circle formed by the laser pointer 103's projection point. By comparing the diameter measured on the point cloud image with the actual measured diameter of the laser point formed by the laser pointer 103, the geometric coincidence between the center of the lidar 102 and the rotation axis of the gimbal can be verified with high precision. Since the road information obtained when the lidar 102 is at the exact center of the gimbal 106 is the most accurate, the higher the geometric coincidence, the higher the accuracy of the detection equipment.

[0065] 13) Perform precise center calibration of prism 101 and lidar 102 to obtain an initial heading angle without zero drift:

[0066] like Figure 6 As shown: First, the total station is activated and prism 101 is locked. Servo motor 107 drives the gimbal 106 to precisely rotate 180 degrees clockwise from zero position. The total station measures and records the coordinates of point B on prism 101. Then, the gimbal 106 precisely rotates 360 degrees counterclockwise, and the total station measures and records the coordinates of point C on prism 101. Finally, based on the two precisely symmetrical points B and C, the coordinates of the midpoint O are calculated. This O point is the precise position of the center of the lidar 102 in the local coordinate system. Using the coordinate relationship between O and B (or C), the precise initial heading angle of the robot chassis 4 relative to the global coordinate system is geometrically calculated. .

[0067] like Figure 6 As shown: This precise initial heading angle This calibration will be used in subsequent multi-sensor fusion to correct and eliminate yaw angle zero drift of the IMU 105 integrated on the gimbal 106. After calibration, the gimbal 106 is reset to zero, and the system enters the data acquisition waiting state.

[0068] 2) Segmented static measurement and data acquisition stage:

[0069] 21) The system controlling the robot's operation adopts a unique working mode of segmented static measurement to ensure that the chassis 4 is completely stationary at the predetermined position during data acquisition. The system then starts each sensor to work synchronously.

[0070] 22) At each static station, the system will acquire the robot's high-precision static pose:

[0071] The RTK positioning module 2 provides the global absolute coordinates of the robot's chassis 4 in real time; at the same time, the total station continuously locks onto and measures the real-time global coordinates of the prism 101. The coordinates of the prism 101 are combined with the geometric relationship obtained in step 1), i.e., the calibration stage, to accurately calculate the robot's absolute pose information at this static position (i.e., the position and attitude of the robot when it performs detection at this point).

[0072] 23) When the signal of RTK positioning module 2 is limited due to obstruction or other reasons, the system can use the extremely high-precision coordinates obtained by the total station measurement as the global coordinates of the robot through coordinate transformation, thus ensuring the continuity and reliability of positioning.

[0073] The multi-sensor fusion module then fuses the global coordinates of the high-precision GNSS receiver, the precise pose of the total station, and the zero-drift heading angle acquired during the calibration phase. First, the timestamps of the high-precision GNSS receiver and the total station need to be synchronized to ensure time synchronization. The system monitors the positioning quality of the high-precision GNSS receiver in real time.

[0074] When the signal is good, the centimeter-level global coordinates provided by the high-precision GNSS receiver are used as the position reference.

[0075] When the high-precision GNSS receiver is blocked, the system seamlessly switches to using millimeter-level coordinates measured by a total station as the global position reference through coordinate transformation, ensuring the continuity of positioning.

[0076] The system uses the zero-drift initial heading angle obtained during the calibration phase using a total station's dual-coordinate method as an external reference value. Through a Kalman filter fusion algorithm, this precise heading angle is used to correct the yaw angle drift of the IMU 105 in real time, completely eliminating the accumulated system error of the IMU 105. This outputs a high-precision, high-reliability global robot pose at this station position.

[0077] 24) After the robot has achieved global pose, the point cloud data acquisition stage of the roadbed and road surface is entered: the servo motor 107 drives the gimbal 106 to rotate precisely, which drives the lidar 102 to start scanning the roadbed and road surface of the road to be measured; the IMU 105 measures and outputs the yaw angle and roll angle of the gimbal 106 in real time; the encoder of the servo motor 107 accurately returns the rotation angle of the gimbal 106.

