Ice rink detection trolley based on UWB positioning and detection method thereof

Abstract

The invention discloses an ice rink detection trolley based on UWB positioning and a detection method thereof, and relates to the technical field of automatic detection robots. The intelligent trolley body realizes motion control through a path planning result, the vehicle-mounted NUC lower computer cooperates with the remote control upper computer to realize remote control and data reading, the UWB positioning module cooperates with the UWB base station to solve the current position, the laser radar sensor is used for obstacle detection and local path re-planning, the IMU attitude sensor acquires an attitude angle, and the vehicle-mounted NUC lower computer cooperates with the remote control upper computer to realize remote control and data reading. The temperature and humidity sensor, the illumination sensor and the infrared temperature sensor are used for respectively measuring environment temperature and humidity, illumination intensity above the ice surface and ice surface temperature, and the communication module is used for realizing real-time data transmission. Multiple sensors are integrated and carried on the basis of an intelligent trolley body, the automation degree is improved, manual intervention is reduced, the positioning precision and repeatability of a detection point location can be guaranteed, the dynamic obstacle avoidance capacity is achieved in combination with path planning, and one-time synchronous collection of multiple parameters of a single point location is achieved.

Description

Technical Field

[0001] This invention relates to the field of automatic inspection robot technology, specifically to an ice rink inspection vehicle based on UWB positioning and its inspection method. Background Technology

[0002] Large ice rinks require regular monitoring of key parameters such as ice surface temperature, air temperature and humidity, and light intensity to ensure the venue meets the requirements for sports competitions and daily use. Currently, the industry mainly relies on manual handheld devices for fixed-point testing, which has the following significant shortcomings: First, the testing efficiency is low, as manual inspections consume a lot of time and manpower, making it difficult to meet the rapid testing needs of large ice rinks; second, data consistency is poor, as differences in operating habits and measurement timing among different operators may lead to measurement errors, affecting the reliability of the test results; third, the testing process is complex, requiring the replacement of multiple sensors to measure different information, lacking an integrated, one-time testing solution, making the operation cumbersome and prone to missing testing points.

[0003] While existing technologies have developed solutions for environmental inspection using robots, their application in large indoor ice rinks still suffers from several shortcomings. Firstly, large indoor ice rinks cannot rely on satellite positioning and navigation. Ordinary wheeled robots are prone to slipping on ice, causing odometry-based positioning methods to fail. Inertial navigation systems accumulate significant errors after prolonged operation, and single UWB positioning accuracy is insufficient to meet the high-precision requirements of inspection points. Furthermore, the lack of its own attitude information prevents direct application to positioning and navigation. Secondly, the detection positions and conditions of various sensors differ, existing solutions lack integrated detection design, and the multiple loading and measurement processes are complex. Moreover, the robot's adaptability to ice surfaces and the effectiveness of multi-sensor fusion are poor, making it difficult to achieve stable and efficient ice rink inspection.

[0004] Therefore, there is an urgent need for an automated detection solution that combines high-precision positioning, multi-sensor fusion, and dedicated ice surface detection to overcome the shortcomings of existing technologies. Summary of the Invention

[0005] To address the shortcomings of the prior art, this invention provides an ice rink inspection vehicle and its inspection method based on UWB positioning. The vehicle integrates multiple sensors on its intelligent body, improving automation, reducing manual intervention, ensuring the positioning accuracy and repeatability of inspection points, and possessing dynamic obstacle avoidance capabilities combined with path planning, enabling simultaneous acquisition of multiple parameters at a single point.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] An ice rink inspection vehicle based on UWB positioning includes an intelligent vehicle body, a control system, a UWB positioning module, a lidar sensor, an IMU attitude sensor, a temperature and humidity sensor, an infrared temperature sensor, a light sensor, and a communication module.

[0008] The intelligent vehicle body is equipped with a low-level controller and anti-slip drive wheels and steering servo to adapt to ice surface movement scenarios. The low-level controller converts the path planning results into drive electrical signals to achieve motion control. A fixed column is placed on top of the intelligent vehicle body.

