Method and terminal for realizing turn-round of unmanned vehicle based on point cloud

Abstract

The invention provides a method and a terminal for realizing turning around of an unmanned vehicle based on point cloud. The method comprises the following steps: acquiring point cloud map data and generating an initial map; acquiring point cloud data of the current environment, matching the point cloud data with the initial map, and determining current position information and attitude information of the vehicle based on the motion information of the vehicle; recording continuous position information of the vehicle in a time sequence, and obtaining a motion track of the vehicle based on the continuous position information; based on the historical data set, when it is detected that the vehicle is about to arrive at the U-turn position, point cloud tracking is used for achieving U-turn within the safe range; according to the method, the point cloud map data is acquired and the initial map is generated, so that the vehicle can smoothly execute the turning operation in the narrow road environment, the problem that the position of the unmanned vehicle needs to be frequently adjusted when the unmanned vehicle turns around on the narrow road is effectively solved, the operation frequency is reduced, and the turning efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of unmanned driving, and in particular to a method and a terminal for realizing U-turn of an unmanned vehicle based on point cloud. Background Art

[0002] With the continuous development of autonomous driving technology, driverless vehicles are gradually becoming practical applications. However, in real-world road environments, especially in limited space such as narrow urban roads, alleyways, and parking lots, vehicle U-turns still face significant challenges. While traditional autonomous driving systems can perform simple driving tasks well, U-turns on narrow roads require frequent adjustments to the vehicle's position due to limited space, irregular obstacle distribution, and diverse environmental conditions. This results in low operational efficiency and increased system instability.

[0003] Traditional methods typically rely on static maps or single sensor information for path planning and obstacle detection. However, these methods have certain limitations. First, static maps cannot reflect real-time environmental changes and may not accurately describe the real-time location of obstacles and road conditions. Second, in dynamic environments, traditional path planning methods often cannot accurately predict the vehicle's motion trajectory and posture changes, causing the vehicle to frequently adjust its position in narrow spaces, increasing the complexity and uncertainty of U-turns. In addition, point cloud matching based on traditional algorithms has low matching accuracy in complex environments, severe occlusion, or high noise levels, which also affects the accuracy of vehicle positioning and posture estimation. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method and terminal for realizing U-turn of an unmanned vehicle based on point cloud, so as to solve the problem that unmanned vehicles have to make U-turns frequently on narrow roads.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is: A method for realizing a U-turn of an unmanned vehicle based on point cloud, comprising the steps of: S1. Obtain point cloud map data and generate an initial map; S2. Acquire point cloud data of the current environment and match it with the initial map, and determine the current position information and posture information of the vehicle based on the vehicle's motion information; S3. Recording the continuous position information of the vehicle in time series, and obtaining the movement trajectory of the vehicle based on the continuous position information; S4. Based on the historical data set, when it is detected that the vehicle is about to reach the U-turn position, the U-turn is realized by using point cloud tracking.

[0006] In order to solve the above technical problems, another technical solution adopted by the present invention is: A terminal for implementing a U-turn of an unmanned vehicle based on a point cloud includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the following steps are performed: S1. Obtain point cloud map data and generate an initial map; S2. Acquire point cloud data of the current environment and match it with the initial map, and determine the current position information and posture information of the vehicle based on the vehicle's motion information; S3. Recording the continuous position information of the vehicle in time series, and obtaining the movement trajectory of the vehicle based on the continuous position information; S4. Based on the historical data set, when it is detected that the vehicle is about to reach the U-turn position, the U-turn is realized by using point cloud tracking.

[0007] The beneficial effects of the present invention are: providing a method and terminal for realizing U-turn of unmanned vehicles based on point cloud, mainly by obtaining point cloud map data and generating an initial map, ensuring that the vehicle can smoothly perform U-turn operations in narrow road environments, thereby effectively solving the problem of unmanned vehicles frequently needing to adjust their positions when turning in narrow roads, reducing the number of operations and improving U-turn efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 1 is a flow chart of a method for implementing a U-turn of an unmanned vehicle based on point cloud in an embodiment of the present invention; Figure 2 This is a flow chart of motion trajectory recording in a method for realizing a U-turn of an unmanned vehicle based on point cloud in an embodiment of the present invention; Figure 3 This is a flow chart of point cloud tracking in a method for realizing a U-turn of an unmanned vehicle based on point cloud in an embodiment of the present invention; Figure 4 Schematic diagram of a terminal for implementing a U-turn of an unmanned vehicle based on point cloud in an embodiment of the present invention; Description of labels: 1. A terminal for enabling unmanned vehicle U-turns based on point clouds; 2. Memory; 3. Processor. DETAILED DESCRIPTION

[0009] To illustrate the technical content, achieved objectives and effects of the present invention in detail, the following description is given in conjunction with the embodiments and accompanying drawings.