[0078] 25) During the rotation of the gimbal 106, the lidar 102 scans and collects the detection information of the roadbed and pavement as point cloud data with high precision. Each point cloud data is recorded synchronously with the angle data of the IMU 105 and the servo motor 107, as well as a unified timestamp (that is, the timestamps of the lidar 102 data and the servo motor 107 data are aligned), providing a basis for subsequent high-precision calculation.

[0079] 26) After the data collection at the current station is completed, the robot moves to the next predetermined position and repeats the above static measurement steps until the measurement task of the entire road section is completed.

[0080] 3) Data post-processing and point cloud stitching analysis stage:

[0081] This stage utilizes the multi-source high-precision data collected in step 2) above to achieve refined resolution, seamless stitching, and automated parameter calculation of road surface point cloud data.

[0082] 31) In the point cloud data refinement and stitching stage, the coordinate transformation within a single station (i.e., the robot at each static station) is first performed: using the roll / yaw angle provided by IMU105 and the rotation angle data provided by servo motor 107, the original point cloud data of the road to be detected measured by LiDAR 102 is transformed into $P_P(x_P, y_P, z_P)$ in the center coordinate system of gimbal 106 through the rotation matrix R and translation matrix T in the homogeneous coordinate transformation matrix. The formula is shown below;

[0083]

[0084] The precise transformation from the coordinate system of LiDAR 102 to the coordinate system of the center of the gimbal 106 is significant because it offsets the displacement changes of LiDAR 102 during the rotation of the gimbal 106 and corrects the tilted scanning plane affected by the road surface inclination to a horizontal reference plane. Since the robot chassis 4 is completely stationary in its static position, the high-precision angle feedback from the motor encoder ensures that each laser measurement point is accurately reconstructed in a local 3D space with the center of the gimbal 106 as the origin. This lays a high-precision foundation for subsequently projecting the point clouds of multiple stations onto the geodetic coordinate system using global pose (RTK / total station data).

[0085] 32) Subsequently, global coordinate registration and stitching are performed: Utilizing the high-precision global pose obtained in the second stage (including the high-precision heading angle after eliminating zero drift), the point cloud data at each static station is accurately transformed and projected onto the global geodetic coordinate system. The projection process involves another coordinate transformation, using the same formula as before. Then, the matching degree of the poses at each station is verified using the point clouds of the overlapping road surface areas between adjacent stations. The Non-rigid Point Cloud Registration with Neural Deformation Pyramid (NDP) algorithm is used for point cloud registration. Once the overlapping road surface area point clouds are registered, the point cloud stitching is complete.

[0086] Through precise inter-station pose connection, seamless high-precision stitching of segmented static point cloud data was successfully achieved, forming a complete and high-precision three-dimensional road surface model.

[0087] 33) In terms of pavement index calculation and analysis, the system analyzes refined and filtered clean pavement point cloud data to accurately calculate the road's planar geometric indices (such as centerline and curve elements), vertical geometric indices (such as longitudinal slope and vertical curves), and surface characteristic indices (such as pavement smoothness IRI and rut depth). These calculation results provide a reliable scientific basis for road quality inspection and acceptance.