[0009] The control system includes an on-board NUC lower-level computer and a remote control upper-level computer. The on-board NUC lower-level computer is installed inside the intelligent vehicle and is used to receive instructions, calculate position and attitude, plan paths, and control sensor data acquisition. The remote control upper-level computer is used to send instructions, receive feedback data, and realize remote control and data reading.

[0010] The UWB positioning module is mounted on the surface of the intelligent vehicle. It works in conjunction with the UWB base stations deployed around the ice rink to calculate the current position of the intelligent vehicle and transmit it to the on-board NUC slave computer.

[0011] The lidar sensor is mounted on the surface of the intelligent vehicle and is used for obstacle detection and local path replanning.

[0012] The IMU attitude sensor is mounted on the surface of the intelligent vehicle body and is used to obtain the attitude angle of the intelligent vehicle body, providing complete attitude data for the on-board NUC lower computer.

[0013] The temperature and humidity sensor and the light sensor are both slidably locked and installed on the column. The infrared temperature sensor is mounted on the front end of the intelligent vehicle body. The three are used to measure the ambient temperature and humidity, the lighting intensity above the ice surface and the ice surface temperature, respectively.

[0014] The communication module is mounted on the rear end of the intelligent vehicle, enabling real-time data transmission between the on-board NUC lower-level computer and the remote control upper-level computer.

[0015] Furthermore, the intelligent vehicle body adopts an Ackerman steering chassis with coaxial swing suspension, with a minimum turning radius of 1.5m and a maximum travel speed of 1.3m / s. The underlying controller is an STM32 controller, which receives path instructions from the on-board NUC lower-level machine through the CAN bus and outputs PWM signals to control the anti-slip drive wheels and steering servo.

[0016] Furthermore, the number of UWB base stations is at least six, and they are respectively arranged at the four vertices and the center of the two long sides of the ice rink to ensure that there are no obstacles between the UWB base stations and that at least four UWB base stations can be seen directly from any point in the ice rink.

[0017] Furthermore, the lidar sensor adopts a multi-line radar mid360, which can scan the environment around the intelligent vehicle in 360° and generate real-time point cloud data.

[0018] A detection method for an ice rink detection vehicle based on UWB positioning includes the following steps:

[0019] Step 1: System initialization and communication establishment. Deploy UWB base stations according to the size and shape of the ice rink and mark the relative coordinates. Establish a global rectangular coordinate system. Combine radar mapping methods to set the initial map of the ice rink and the initial obstacle range. Place the intelligent vehicle body at the initial point and establish communication between the remote operation host computer and the vehicle-mounted NUC slave computer.

[0020] Step 2: Task parameter setting and start-up. Set the movement speed of the intelligent vehicle body through remote operation host computer, and input the coordinates of the detection points of each target in the global rectangular coordinate system to form a detection point sequence. Then send the start command, and the intelligent vehicle body begins to perform the detection task.

[0021] Step 3: Global path planning and adaptive movement on the ice surface. The onboard NUC lower-level computer combines the IMU attitude sensor and the UWB positioning module for fusion positioning, generates a global path based on the path planning algorithm, and controls the intelligent vehicle to drive along the path.

[0022] Step 4: Real-time positioning and dynamic obstacle avoidance. During the movement of the intelligent vehicle, the on-board NUC lower-level computer updates the position in real time through the IMU attitude sensor and UWB positioning module, and combines the LiDAR sensor to obtain the surrounding environment information to determine whether there are obstacles. When an obstacle is detected, the local path planner replans the local path.

[0023] Step 5: Fixed-point data synchronous collection and upload. When the intelligent vehicle reaches any detection point, it stops moving. The on-board NUC lower computer sends synchronous collection instructions to the infrared temperature sensor, temperature and humidity sensor and light sensor. It reads and records the ice surface temperature, air temperature and humidity and light intensity in sequence, and sends them to the remote operation upper computer through the communication module.

[0024] Step Six: Task Completion and Automatic Return. The remote control host computer determines whether all detection points have been completed. If not, it continues according to the planned global path. If all detection points have been completed, the automatic return program is started, and the intelligent vehicle returns to the initial point.

