ROS-oriented unmanned vehicle positioning precision improvement method and system

Abstract

The invention relates to an ROS-oriented unmanned vehicle positioning precision improvement method and system. The system comprises a BDS receiver, a sensor and an ROS controller. Firstly, a BDS receiver obtains current position information of a trolley and transmits the current position information to an ROS controller; the sensor detects the surrounding environment and transmits data to the ROS controller; the ROS controller calculates the optimal path of the trolley by using a path planning algorithm based on the BDS position information and the sensor data; and then, the ROS controller transmits the path information to the actuator, and the actuator controls the trolley to run along the planned path. By combining BDS navigation position information and sensor data, the system can realize efficient path planning, and the navigation precision and efficiency of the ROS trolley are improved.

Description

Technical Field

[0001] The present invention belongs to the field of navigation and positioning, and specifically relates to a method and system for improving positioning accuracy of a ROS unmanned vehicle. Background Art

[0002] In the field of modern autonomous driving and unmanned vehicle navigation, positioning accuracy is a key indicator, which is used to determine the specific position of mobile unmanned vehicles in complex environments to achieve safe and efficient navigation. Through the positioning method combined with BDS and the sensor fusion algorithm, the influence of IMU accumulation error and drift error caused by long-term use can be reduced.

[0003] Beidou Satellite Navigation and Positioning System (BDS) is a widely used positioning technology with advantages such as global coverage and high precision. It has become an indispensable part of the field of autonomous driving and robot navigation. However, using BDS positioning alone has problems such as insufficient accuracy and occlusion, especially in urban environments or densely built areas. Therefore, combining BDS positioning with other sensor information can improve the accuracy and robustness of positioning.

[0004] An inertial measurement unit (IMU) is a device that integrates multiple inertial sensors to measure and calculate the acceleration and angular velocity of an object. A typical IMU includes a three-axis accelerometer and a three-axis gyroscope, and sometimes a magnetometer to provide orientation information. These sensors can measure the linear acceleration, angular velocity, and orientation of an object in real time without relying on external references, so they are particularly useful in environments without BDS signals or limited space.

[0005] In ROS, the IMU system publishes data through the sensor message publisher (sensor_msgs). Common message types include acceleration, angular velocity, and magnetic field strength. These data can be transmitted to other nodes, such as navigation nodes or control nodes, through ROS topics to perform tasks such as navigation, path planning, and attitude control.

[0006] Typically, IMU data in ROS needs to be filtered and fused to improve the accuracy and stability of attitude estimation. Common methods include using a Kalman filter or an extended Kalman filter (EKF) to fuse the accelerometer, gyroscope, and magnetometer data from the IMU to remove noise and provide a more accurate attitude estimate.

[0007] The application of IMU system in ROS car is very important, it provides key real-time data for navigation, positioning and control, especially in environments without BDS signal or limited BDS signal, such as indoor or urban canyon. Through the flexibility and message passing mechanism of ROS, IMU system can be seamlessly integrated with other sensors (such as lidar or visual sensor), thus enhancing the perception and decision-making ability of the car.

[0008] The existing unmanned vehicle positioning methods have the following main shortcomings:

[0009] 1) Single dependency problem: When relying solely on BDS for positioning, there will be problems such as signal blocking, multipath effect, and insufficient signal strength. Especially in urban canyons, tunnels, high-rise dense areas or indoor environments, the stability and accuracy of BDS signals will be greatly reduced.

[0010] Relying solely on IMU for positioning will result in a long-term integration drift problem, that is, as time goes by, the positioning accuracy will gradually decrease due to the accumulation of sensor noise and integration errors.

[0011] 2) Insufficient accuracy and robustness: The accuracy and robustness of a single sensor system are usually not as good as those of a multi-sensor fusion system. Using BDS alone may not provide high enough accuracy to meet the high requirements of autonomous driving; and using IMU alone will make it difficult to maintain stable positioning accuracy for a long time.

[0012] 3) Poor environmental adaptability: Existing technologies often have difficulty adapting to complex and changing environmental conditions. In complex scenarios such as large changes in light, many dynamic obstacles, and unclear road markings, navigation systems based on vision or lidar may be greatly affected.