[0010] Please refer to Figure 1 A method for realizing a U-turn of an unmanned vehicle based on point cloud comprises the following steps: S1. Obtain point cloud map data and generate an initial map; S2. Obtain point cloud data of the current environment and match it with the initial map, and determine the current position and posture information of the vehicle based on the vehicle's motion information; S3. Recording the continuous position information of the vehicle in time series, and obtaining the movement trajectory of the vehicle based on the continuous position information; S4. Based on the historical data set, when it is detected that the vehicle is about to reach the U-turn position, the U-turn is realized by using point cloud tracking.

[0011] As can be seen from the foregoing description, the beneficial effects of the present invention are that, by acquiring environmental point cloud data and generating an initial map, the vehicle can accurately perceive its surroundings, enabling efficient U-turns in confined spaces. Preferably, this method utilizes the high-precision three-dimensional point cloud data provided by laser radar (LiDAR), combined with the vehicle's motion information, posture information, historical data, and trajectory. Through precise point cloud matching and tracking technology, it can efficiently complete U-turns within confined spaces, reducing the frequency of manual intervention and increasing the degree of automation. By combining point cloud maps with motion information, the system can accurately predict the vehicle's trajectory and position, ensuring accuracy and safety during U-turns, thereby resolving the issues of low operational efficiency and instability associated with frequent adjustments in traditional methods.

[0012] In some embodiments, step S2 specifically includes the steps of: S21. Optimize the point cloud data of the current environment based on the normal distribution transformation algorithm; S22, matching the optimized point cloud data of the current environment with the initial map, and estimating and generating the initial position information and posture information of the vehicle; S23. Determine the current position information and posture information of the vehicle based on the motion information of the vehicle and the estimated initial position information and posture information.

[0013] As can be seen from the above description, by introducing the Normal Distribution Transformation (NDT) algorithm to optimize the point cloud data of the current environment, the accuracy and robustness of point cloud matching are further improved. Specifically, the NDT algorithm is used to convert point cloud data into a grid map or Gaussian mixture model. The principle behind this is that the NDT algorithm can model the point cloud using a Gaussian distribution, improving the noise problem in the point cloud data. This allows the vehicle's positioning and posture information to be accurately acquired even in complex environments or with significant occlusion. At the same time, combining matching with the vehicle's motion information can effectively reduce matching errors caused by instantaneous motion, improve the system's real-time performance and accuracy, and avoid the position drift that can occur when relying on static maps.

[0014] Specifically, step S23 includes the following steps: S231, using the vehicle's motion information and the estimated initial position information and posture information, predicting the vehicle's current position information and posture information; S232: Determine a search range based on the predicted current position information and posture information; S233. Perform point cloud matching within the search range using an iterative closest point algorithm, and determine the vehicle's current position and posture information based on the matching results.

[0015] As can be seen from the above description, accurate search range prediction is achieved by combining the vehicle's motion information with initially estimated position and attitude information. Vehicle motion information, such as IMU (Inertial Measurement Unit) data and odometer information, can provide real-time speed and acceleration data. Based on this information, the system can predict the vehicle's current position and automatically adjust the search range, thereby performing point cloud matching more efficiently. Traditional point cloud matching methods often rely solely on static or known maps, ignoring the impact of real-time vehicle motion. However, this invention dynamically adjusts the search range, enabling more flexible and accurate positioning and navigation of vehicles in confined environments, significantly reducing operational complexity and computational burden.

[0016] Preferably, the search range is determined as follows: The search range is based on the vehicle's length, width, and wheelbase for narrow U-turns. With the vehicle at the center, the front-to-back range covers 1.5 to 2 times the vehicle's length, and the left-to-right range covers 1.5 to 2 times the vehicle's width. Based on this data, a quasi-rectangular range is plotted on the map.

[0017] Adjust the search range based on obstacle information. The search range should avoid obstacles, with obstacles as the boundary, and leave a distance of 0.3 to 0.5 meters around them as a safety margin.

[0018] Preferably, step S233 specifically includes the steps of: The vehicle's motion information is used as the initial estimation data and auxiliary constraints of the iterative closest point algorithm. The source point cloud data of the current environment and the target point cloud data of the initial gradient are read. The iterative closest point algorithm is used to perform point cloud matching, and the current position and posture information of the vehicle are output based on the matching results.