[0025] Furthermore, in step three, the path planning algorithm discretizes the ice rink into grids with adjustable resolution based on a grid map model, and the planning process employs an improved... The algorithm has a cost function of f(n) = g(n) + h(n) + c(n), where g(n) represents the actual movement cost from the starting point to the current node n, h(n) is the heuristically estimated cost from the current node n to the target point, and c(n) is the turning penalty term, the value of which is positively correlated with the turning angle of the current node relative to the predecessor node.

[0026] Furthermore, in step three, the fusion positioning of the IMU attitude sensor and the UWB positioning module adopts an extended Kalman filter framework, with embedded residual chi-square detection and adaptive noise adjustment mechanism, and ice surface adhesion condition constraints are introduced in the state prediction stage to limit the instantaneous angular velocity range.

[0027] Furthermore, in step four, the local path planner adopts the dynamic window method, and its evaluation function introduces trajectory stability evaluation based on speed, orientation and obstacle avoidance scores. The trajectory stability evaluation simulates and calculates the expected lateral acceleration of the candidate trajectory. The greater the lateral acceleration, the lower the score.

[0028] Compared with the prior art, the beneficial effects of the present invention are:

[0029] 1. High positioning accuracy: By integrating UWB and IMU positioning, the shortcomings of a single sensor are overcome. Combined with an improved path planning algorithm and ice surface adhesion constraints, centimeter-level positioning accuracy is achieved in large ice rinks, ensuring accurate repeatability of detection points.

[0030] 2. High detection efficiency: Automated detection avoids the complicated work of manual detection. Multiple sensors are integrated on the intelligent vehicle body to realize the one-time synchronous acquisition of multiple parameters at a single point. The detection work that originally required the cooperation of multiple people is simplified to single-person operation, saving time and effort.

[0031] 3. Excellent data acquisition: Synchronous acquisition by multiple sensors avoids errors caused by human operation. Data is visualized in real time through a host computer, allowing for real-time observation of acquisition results and ensuring data consistency and reliability.

[0032] 4. Flexible deployment: The location of UWB base stations can be adjusted according to the size of the ice rink, and the system can be expanded to other indoor venue detection scenarios to adapt to indoor venues of different sizes and shapes. Attached Figure Description

[0033] Figure 1 This is an isometric view of the ice rink inspection trolley structure of the present invention;

[0034] Figure 2 This is a flowchart of the detection method of the present invention.

[0035] In the diagram: 1. Temperature and humidity sensor; 2. Support 1; 3. Column; 4. Support 2; 5. Communication module; 6. UWB positioning module; 7. Battery; 8. Protective shell; 9. Smart car body; 10. Onboard NUC lower unit; 11. Infrared temperature sensor; 12. Support 3; 13. Mounting bracket; 14. LiDAR sensor; 15. IMU attitude sensor; 16. Light sensor. Detailed Implementation

[0036] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the invention, not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0037] like Figure 1 As shown, an ice rink inspection vehicle based on UWB positioning includes an intelligent vehicle body 9, a control system, a UWB positioning module 6, a lidar sensor 14, an IMU attitude sensor 15, a temperature and humidity sensor 1, an infrared temperature sensor 11, a light sensor 16, and a communication module 5, wherein:

[0038] The intelligent vehicle body 9 adopts an Ackermann steering chassis and is equipped with a low-level controller. The Ackermann steering chassis adapts to ice surface movement scenarios and includes anti-slip drive wheels (motor-driven) and a steering servo. The low-level controller converts path planning results into drive electrical signals to control the movement of the anti-slip drive wheels and steering servo, achieving motion control. Furthermore, the intelligent vehicle body 9 internally houses a battery 7 as a power source for the electrical components. An external protective shell 8 is installed on the intelligent vehicle body 9 to support and protect other components. A column 3 is fixedly mounted on top of the intelligent vehicle body 9 for mounting the temperature and humidity sensor 1 and the light sensor 16. In one embodiment, the Ackerman steering chassis can also be equipped with a coaxial swing suspension, with a minimum turning radius of 1.5m and a maximum travel speed of 1.3m / s. The drive motor is a DC motor with a reduction ratio of 1:27 and a rated torque of 1.3N·m, ensuring stability when driving on ice. The underlying controller is an STM32 controller, which receives path instructions from the on-board NUC lower unit 10 via the CAN bus and outputs PWM signals to control the anti-slip drive wheels and steering servo motor, thereby realizing motion control of the intelligent vehicle body 9.