[0013] 4) Computational complexity and real-time performance: Some complex positioning algorithms and data processing methods may require higher computing resources, which may affect the real-time performance and response speed of the system.

[0014] 5) System integration and compatibility: Existing technologies may have deficiencies in system integration and compatibility. For example, there may be incompatible communication protocols and inconsistent data formats between sensors of different brands and models, which will increase the difficulty and cost of system integration.

[0015] 6) Lack of comprehensive perception capabilities: Relying on one or a few sensors alone may not provide comprehensive environmental perception capabilities. Although visual sensors can provide rich image information, they will fail in low light or severe occlusion; although lidar can accurately measure distance and shape, its ability to recognize materials and colors is limited. Summary of the invention

[0016] The purpose of the present invention is to provide a method and system for improving the positioning accuracy of a ROS unmanned vehicle, which can achieve high-precision positioning and improve the navigation accuracy and efficiency of the ROS vehicle.

[0017] The technical solution to achieve the purpose of the present invention is: a method for improving the positioning accuracy of a ROS unmanned vehicle, comprising:

[0018] Get BDS location information: The system obtains the current location information of the ROS unmanned vehicle through the BDS receiver; the BDS receiver receives signals from the satellite, calculates the longitude and latitude coordinates of the ROS unmanned vehicle, and transmits them to the ROS controller;

[0019] Transform the data provided by BDS: The system converts the longitude and latitude coordinate information provided by the BDS receiver into the position coordinate information in the ROS coordinate system through the coordinate conversion matrix, and publishes a topic to transmit the data to the ROS controller;

[0020] IMU data acquisition: The system uses the IMU inertial navigation system to obtain the initial position information of the ROS unmanned vehicle. The data provided by the IMU inertial navigation system includes the relative distance and attitude angle information between the ROS unmanned vehicle and the starting point, and publishes topics to transmit the data to the ROS controller.

[0021] Fusion of BDS data and IMU data: The system fuses the position data after BDS coordinate transformation and IMU data through Kalman filtering at the data fusion node to correct the position information provided by the IMU inertial navigation system and obtain the final position information.

[0022] A system for improving positioning accuracy of ROS unmanned vehicles, used to implement the above method, the system includes:

[0023] ROS unmanned vehicle, equipped with IMU inertial navigation system;

[0024] BDS receiver, used to receive signals from the Beidou satellite system and obtain location information;

[0025] The ROS controller is used to receive BDS position information and IMU sensor data, and obtain more accurate position information after fusion.

[0026] An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the steps of the above method are implemented when the processor executes the program.

[0027] A computer-readable storage medium stores a computer program, which implements the steps of the above method when executed by a processor.

[0028] A computer program product comprises a computer program, which implements the steps of the above method when executed by a processor.

[0029] Compared with the prior art, the present invention has the following significant advantages:

[0030] (1) Path planning system that comprehensively utilizes BDS navigation and sensor data: The innovation of this system lies in the organic combination of Beidou Satellite Navigation System (BDS) navigation position information and ROS car sensor data, which are comprehensively processed and utilized through the ROS controller; the environmental data provided by the sensor and the BDS navigation position information are used together for path planning to optimize the car's driving route. This comprehensive utilization can significantly improve the accuracy and efficiency of path planning.

[0031] (2) Precise positioning technology based on multi-sensor data fusion: This system realizes dynamic positioning of the precise position of the ROS car by integrating multiple sensor data (such as vision, radar, etc.) and BDS navigation information. This multi-sensor data fusion technology not only improves the accuracy and stability of the car's positioning, but also enhances its navigation capabilities in complex environments, providing a reliable solution for various application scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 Flowchart of a method to improve IMU inertial navigation positioning accuracy using BDS navigation position information.

[0033] Figure 2 Topics and node images for the ROS location package.

[0034] Figure 3 Schematic diagram of coordinate transformation.