[0019] As can be seen from the preceding description, the Iterative Closest Point (ICP) algorithm, combined with vehicle motion information as an initial estimate and constraints, accurately updates the vehicle's position and posture by matching the source and target point cloud data of the current environment. The ICP algorithm can optimize the vehicle's trajectory through iterative adjustments in dynamic environments. Incorporating vehicle motion information such as IMU data, the algorithm effectively avoids errors caused by environmental complexity or dynamic changes, thereby improving system stability and robustness.

[0020] Among them, during the ICP algorithm iteration, the rotation and translation in the calculated results are corrected according to the IMU data to improve the stability and reliability of the matching, especially when the vehicle motion changes rapidly, to avoid the errors caused by relying solely on power matching.

[0021] After reading the source point cloud data of the current environment and the target point cloud data of the initial gradient, the source and target point clouds are preprocessed. This process involves setting the initial transformation matrix and time interval, and then starting multiple iterations. In each iteration, the current IMU data (vehicle motion data) is acquired, the IMU data is fused into the initial transformation matrix, ICP registration is performed, and the initial transformation matrix is updated. The final transformation is then applied to the source point cloud.

[0022] In addition, the NDT algorithm is used to optimize the selected area to improve the matching effect.

[0023] In some embodiments, step S4 specifically includes: S41. When detecting that the vehicle is about to reach a U-turn position, predicting the path and posture change of the vehicle based on the historical data set and the movement trajectory of the vehicle; S42. Using a point cloud tracking method, adjust the vehicle's motion according to the predicted path within the search range to ensure that the vehicle can perform a U-turn within a safe range; S43. During point cloud tracking, the system monitors the vehicle's surrounding environment and obstacle information in real time, and ensures the accuracy and safety of U-turns by adjusting the vehicle's steering, throttle, and brake control parameters. S44. After the vehicle completes the U-turn operation, the current position and posture information of the vehicle is updated, and the motion trajectory data at this time is recorded to ensure that the system can continuously track the position and posture of the vehicle.

[0024] As can be seen from the above description, by utilizing historical data sets for point cloud tracking, when a vehicle is detected approaching a U-turn position, the system can accurately predict the vehicle's turn path and ensure that the turn is completed within a safe range. This method not only incorporates the vehicle's real-time motion trajectory but also monitors surrounding obstacles in real time, dynamically adjusting the path to enable the vehicle to successfully turn in confined environments. Through precise control of point cloud tracking, the system reduces U-turn failures caused by space constraints and improves the success rate of vehicle U-turn operations.

[0025] For details, please refer to Figure 2 The specific process of recording the motion trajectory in step S44 is as follows: 1) Start the operation Operate the vehicle (manually) into the road section where a U-turn is required, ensure that the vehicle maintains a stable driving state, and start recording the vehicle's U-turn process.

[0026] 2) Determine whether the U-turn operation can be completed Based on the vehicle's current position and the surrounding environment ahead, the system determines whether there is sufficient space to make a U-turn. If so, the system proceeds to the next step; if not, the system stops and exits.

[0027] 3) Path data storage If the path space meets the conditions, the system stores the corresponding point cloud data along the vehicle's driving path at intervals of 0.1 meters, including the vehicle's coordinates, direction status, and vehicle position data.

[0028] 4) Determine whether the vehicle ahead can continue to move forward Determines whether the vehicle can successfully reach the U-turn position. If there is an obstacle or insufficient space ahead, the vehicle will pause and return to its previous position. If there is sufficient space ahead to continue, the U-turn will be executed.

[0029] 5) Determine whether there is enough space at the rear of the vehicle Determine whether there is enough space at the rear of the vehicle to make a U-turn. If there is, proceed with the U-turn. If there is insufficient space, adjust the vehicle's position and ensure there is enough space at the rear to make the U-turn.

[0030] 6) Route adjustment and U-turn execution After confirming that there is sufficient space at the rear of the vehicle, the system automatically adjusts the steering wheel and performs a U-turn, and the vehicle performs a U-turn according to the predetermined path.

[0031] 7) Complete the U-turn and record it After the U-turn operation is completed, the system records the complete path and related information of the U-turn to ensure that it can be used for reference and optimization in the future. After the operation is completed, the U-turn process is exited and the record is completed.

[0032] Preferably, please refer to Figure 3 The steps of point cloud tracking in step S42 are as follows: 1) Start the vehicle U-turn cycle function Activate the vehicle's automatic U-turn system to ensure that the system can make path decisions and U-turns based on current road conditions and vehicle position.

[0033] 2) Select a suitable loop starting point The system traverses all available road sections and selects a valid starting point closest to the vehicle's current point as the first loop point for the U-turn.