[0039] The control system includes an on-board NUC lower-level computer 10 and a remote control upper-level computer. The remote control upper-level computer sends parameters such as the vehicle's movement speed and the detection point sequence generated based on coordinates to the on-board NUC lower-level computer 10, and receives the real-time status of the intelligent vehicle body 9 and data from various sensors from the on-board NUC lower-level computer 10, thereby realizing remote control and data reading. The on-board NUC lower-level computer 10 is installed inside the intelligent vehicle body 9 and is used to receive commands from the remote control upper-level computer, calculate its own position and attitude by combining the position information of the UWB positioning module 6 and the attitude information of the IMU attitude sensor 15, plan a global path according to the target positions to be reached in sequence, and plan a local path by integrating the point cloud data of the lidar sensor 14 to achieve obstacle avoidance, and control the synchronous data acquisition of various sensors.

[0040] The UWB positioning module 6 works in conjunction with UWB base stations arranged around the ice rink. The number of UWB base stations is at least six (taking a standard speed skating ice rink of 60m×30m as an example, they are arranged at the four vertices and the center of the two long sides of the ice rink, respectively. The positions of the UWB base stations are determined according to the known ice rink conditions. The UWB base stations are placed on tripods at a height of 2m. There are no obstacles between the UWB base stations, and it is ensured that at least four UWB base stations can be seen directly from any point in the ice rink). The number of base stations can be increased according to the positioning accuracy requirements. The relative coordinates are calculated by calibrating the relative positions of the UWB base stations, thereby establishing a global rectangular coordinate system. The UWB positioning module 6 is mounted on the surface of the intelligent vehicle body 9. The current position of the intelligent vehicle body 9 is calculated by reading the distance to each UWB base station, and the position information is transmitted to the on-board NUC lower computer 10.

[0041] The lidar sensor 14 is mounted on the surface of the intelligent vehicle body 9 via a mounting bracket 13. It is used for obstacle detection and local path replanning. In actual working scenarios, obstacles may interfere with the path planned from an obstacle-free map, affecting normal movement. Therefore, the lidar sensor 14 scans the surrounding environment to generate real-time point cloud data for obstacle detection. When the onboard NUC lower-level computer 10 determines the presence of an obstacle based on the point cloud data, it triggers local path replanning to ensure the driving safety of the intelligent vehicle body 9. In one embodiment, the lidar sensor 14 is a multi-line radar mid360, capable of scanning the surrounding environment of the intelligent vehicle body 9 360° and generating real-time point cloud data.

[0042] The IMU attitude sensor 15 is mounted on the surface of the intelligent vehicle body 9 to obtain the attitude angle of the intelligent vehicle body 9, supplement the attitude information missing by the UWB positioning module 6, provide the vehicle-mounted NUC lower computer 10 with complete attitude data of the intelligent vehicle body 9, and assist the path planning algorithm in generating a detection path adapted to driving on ice.

[0043] The temperature and humidity sensor 1 is slidably locked onto the column 3 via the support 2. The initial position is 1.2 meters away from the ice surface and the height can be adjusted. It is used to measure the ambient temperature and humidity. The temperature measurement accuracy is ±0.5℃ and the humidity measurement accuracy is ±5%RH, which meets the measurement height and accuracy requirements.

[0044] The infrared temperature sensor 11 is mounted on the front end of the intelligent vehicle body 9 via support 3 12. It measures the ice surface temperature in a non-contact manner, with a range of -30℃ to 10℃ and a measurement accuracy of ±0.5℃. The installation position is about 5 cm away from the ice surface to avoid scratching, which solves the problem of waiting for temperature equilibrium of contact sensors and improves the temperature measurement efficiency.