[0035] Figure 4 This is a schematic diagram of the Kalman filter transformation simulation. DETAILED DESCRIPTION

[0036] The present invention provides a method and system for realizing accurate positioning of a ROS (Robot Operating System) car by fusing navigation position information of a global positioning system (BDS) and inertial navigation data of an IMU. The method uses BDS to obtain the initial data of the IMU inertial navigation instead of using SLAM mapping to obtain the initial coordinates when positioning the ROS car using the IMU inertial navigation. In the later IMU inertial navigation data (imu_data) node, the Kalman filter is used to add the fused data of the IMU inertial navigation sensor position data (imu_sonser_spec) and Beidou positioning data (bds_data) to the node, and the dynamic information (relative distance and attitude angle) of the IMU is integrated with the position information of the BDS, and the correction of the BDS is used to update the system state, so as to reduce the long-term accumulated error of the IMU, thereby improving the positioning accuracy and stability.

[0037] Combination Figure 1 , the method specifically comprises the following steps:

[0038] Get BDS location information: The system obtains the current location information of the car through the BDS receiver. The BDS receiver receives signals from satellites, calculates the latitude and longitude coordinates of the car, and transmits them to the ROS controller.

[0039] Transform the data provided by BDS: The system converts the longitude and latitude coordinate information provided by BDS into position coordinate information in the ROS coordinate system through the coordinate conversion matrix, and publishes a topic to transmit the data to the ROS controller.

[0040] IMU data acquisition: The system uses the IMU inertial navigation system to obtain the initial position information of the car. The data provided by the IMU inertial navigation system includes the relative distance and attitude angle information between the car and the starting point, and publishes topics to transmit the data to the ROS controller.

[0041] Fusion of BDS data and IMU data: The system fuses the position data after BDS coordinate transformation and IMU data through Kalman filtering at the data fusion node to correct the position information provided by the IMU and obtain more accurate position information.

[0042] Through the above steps, the system can realize the navigation and positioning of the ROS car based on the BDS navigation position information and the IMU inertial navigation information, thereby making the car positioning more accurate and improving the navigation accuracy and efficiency of the car. The following is a detailed description in conjunction with the accompanying drawings.

[0043] A system for improving positioning accuracy of ROS unmanned vehicles, comprising:

[0044] ROS car, equipped with IMU inertial navigation system; including:

[0045] (1) Mobile chassis: usually driven by a motor, can be wheeled or tracked, providing mobility.

[0046] Single-board computer: such as Raspberry Pi or NVIDIA Jetson, running ROS (Robot Operating System).

[0047] (2) LiDAR: used for environmental perception, obstacle detection, and map construction.

[0048] (3) Camera: used for visual navigation and identification of target objects.

[0049] (4) Ultrasonic sensor: used for close-range obstacle detection.

[0050] (5) Accelerometer: measures linear acceleration.

[0051] (6) Gyroscope: measures angular velocity and helps determine changes in direction.

[0052] (7) Magnetometer: measures the Earth's magnetic field for directional reference.

[0053] As an experimental platform, ROS integrates various sensors and control systems to support autonomous navigation, path planning and environmental interaction. The powerful communication capabilities of ROS are used to achieve data exchange and collaborative work between modules. The IMU can provide the attitude and motion status information of the car in three-dimensional space. Especially when the GPS signal is weak or invalid, the IMU can compensate the position data, realize instant dynamic positioning and motion estimation, and improve the navigation accuracy and response speed of the car.

[0054] BDS sensors and receivers, including:

[0055] (1) BDS receiver: responsible for receiving signals from the Beidou Satellite System (BDS) and obtaining location information.

[0056] (2) Antenna: used to enhance signal reception capability, usually an integrated design.

[0057] (3) Processor: Decodes the received satellite signals and converts them into coordinate and time information.

[0058] Function: Provide global positioning services to ensure that the car can obtain accurate location information. Combined with IMU data, it can be filtered and fused more efficiently to provide highly reliable information for the car's navigation and positioning.

[0059] ROS controller, used to receive BDS location information and sensor data

[0060] First, the BDS receiver obtains the current position information of the car and transmits it to the data processing node in the ROS controller. The IMU sensor detects the surrounding environment and position information and transmits the data to the data fusion node of the ROS controller. The ROS controller obtains more accurate position information after fusing the BDS position information and IMU data. This system can achieve high-precision positioning and improve the navigation accuracy and efficiency of the ROS car.