[0088] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An intelligent robot for road geometry parameter detection and acceptance, characterized in that, The utility model provides a kind of road detection device, including gimbal measurement module (1), RTK positioning module (2), chassis (4), RTK base station and total station instrument erected in the to-be-measured area of to-be-measured road;The gimbal measurement module (1) and RTK positioning module (2) are both installed on chassis (4);The chassis (4) moves using intermittent moving mode of walking a section and then stopping down on the to-be-detected road;The RTK positioning module (2) is electrically connected with RTK base station, for providing the global coordinate of chassis (4) on the to-be-detected road in real time;The gimbal measurement module (1) is used to detect the roadbed pavement of to-be-detected road when chassis (4) stops moving, and is electrically connected with total station instrument;The total station instrument is used to lock the real-time global coordinate of gimbal measurement module (1) on to-be-detected road; The gimbal measurement module (1) includes prism (101), laser radar (102), laser pen (103), cam mechanism (104), IMU (105), gimbal (106), servo motor (107) and base (108); The base (108) is installed on chassis (4);The gimbal (106) is installed above base (108);The laser radar (102), cam mechanism (104) and IMU (105) are all installed above gimbal (106), and cam mechanism (104) is connected with laser pen (103);The prism (101) is installed above laser radar (102);The servo motor (107) is installed below gimbal (106), and is used to drive prism (101), laser radar (102), laser pen (103), cam mechanism (104), IMU (105) and gimbal (106) to rotate synchronously;And the encoder of servo motor (107) is used to record the rotation angle of gimbal (106); The laser radar (102) is used to scan the roadbed pavement of to-be-detected road statically;The cam mechanism (104) is used to adjust the height of laser pen (103);The IMU (105) is used to measure the yaw angle and roll angle of gimbal (106);The total station instrument is used to measure the real-time coordinate of prism (101) on to-be-detected road.

2. The intelligent robot for road geometry parameter detection and acceptance as claimed in claim 1, wherein, The real-time coordinates of the prism (101) and the rotation angle of the holder (106) are used to calculate the coordinates of the laser radar (102); the coordinates of the prism (101) are used to calculate the initial heading angle of the robot chassis (4) ; the initial heading angle is used to correct and eliminate the yaw angle zero drift of the IMU (105) integrated on the holder (106), and the pose information of the robot on the road to be detected is obtained.

3. The control system of an intelligent robot for road geometry parameter detection and acceptance according to any one of claims 1-2, characterized in that, The cloud digital platform includes IoT communication module and cleaning module;The IoT communication module is used to listen and receive job data, robot task information and original data of robot uploaded by robot end and APP control end using MQTT protocol in real time, and transmit the received data to cleaning module for preprocessing to form multi-modal data.

4. The control system of an intelligent robot for road geometry parameter detection and acceptance according to claim 3, characterized in that, The cloud digital platform includes IoT communication module and cleaning module;The IoT communication module is used to listen and receive job data, robot task information and original data of robot uploaded by robot end and APP control end using MQTT protocol in real time, and transmit the received data to cleaning module for preprocessing to form multi-modal data.

5. The control system of the intelligent robot for road geometry parameter detection and acceptance according to claim 4, wherein the raw data of the robot comprises a global pose of the robot in a static station on a road to be detected; the operation data comprises point cloud data containing roadbed and pavement detection information obtained by the laser radar (102) after segment scanning of the road to be detected; and angle data measured by the IMU (105) and the servo motor (107) when each point cloud data is measured.

6. The control system of an intelligent robot for road geometry parameter detection and acceptance according to claim 5, characterized in that, The cloud digital platform comprises a data processing service module; the data processing service module receives multi-modal data, splices point cloud data collected by the robot in segments and pose information corrected by the total station in the received multi-modal data, and performs three-dimensional reconstruction and geometry parameter calculation to form a pavement three-dimensional model of the road to be detected.

7. The control system of an intelligent robot for road geometry parameter detection and acceptance according to claim 6, characterized in that, The cloud digital platform comprises a map management service module; the map management service module is used for displaying map data of a road to be measured, processing CAD drawings of the road to be measured, importing line element table, and rendering a model.

8. The control system of an intelligent robot for road geometry parameter detection and acceptance according to claim 7, characterized in that, The cloud digital platform comprises a job monitoring service module; the job monitoring service module is used for real-time tracking and recording of a job task, a measurement result, and a current state of the robot.

9. The control system of an intelligent robot for road geometry parameter detection and acceptance according to claim 8, characterized in that, The cloud digital platform comprises a system management service module; the management service module is installed in an APP control end and is used for managing user rights, information, and the robot.

Images