[0045] The light sensor 16 is slidably locked onto the column 3 via the support 2 4. Its initial position is 1 meter away from the ice surface and its height can be adjusted. It is used to measure whether the lighting intensity above the ice surface meets the usage requirements. The range is 0-20000 lux and the measurement accuracy is ±5%.

[0046] The communication module 5 is mounted on the rear end of the intelligent vehicle body 9. It uses a high-power router and establishes a local area network to realize real-time data transmission between the on-board NUC lower computer 10 and the remote operation upper computer over long distances.

[0047] like Figures 1-2 As shown, a detection method for an ice rink detection vehicle based on UWB positioning includes the following steps:

[0048] Step 1: System Initialization and Communication Establishment

[0049] Based on the size and shape of the ice rink, UWB base stations are deployed around the rink, and the relative coordinates of each UWB base station are calibrated. A global rectangular coordinate system O-XYZ is established (with the UWB base station located at a certain vertex of the ice rink as the origin O, the X-axis along the long side of the ice rink rectangle, the Y-axis along the short side of the ice rink rectangle, and the Z-axis vertically upward). Combining radar mapping methods, the initial map of the ice rink and the initial obstacle range are set. The intelligent vehicle body 9 is placed at the initial point on the edge of the ice rink, and the normal and stable operation of various functions is checked. The remote operation host computer is deployed and turned on, so that the remote operation host computer connects to the communication module 5 to complete the communication establishment with the vehicle-mounted NUC lower computer 10.

[0050] Step 2: Task Parameter Settings and Startup

[0051] According to the requirements of the ice rink inspection task, the movement speed of the intelligent vehicle body 9 is set in the visual interface of the remote operation host computer, and the coordinates of the detection points of each target in the global rectangular coordinate system are entered in sequence to form a detection point sequence. Then, a start command is sent, and the remote operation host computer sends the movement speed and detection point sequence to the vehicle-mounted NUC lower computer 10, and the intelligent vehicle body 9 begins to carry out the inspection task.

[0052] Step 3: Global Path Planning and Adaptive Ice Surface Movement

[0053] The onboard NUC slave computer 10, based on a path planning algorithm, combines the attitude data from the IMU attitude sensor 15 with the position information from the UWB positioning module 6 to generate a global path, and then transmits this global path to the intelligent vehicle body 9 so that it can travel according to the path. Specifically:

[0054] After receiving the detection point sequence and specified parameters sent by the remotely operated host computer, the vehicle-mounted NUC lower-level computer 10 starts a path planning algorithm to generate a global path. This path planning algorithm discretizes the ice rink into grids with adjustable resolution based on a raster map model, and the planning process adopts an improved method. The algorithm has a cost function of f(n) = g(n) + h(n) + c(n), where g(n) represents the actual movement cost from the starting point to the current node n, h(n) is the heuristically estimated cost from the current node n to the target point, and c(n) is the turning penalty term, the value of which is positively correlated with the turning angle of the current node relative to the predecessor node, and is set based on the Ackerman turning geometry model and the ice surface adhesion coefficient.

[0055] To improve driving efficiency on ice, the cost function takes into account the Ackerman steering characteristics of the intelligent vehicle body 9, and applies an additional penalty coefficient to paths with excessive steering angles, thereby generating a smooth global path that balances path length and driving stability.

[0056] While generating the global path, the onboard NUC lower-level computer 10 starts a high-precision positioning system based on the fusion of the IMU attitude sensor 15 and the UWB positioning module 6 to provide centimeter-level positioning accuracy in icy environments. This positioning system specifically includes:

[0057] First, based on the global rectangular coordinate system O-XYZ, a vehicle coordinate system O-XYZ is established for the intelligent vehicle body 9 (with the geometric center of the intelligent vehicle body 9 as the origin o, the x-axis pointing in the direction of forward movement, the y-axis pointing to the left side of the vehicle body, and the z-axis pointing vertically upward). The state variables of the position and attitude of the intelligent vehicle body 9 in the ice rink plane are defined as a=[X o Y o ,θ,v,ω] T , of which (X o Y o) represents the planar position coordinates of the intelligent vehicle body 9 in the global coordinate system, θ is the heading angle, v is the linear velocity, and ω is the angular velocity.