[0061] Compared with the positioning method that relies solely on BDS, this method solves the problems of single dependence, insufficient accuracy and robustness, poor environmental adaptability, computational complexity and real-time performance.

[0062] The system combining BDS and IMU can improve computing efficiency and real-time performance through algorithm optimization and hardware acceleration.

[0063] Systems combining BDS and IMU usually need to consider these factors to ensure the stability and reliability of the system.

[0064] The system combining BDS and IMU can form a more comprehensive and accurate perception capability by integrating with other sensors (such as vision, lidar, etc.).

[0065] In order to better understand the above scheme, combined with Figure 2 to Figure 4 , specifically describe the implementation steps of the method of the present invention:

[0066] Get BDS location information: The system obtains the current location information of the car and the location information of the target point through the BDS receiver. The BDS receiver receives signals from satellites, calculates the longitude and latitude coordinates of the car and the target point in the WGS84 geodetic coordinate system, and transmits them to the ROS controller.

[0067] IMU data collection: On the ROS car, IMU (inertial measurement unit) data collection is done by connecting the IMU sensor to the main control board, installing the corresponding ROS driver and starting the node to publish accelerometer, gyroscope and magnetometer data to ROS topics. Other ROS nodes subscribe to these topics to obtain real-time attitude and motion information for navigation, path planning and motion control.

[0068] Coordinate conversion: Import the map created in the previous step into the move_base navigation package, and then use the position of the car as the origin of the ROS car coordinate system we need, and assume that the latitude and longitude of the origin are (\phi_0) (latitude) and (\lambda_0)

[0069] (longitude), and convert the latitude and longitude coordinates of the origin to the coordinate values ​​in the Earth-centered Earth-fixed coordinate system (ECEF):

[0070] Import the map created in the previous step into the move_base navigation package, and then use the position of the car as the origin of the ROS car coordinate system we need, and assume that the longitude and latitude of the origin are (\phi_0) (latitude) and (\lambda_0) (longitude), and convert the longitude and latitude coordinates of the origin to the coordinate values ​​in the Earth-centered Earth-fixed coordinate system (ECEF):

[0071] [X0=(N0+·h0)·cos(φ0)·cos(λ0)][Y0=(N0+h0)·cos(φ0)·sin(λ0)][Z0=((1-e 2 )·N0+h0)·sin(φ0)]

[0072] Among them, X0, Y0, and Z0 are the coordinate values ​​of the origin in the Earth-centered Earth-fixed coordinate system, N0 is the radius of curvature of the car's circumference, h0 is the vertical distance from the car's location to the earth's surface, Φ0 is the latitude of the origin, and λ0 is the longitude of the origin.

[0073] Similarly, the coordinates of the navigation end point in the ECEF coordinate system can be obtained by the same method.

[0074] [X=(N+h)·cos(φ)·cos(λ)][Y=(N+h)·cos(φ)·sin(λ)][Z=((1-e 2 )]·N+h)·sin(φ)]

[0075] Wherein, X, Y, and Z are the coordinate values ​​of the navigation endpoint in the Earth-centered Earth-fixed coordinate system, N is the radius of curvature of the azimuth circle of the navigation endpoint, h is the vertical distance from the location of the navigation endpoint to the surface of the earth, Φ is the latitude of the navigation endpoint, and λ is the longitude of the navigation endpoint.

[0076] In this way, the offset of the navigation endpoint relative to the origin in the Earth-centered Earth-fixed coordinate system can be obtained.

[0077] According to the offset of the target point in the ECEF coordinate system, the coordinate value of the target point in the final ROS car coordinate system can be obtained by converting it to the ROS car coordinate system we need through the direction cosine matrix.

[0078] Rotation matrix:

[0079] [R=[-sin(λ0) cos(λ0) 0-sin(φ0)cos(λ0) -sin(φ0)sin(λ0) cos(φ0) cos(φ0)cos(λ0) cos(φ0)sin(λ0) sin(φ0)]]

[0080] Among them, R is the coordinate matrix of the target point in the final ROS car coordinate system.