[0034] 3) Determine whether the loop point is valid Based on the selected starting point, the system determines whether the point is suitable for continuing the operation. If it is valid, it proceeds to the next step; if not, it ends the current operation.

[0035] 4) The vehicle moves forward and cycles using the front wheel alignment mode After confirming that the loop point is valid, the vehicle starts to move forward and performs a U-turn using the vehicle's front wheel alignment mode. At this time, the vehicle automatically adjusts according to the set path.

[0036] 5) Determine whether you have entered the correct cycle stage The system determines whether the vehicle has entered the correct phase of the U-turn cycle. If so, the system continues the U-turn; if not, it adjusts the route.

[0037] 6) Determine whether the last loop point is valid As the vehicle continues to move forward, the system checks the validity of the last loop point. If valid, execution continues; if not, a path switch is required to select the next appropriate loop point.

[0038] 7) End of loop When the system confirms that the U-turn operation is completed, it determines whether to end the cycle. If the cycle is completed, the operation is successful and the U-turn ends.

[0039] 8) Switch to the next search point If the current loop point is invalid or the U-turn fails, the system switches to the next search point and continues the loop operation until a valid loop point is found and the U-turn is completed.

[0040] 9) End the operation After the U-turn operation is successfully completed, the loop ends and the system returns to the initial state, ready for the next operation.

[0041] Please refer to Figure 4 A terminal 1 for realizing a U-turn of an unmanned vehicle based on a point cloud includes a memory 2, a processor 3, and a computer program stored in the memory and executable on the processor 3. When the processor 3 executes the computer program, the steps in a method for realizing a U-turn of an unmanned vehicle based on a point cloud are completed.

[0042] In summary, the present invention provides a method and terminal for implementing U-turns for unmanned vehicles based on point clouds. By integrating high-precision point cloud data, vehicle motion information, historical data, and point cloud tracking technology, the method significantly improves the accuracy and efficiency of U-turns in complex and confined environments. Specifically, the present invention exhibits significant beneficial effects in the following aspects: This invention utilizes high-precision three-dimensional point cloud data provided by LiDAR, combined with the vehicle's real-time motion information, posture information, and historical data, to enable efficient U-turns in narrow or obstacle-filled environments. Through point cloud matching and point cloud tracking technology, the vehicle's trajectory can be accurately predicted and controlled, reducing operational delays and inefficiencies caused by frequent position adjustments. Traditional unmanned driving systems typically require frequent manual intervention, especially when making U-turns in confined spaces. However, this invention uses precise point cloud data sensing and path prediction to automatically identify and adjust the vehicle's position and posture, reducing reliance on manual intervention and enabling the vehicle to automatically and efficiently complete U-turns in complex environments. By introducing the Normal Distribution Transform (NDT) algorithm to optimize point cloud data, the vehicle's positioning and posture information can be accurately acquired even in complex environments or with significant occlusion. Combined with vehicle motion information (such as IMU and odometer data), the system can dynamically adjust the search range and optimize point cloud matching, improving matching accuracy and the system's real-time performance. The Iterative Closest Point (ICP) algorithm, combined with vehicle motion information, optimizes the point cloud matching process in dynamic environments through continuous iterative adjustments. Especially in situations where vehicle motion changes rapidly, the inclusion of IMU data enables precise corrections to rotation and translation calculations, thereby improving the stability and robustness of point cloud matching and reducing errors caused by environmental changes. This invention provides an intelligent decision-making process for path selection and loop determination, automatically identifying and selecting appropriate loop points and executing loops until a U-turn is successfully completed. The system automatically switches to the next search point based on real-time feedback and adjusts the path in the event of a U-turn failure, minimizing the risk of U-turn failure and improving the autonomous driving system's adaptability in confined spaces.

[0043] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent transformations made using the contents of the present invention's description and drawings, or directly or indirectly applied in related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for realizing U-turn of an unmanned vehicle based on point cloud, characterized by: Including steps: S1. Obtain point cloud map data and generate an initial map; S2. Acquire point cloud data of the current environment and match it with the initial map, and determine the current position information and posture information of the vehicle based on the vehicle's motion information; S3. Recording the continuous position information of the vehicle in time series, and obtaining the movement trajectory of the vehicle based on the continuous position information; S4. Based on the historical data set, when it is detected that the vehicle is about to reach the U-turn position, the U-turn is realized by using point cloud tracking.

2. The method for realizing U-turn of an unmanned vehicle based on point cloud according to claim 1, characterized in that: The step S2 specifically includes the following steps: S21. Optimize the point cloud data of the current environment based on the normal distribution transformation algorithm; S22, matching the optimized point cloud data of the current environment with the initial map, and estimating and generating initial position information and posture information of the vehicle; S23. Determine the current position information and posture information of the vehicle based on the motion information of the vehicle and the estimated initial position information and posture information.