[0058] Secondly, an inertial navigation prediction model is constructed based on the IMU attitude sensor 15. The IMU attitude sensor 15 outputs triaxial acceleration and angular velocity data in real time. After zero-bias calibration, gravity compensation, and filtering, combined with the kinematic constraints of the Ackerman steering chassis, the state variable a is predicted forward at a discrete time step Δt: the prediction is based on the integral of the angular velocity ω over the heading angle θ, and the prediction is based on the linear acceleration and the current heading angle θ. o With Y o The displacement increment in the direction is updated synchronously with the linear velocity v and angular velocity ω.

[0059] Then, an absolute position observation model is constructed based on the UWB positioning module 6. The distance is measured between the UWB positioning module 6 and the UWB base stations deployed around the ice rink. Under the premise of knowing the coordinates of the UWB base stations, the absolute position observation value of the intelligent vehicle body 9 in the global coordinate system is obtained by polygonal measurement. When the number of visible UWB base stations is greater than or equal to four, the position is solved by the weighted least squares method, and the reliability and observation noise level of the observation are evaluated based on the solution residual.

[0060] To further enhance the robustness of positioning in ice rink environments, an extended Kalman filter (EKF) framework can be employed for fusion, using the IMU attitude sensor 15 as the predictor and the UWB positioning module 6 as the observer. The EKF incorporates residual chi-square detection and adaptive noise adjustment mechanisms. When the information between the UWB observation and the IMU prediction exceeds a preset threshold, it is determined that the UWB observation may be affected by multipath reflection or occlusion interference. The system automatically increases the observation noise covariance matrix at that moment, reducing the weight of that observation in the fusion result, effectively smoothing positioning glitches caused by signal reflection. Simultaneously, through sliding window analysis of the statistical characteristics of information over a period of time, the ratio of IMU prediction noise to UWB observation noise is dynamically adjusted, allowing the fusion algorithm to adaptively converge to an appropriate weight configuration based on the current environment.

[0061] Furthermore, considering the low coefficient of adhesion on ice, the intelligent vehicle body 9 is prone to sideslip during sharp turns or rapid acceleration and deceleration. Therefore, ice adhesion condition constraints are introduced during the state prediction stage: based on the set maximum allowable lateral acceleration and the current speed, the range of achievable instantaneous angular velocity is limited to suppress unreasonable large-angle sharp turn predictions; when the lateral acceleration measured by the IMU attitude sensor 15 deviates significantly from the theoretical value calculated from the speed and curvature, the sideslip monitoring mechanism is triggered to correct the speed and attitude estimation, thereby avoiding motion estimation distortion caused by slippage.

[0062] Through the fusion positioning of IMU attitude sensor 15 and UWB positioning module 6, the accuracy of the fusion positioning data obtained in the ice rink environment can be consistently better than 10 cm. Even in areas where the local UWB signal is briefly lost or multipath reflection is severe, the trajectory of the intelligent vehicle body 9 can still remain continuous and smooth, which significantly improves the repeatability accuracy and positioning robustness of the detection points.

[0063] Finally, the onboard NUC lower-level computer 10 sends the planned sequence of detection points and their corresponding pose information to the intelligent vehicle body 9 via the CAN bus at a fixed frequency. The underlying controller then controls the anti-slip drive wheels and steering servo in a closed loop to move smoothly along the planned path.

[0064] Step 4: Real-time positioning and dynamic obstacle avoidance

[0065] During the movement of the intelligent vehicle body 9, the onboard NUC lower-level computer 10 updates its position in real time through the IMU attitude sensor 15 and the UWB positioning module 6, and obtains surrounding environmental information in conjunction with the lidar sensor 14 to determine whether there are obstacles. The onboard NUC lower-level computer 10 sets a fan-shaped area with a radius of 3 meters centered on the intelligent vehicle body 9 as a warning zone, and can additionally set an emergency obstacle avoidance zone within a radius of 2 meters. When an obstacle is detected entering the warning zone, the local path is replanned by the local path planner. The local path planner adopts the dynamic window method, and its evaluation function introduces trajectory stability evaluation (simulating the expected lateral acceleration of the candidate trajectory, the greater the lateral acceleration, the lower the score) on the basis of conventional speed, orientation, and obstacle avoidance scores. When an obstacle is detected entering the emergency obstacle avoidance zone, an emergency braking command is immediately sent to ensure safety. After the intelligent vehicle body 9 successfully avoids the obstacle, it automatically returns to the global path and continues to move towards the next detection point.