[0081] And get the final position coordinates of the end point under the ROS car coordinates (ENU):

[0082] [E=R 11 ΔX+R 12 ΔY+R 13 ΔZ][N=R 21 ΔX+R 22 ΔY+R 23 ΔZ][U=R 31 ΔX+R 32 ΔY+R 33 ΔZ]

[0083] Among them, E, N, and U are the final end point position coordinate values ​​in the ROS car coordinate system.

[0084] Next, the obstacle position information provided by the sensor is calculated.

[0085] Through mathematical calculations of trigonometric functions, we can obtain the position coordinates of the obstacle in the ENU coordinate system: X = l*cos(a); Y = l*sin(a); Z = 0 (ROS car cannot move in the vertical direction)

[0086] Among them, l is the distance information of the car obtained by the sensor, and a is the angle information obtained by the sensor.

[0087] Among them, the relative movement distance of the car can be calculated by the following method:

[0088] [ΔX=X-X0][ΔY=Y-Y0][ΔZ=Z-Z0]

[0089] Among them, ΔX, ΔY, and ΔZ are the relative movement distances of the car in the horizontal, vertical, and vertical directions, respectively.

[0090] Kalman filter: Kalman filter is a state estimator. Here we define the speed and position of the car as state quantities, so the state quantity of the car can be expressed by the position vector of the car. and the velocity vector The equation represents that When we have the state at time k-1 Then, if the acceleration remains unchanged, we can get the state at the next moment in

[0091]

[0092] The acceleration value in the above equation can be provided by the IMU inertial navigation system, thereby obtaining the state equation of the car provided by the IMU inertial navigation information, where x k is a computationally derived value obtained by calculation.

[0093] At the same time, similar to the above analysis, the BDS sensor can also provide a state quantity about the car based on the information provided by the BDS sensor itself. The state quantity g containing the car position and speed information provided by the BDS satellite navigation information is listed. k , as the observation value in Kalman filtering, list the observation state space expression g k =H k x k +n k , where H k is the observation model, which maps the real state space of the system into the observation space, n k is the noise term, which is what we often call observation noise, and it obeys the normal distribution.

[0094] According to the Kalman filter algorithm, we first solve the prior estimate based on the state information of the car itself provided by the IMU inertial navigation system This is a formula obtained through calculation, where A is the state matrix and B is the control matrix. At the same time, according to the measurement results of BDS, the posterior estimate of the car can be expressed in is the pair x obtained from the measured value k The Kalman filter calculates the final optimal estimate based on the inaccurate prior estimate and the inaccurate measurement estimate.

[0095] The final formula for the Kalman filter is Substituting the priori estimate and the posterior estimate obtained previously into the final formula of the Kalman filter, we can get the Kalman filter gain K k In order to optimize the parameters, we need to find the most suitable K k value, so that the final estimated value tends to the true value x k The degree of this trend is expressed as error e k To indicate that

[0096] In order to achieve the above optimization goal, the error e is required k Minimum, error e k is a linear combination of normally distributed variables, so it still obeys the normal distribution, e k The probability density of k )~(0,p), its variance can be obtained from the covariance matrix (Here, e k Take two items as an example) and the final simplified result is in The prior error The covariance matrix of P k When the trace tr(p) of the covariance matrix is ​​minimized, the error e k The variance is the smallest. At this time, the error e k The trace of the covariance matrix, tr(p), is closest to the expected value 0, which meets our needs. Simplifying tr(p), we get Where R is the covariance matrix of the noise error. In order to obtain tr(p) with respect to K k To find the minimum value of a variable, we use tr(p) to calculate K k Taking the partial derivative we get when When the value is 0, tr(p) takes the minimum value, and the K at this time can be solved k Size This is the Kalman gain value we need. Based on the calculated gain, we can update the formula of the final estimate to get a more accurate estimate.

[0097] Node creation and topic publishing: In order to implement the coordinate transformation and data optimization of the Kalman filter system mentioned in the above steps in the ROS car, it is necessary to create new data processing nodes bds_data_processing and data_processing. For the coordinate transformation node bds_data_processing, it completes the processing of data provided by the BDS satellite signal, so it is necessary to subscribe to the BDS data topic published by the BDS node. Similarly, the Kalman filter processing node data_processing is a fusion optimization of the data provided by the IMU system and the BDS satellite data, so it is necessary to subscribe to the IMU inertial navigation system node imu_sonser_spec and the data topics imu_data and bds_ros_data published after the BDS coordinate transformation, so as to realize the call and processing of the two data, and finally publish the optimized positioning information topic location_data to be subscribed by subsequent nodes.