3. The method for realizing U-turn of an unmanned vehicle based on point cloud according to claim 2, characterized in that: The step S23 specifically includes the following steps: S231, using the vehicle's motion information and the estimated initial position information and posture information, predicting the vehicle's current position information and posture information; S232: Determine a search range based on the predicted current position information and posture information; S233. Perform point cloud matching within the search range using an iterative closest point algorithm, and determine the current position information and posture information of the vehicle based on the matching results.

4. The method for realizing U-turn of an unmanned vehicle based on point cloud according to claim 3, characterized in that: The step S233 specifically includes the following steps: The vehicle's motion information is used as the initial estimation data and auxiliary constraints of the iterative closest point algorithm. The source point cloud data of the current environment and the target point cloud data of the initial gradient are read. The iterative closest point algorithm is used to perform point cloud matching, and the current position and posture information of the vehicle are output based on the matching results.

5. The method for realizing U-turn of an unmanned vehicle based on point cloud according to claim 1, characterized in that: The step S4 specifically includes: S41. When detecting that the vehicle is about to reach a U-turn position, predicting the path and posture change of the vehicle based on the historical data set and the movement trajectory of the vehicle; S42. Using a point cloud tracking method, adjust the vehicle's motion according to the predicted path within the search range to ensure that the vehicle can perform a U-turn within a safe range; S43. During point cloud tracking, the system monitors the vehicle's surrounding environment and obstacle information in real time, and ensures the accuracy and safety of U-turns by adjusting the vehicle's steering, throttle, and brake control parameters. S44: After the vehicle completes the U-turn operation, the current position and posture information of the vehicle is updated, and the motion trajectory data at this time is recorded.

6. A terminal for realizing U-turn of an unmanned vehicle based on point cloud, characterized by: The invention comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the following steps are performed: S1. Obtain point cloud map data and generate an initial map; S2. Acquire point cloud data of the current environment and match it with the initial map, and determine the current position information and posture information of the vehicle based on the vehicle's motion information; S3. Recording the continuous position information of the vehicle in time series, and obtaining the movement trajectory of the vehicle based on the continuous position information; S4. Based on the historical data set, when it is detected that the vehicle is about to reach the U-turn position, the U-turn is realized by using point cloud tracking.

7. The terminal for realizing U-turn of an unmanned vehicle based on point cloud according to claim 6, characterized in that: The step S2 specifically includes the following steps: S21. Optimize the point cloud data of the current environment based on the normal distribution transformation algorithm; S22, matching the optimized point cloud data of the current environment with the initial map, and estimating and generating initial position information and posture information of the vehicle; S23. Determine the current position information and posture information of the vehicle based on the motion information of the vehicle and the estimated initial position information and posture information.

8. The terminal for realizing U-turn of an unmanned vehicle based on point cloud according to claim 7, characterized in that: The step S23 specifically includes the following steps: S231, using the vehicle's motion information and the estimated initial position information and posture information, predicting the vehicle's current position information and posture information; S232: Determine a search range based on the predicted current position information and posture information; S233. Perform point cloud matching within the search range using an iterative closest point algorithm, and determine the current position information and posture information of the vehicle based on the matching results.

9. The terminal for realizing U-turn of an unmanned vehicle based on point cloud according to claim 8, characterized in that: The step S233 specifically includes the following steps: The vehicle's motion information is used as the initial estimation data and auxiliary constraints of the iterative closest point algorithm. The source point cloud data of the current environment and the target point cloud data of the initial gradient are read. The iterative closest point algorithm is used to perform point cloud matching, and the current position and posture information of the vehicle are output based on the matching results.

10. The terminal for realizing U-turn of an unmanned vehicle based on point cloud according to claim 6, characterized in that: The step S4 specifically includes: S41. When detecting that the vehicle is about to reach a U-turn position, predicting the path and posture change of the vehicle based on the historical data set and the movement trajectory of the vehicle; S42. Using a point cloud tracking method, adjust the vehicle's motion according to the predicted path within the search range to ensure that the vehicle can perform a U-turn within a safe range; S43. During point cloud tracking, the system monitors the vehicle's surrounding environment and obstacle information in real time, and ensures the accuracy and safety of U-turns by adjusting the vehicle's steering, throttle, and brake control parameters. S44: After the vehicle completes the U-turn operation, the current position and posture information of the vehicle is updated, and the motion trajectory data at this time is recorded.

Images