[0066] Step 5: Synchronous Collection and Uploading of Fixed-Point Data

[0067] When the on-board NUC lower-level computer 10 detects that the intelligent vehicle body 9 has reached any detection point, it controls the intelligent vehicle body 9 to stop moving (the stop time can be set to 1 minute, which is mainly to eliminate reading fluctuations caused by movement vibration), and enters the fixed-point data acquisition mode. The on-board NUC lower-level computer 10 sends synchronous acquisition commands to the infrared temperature sensor 11, temperature and humidity sensor 1, and light sensor 16, and sequentially reads and records the detection data such as ice surface temperature, air temperature and humidity, and light intensity, and sends them to the remote operation upper computer through the communication module 5 to complete data storage and visualization display.

[0068] Step Six: Mission Completion and Automatic Return

[0069] After receiving and storing the detection data of the current detection point, the remote operation host computer determines whether all preset detection points have been detected. If not, it continues according to the planned global path. If all detection points have been detected, the remote operation host computer sends a task completion command to the vehicle-mounted NUC slave computer 10, starts the automatic return program, and reverses the return path according to the planned path until the intelligent vehicle body 9 returns to the initial point, and the detection task is completed.

[0070] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0071] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. An ice rink inspection vehicle based on UWB positioning, characterized in that: It includes the intelligent vehicle body (9), control system, UWB positioning module (6), lidar sensor (14), IMU attitude sensor (15), temperature and humidity sensor (1), infrared temperature sensor (11), light sensor (16) and communication module (5). The intelligent vehicle body (9) is equipped with a bottom-level controller and anti-slip drive wheels and steering servo to adapt to ice surface movement scenarios. The path planning result is converted into drive electrical signals through the bottom-level controller to achieve motion control. A fixed column (3) is fixed on the top of the intelligent vehicle body (9). The control system includes an on-board NUC lower unit (10) and a remote control upper unit. The on-board NUC lower unit (10) is installed inside the intelligent vehicle body (9) and is used to receive instructions, calculate position and attitude, plan paths and control sensor data acquisition. The remote control upper unit is used to send instructions, receive feedback data and realize remote control and data reading. The UWB positioning module (6) is mounted on the surface of the intelligent vehicle body (9). It works in conjunction with the UWB base stations arranged around the ice rink to calculate the current position of the intelligent vehicle body (9) and transmit it to the on-board NUC lower computer (10). The lidar sensor (14) is mounted on the surface of the intelligent vehicle body (9) and is used for obstacle detection and local path replanning; The IMU attitude sensor (15) is mounted on the surface of the intelligent vehicle body (9) to obtain the attitude angle of the intelligent vehicle body (9) and provide complete attitude data for the vehicle-mounted NUC lower computer (10). The temperature and humidity sensor (1) and the light sensor (16) are both slidably locked and installed on the column (3), and the infrared temperature sensor (11) is mounted on the front end of the intelligent vehicle body (9). The three are used to measure the ambient temperature and humidity, the lighting intensity above the ice surface and the ice surface temperature, respectively. The communication module (5) is mounted on the rear end of the intelligent vehicle body to realize real-time data transmission between the vehicle-mounted NUC lower computer (10) and the remote control upper computer.

2. The ice rink detection vehicle based on UWB positioning according to claim 1, characterized in that: The intelligent vehicle body (9) adopts an Ackerman steering chassis with a coaxial swing suspension, with a minimum turning radius of 1.5m and a maximum travel speed of 1.3m / s. The underlying controller is an STM32 controller, which receives path instructions from the vehicle-mounted NUC lower unit (10) via the CAN bus and outputs PWM signals to control the anti-slip drive wheels and steering servo.