[0098] Through the above steps, the system can realize the ROS car path planning based on the BDS navigation position information, thereby realizing the intelligent navigation function and improving the navigation accuracy and efficiency of the car.

[0099] It should be understood that in order to simplify the present invention and help those skilled in the art understand the various aspects of the present invention, in the above description of the exemplary embodiments of the present invention, the various features of the present invention are sometimes described in a single embodiment or described with reference to a single figure. However, the present invention should not be interpreted as the features included in the exemplary embodiments are all necessary technical features of the patent claims.

Claims

1. A method for improving positioning accuracy of ROS unmanned vehicles, characterized in that: include: Get BDS location information: The system obtains the current location information of the ROS unmanned vehicle through the BDS receiver; The BDS receiver receives signals from the satellite, calculates the longitude and latitude coordinates of the ROS vehicle, and transmits them to the ROS controller; Transform the data provided by BDS: The system converts the longitude and latitude coordinate information provided by the BDS receiver into the position coordinate information in the ROS coordinate system through the coordinate conversion matrix, and publishes a topic to transmit the data to the ROS controller; IMU data acquisition: The system uses the IMU inertial navigation system to obtain the initial position information of the ROS unmanned vehicle. The data provided by the IMU inertial navigation system includes the relative distance and attitude angle information between the ROS unmanned vehicle and the starting point, and publishes topics to transmit the data to the ROS controller. Fusion of BDS data and IMU data: The system fuses the position data after BDS coordinate transformation and IMU data through Kalman filtering at the data fusion node to correct the position information provided by the IMU inertial navigation system and obtain the final position information.

2. The method for improving positioning accuracy of ROS unmanned vehicles according to claim 1 is characterized in that: When using the IMU inertial navigation system to locate the ROS car, the BDS satellite system is used to obtain the position coordinates, and in the IMU inertial navigation system data processing node, the dynamic information of the IMU inertial navigation system is integrated with the position information of the BDS with the help of the Kalman filter, and the system status is updated using the correction of the Kalman filter.

3. The method for improving positioning accuracy of ROS unmanned vehicles according to claim 1 is characterized in that: The dynamic information of the IMU inertial navigation system includes relative distance and attitude angle.

4. The method for improving positioning accuracy of ROS unmanned vehicles according to claim 1 is characterized in that: The BDS receiver obtains location information through the BDS satellite system, and publishes the topic bds_data containing the obtained location information in the ROS system.

5. The method for improving positioning accuracy of ROS unmanned vehicles according to claim 1 is characterized in that: The IMU inertial navigation system provides the acceleration and attitude angular acceleration information of the ROS unmanned vehicle through the gyroscope and accelerometer. After internal integration, the relative movement distance and current attitude angle of the vehicle based on the initial coordinates are obtained. The IMU inertial navigation system publishes a topic containing the relative movement distance and current attitude angle information of the vehicle based on the initial coordinates in the ROS system.

6. A system for improving positioning accuracy of ROS unmanned vehicles, characterized in that: For implementing the method described in any one of claims 1 to 5, the system comprises: ROS unmanned vehicle, equipped with IMU inertial navigation system; BDS receiver, used to receive signals from the Beidou satellite system and obtain location information; The ROS controller is used to receive BDS position information and IMU sensor data, and obtain more accurate position information after fusion.

7. The system for improving positioning accuracy of ROS unmanned vehicles according to claim 6 is characterized in that: The ROS unmanned vehicle also includes: The mobile chassis is driven by a motor to provide mobility; A single-board computer to run ROS; LiDAR for environmental perception, obstacle detection, and map building; Camera, used for visual navigation and identification of target objects; Ultrasonic sensors for close-range obstacle detection; Accelerometer, used to measure linear acceleration; A gyroscope, which measures angular velocity and determines changes in direction; Magnetometer, which measures the Earth's magnetic field for directional reference.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the method according to any one of claims 1 to 5 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images