3. The ice rink detection vehicle based on UWB positioning according to claim 1, characterized in that: The number of UWB base stations is at least six, and they are respectively arranged at the four vertices and the center of the two long sides of the ice rink, ensuring that there are no obstacles between the UWB base stations and that at least four UWB base stations can be seen directly from any point in the ice rink.

4. The ice rink detection vehicle based on UWB positioning according to claim 1, characterized in that: The lidar sensor (14) adopts a multi-line radar mid360, which can scan the environment around the intelligent vehicle body (9) in 360° and generate real-time point cloud data.

5. A detection method for an ice rink detection vehicle based on UWB positioning, characterized in that: The ice rink detection vehicle based on UWB positioning according to claim 1, its detection method includes the following steps: Step 1: System initialization and communication establishment. Deploy UWB base stations and mark relative coordinates according to the size and shape of the ice rink. Establish a global rectangular coordinate system. Combine radar mapping method to set the initial map of the ice rink and the initial obstacle range. Place the intelligent vehicle body (9) at the initial position and establish communication between the remote operation host computer and the vehicle-mounted NUC slave computer (10). Step 2: Task parameter setting and start-up. The movement speed of the intelligent vehicle body (9) is set by remotely operating the host computer, and the coordinates of the detection points of each target in the global rectangular coordinate system are input in sequence to form a detection point sequence. Then, the start command is sent and the intelligent vehicle body (9) begins to perform the detection task. Step 3: Global path planning and ice surface adaptive movement. The vehicle-mounted NUC lower computer (10) combines the fusion positioning of the IMU attitude sensor (15) and the UWB positioning module (6) to generate a global path based on the path planning algorithm, and controls the intelligent vehicle body (9) to make it travel according to the path. Step 4: Real-time positioning and dynamic obstacle avoidance. During the movement of the intelligent vehicle body (9), the on-board NUC lower computer (10) updates the position in real time through the IMU attitude sensor (15) and the UWB positioning module (6), and obtains the surrounding environment information by combining the laser radar sensor (14) to determine whether there are obstacles. When an obstacle is detected, the local path is replanned by the local path planner. Step 5: Fixed-point data synchronous acquisition and upload. When the intelligent vehicle body (9) reaches any detection point, it stops moving. The vehicle-mounted NUC lower computer (10) sends synchronous acquisition instructions to the infrared temperature sensor (11), temperature and humidity sensor (1) and light sensor (16), and reads and records the ice surface temperature, air temperature and humidity and light intensity in sequence, and sends them to the remote operation upper computer through the communication module (5). Step 6: Task completion and automatic return. The remote operation host computer determines whether all detection points have been completed. If not, it continues according to the planned global path. If all detection points have been completed, the automatic return program is started, and the intelligent vehicle body (9) returns to the initial point.

6. The detection method for an ice rink detection vehicle based on UWB positioning according to claim 5, characterized in that: In step three, the path planning algorithm discretizes the ice rink into grids with adjustable resolution based on a grid map model, and the planning process employs an improved method. The algorithm has a cost function of f(n) = g(n) + h(n) + c(n), where g(n) represents the actual movement cost from the starting point to the current node n, h(n) is the heuristically estimated cost from the current node n to the target point, and c(n) is the turning penalty term, the value of which is positively correlated with the turning angle of the current node relative to the predecessor node.

7. The detection method for an ice rink detection vehicle based on UWB positioning according to claim 5, characterized in that: In step three, the fusion positioning of the IMU attitude sensor (15) and the UWB positioning module (6) adopts an extended Kalman filter framework, with embedded residual chi-square detection and adaptive noise adjustment mechanism. In the state prediction stage, ice surface adhesion condition constraints are introduced to limit the instantaneous angular velocity range.

8. The detection method for an ice rink detection vehicle based on UWB positioning according to claim 5, characterized in that: In step four, the local path planner uses the dynamic window method. Its evaluation function introduces trajectory stability evaluation based on speed, orientation and obstacle avoidance scores. The trajectory stability evaluation simulates and calculates the expected lateral acceleration of the candidate trajectory. The greater the lateral acceleration, the lower the score.

Images