A positioning method and positioning device for a vehicle

Abstract

The present invention provides a positioning method and positioning device for a vehicle. The positioning method includes: the vehicle is provided with a first sensor and a second sensor, and the positioning method includes: using a Kalman filter to filter the first sensor data from the first sensor to obtain and record a prediction data queue for the positioning of the vehicle; based on the second sensor data from the second sensor and the time delay of the second sensor data, the prediction data queue is corrected and fused to obtain a positioning result for the vehicle. The positioning device includes a filtering module, a correction module, a judgment module and a data removal module. The present invention solves the technical problem that during the positioning process of the vehicle, the sensor has no way to process and fuse the delay of the collected data, resulting in the sensor being unable to accurately locate and identify the motion state and position of the vehicle.

Description

Technical Field

[0001] The present invention relates to the technical field of communications, and in particular to a positioning method and a positioning device for a vehicle. Background Art

[0002] With the advancement of modern technology and scientific development, people's requirements for vehicles are becoming increasingly higher. Beyond meeting traditional driving requirements, vehicles also require intelligent control, such as driverless cars, adaptive cruise control, active safety features, and fully automated parking. Sensor modules for environmental and situational awareness (such as lidar, millimeter-wave radar, and cameras) are crucial components of intelligent control. Processing sensor data and correcting and fusing data collected by multiple sensors are crucial. Therefore, positioning requires both real-time positioning solution output and optimal state estimation.

[0003] However, in the prior art, there is no method to process and fuse the data delay, resulting in the sensor being unable to accurately locate and identify the motion state and position of the vehicle. Summary of the Invention

[0004] The present invention solves the technical problem that during the positioning process of a vehicle, the sensor has no way to process and fuse the collected data to correct the delay, resulting in the sensor being unable to accurately locate and identify the motion state and position of the vehicle, thereby achieving the technical effect of real-time accurate positioning and state identification of the vehicle.

[0005] In order to solve the above problems, the present invention provides a positioning method and a positioning device for a vehicle.

[0006] In a first aspect, the present invention provides a positioning method for a vehicle, wherein the vehicle is provided with a first sensor and a second sensor, and the positioning method comprises: using a Kalman filter to filter first sensor data from the first sensor to obtain and record a prediction data queue for the positioning of the vehicle; and correcting and fusing the prediction data queue based on second sensor data from the second sensor and the degree of time delay of the second sensor data to obtain a positioning result for the vehicle.

[0007] Compared with existing technologies, this technical solution achieves the following technical effects: a vehicle is equipped with a first sensor and a second sensor. Furthermore, the first sensor can be an IMU sensor, which can meet real-time prediction requirements and maintain accuracy in a short timeframe, while also offering short processing times and acceptable latency. The second sensor can be of various types, such as vision, laser, GNSS, or wheel speedometers, and there can be multiple second sensors. The positioning method includes: after collecting first sensor data, the first sensor performs filtering processing using a Kalman filter to generate a prediction data queue, and then correcting and fusing the first sensor data based on the time delay of the second sensor data to obtain a positioning result. Kalman filtering technology is widely used and utilizes a decay function model as a matrix. The exponentiation of the matrix represents a recursive function, ultimately resulting in a type of logarithmic difference equation. The optimal solution of the difference equation can be predicted using the covariance. Using a Kalman filter to filter the first sensor data and fusing and correcting the first sensor data with the second sensor data allows for more accurate positioning of the vehicle and identification of its status.

[0008] In one example of the present invention, the prediction data queue is corrected and fused based on the second sensor data from the second sensor and the degree of time delay of the second sensor data, including: determining whether the second sensor data has a time delay; if the second sensor data has a time delay, correcting and fusing the oldest data in the prediction data queue based on the second sensor data; if the second sensor data does not have a time delay, correcting and fusing the latest data in the prediction data queue based on the second sensor data.

[0009] Compared with the existing technology, the technical effect achieved by adopting this technical solution is: before correcting and fusing the second sensor data of the second sensor, it is necessary to first judge whether there is a time delay in the second sensor data, and obtain two different results by judging whether there is a delay in different second sensor data. Two different fusion correction methods are used to improve the efficiency of the correction and fusion processing of the prediction data queue, and orderly judge and process the second sensor data.

[0010] In one example of the present invention, determining whether there is a time delay in the second sensor data includes: when the second sensor generates the second sensor data, adding the second sensor data to an observation data priority queue; comparing the length of the observation data priority queue with a delay time threshold to obtain a comparison result; and determining whether there is a time delay in the second sensor data based on the comparison result.

[0011] Compared with the existing technology, the technical effect achieved by adopting this technical solution is: judging whether there is a delay in the second sensor data by comparing whether the length of the priority queue of the observed data reaches the time threshold required for fusion, unifying the standard for data delay judgment, and effectively judging the second sensor data.

[0012] In one example of the present invention, based on the comparison result, it is determined whether the second sensor data has a time delay, including: when the length of the observation data priority queue reaches the delay time threshold, determining that the second sensor data has a time delay; when the length of the observation data priority queue does not reach the delay time threshold, determining that the second sensor data does not have a time delay.

[0013] Compared with the existing technology, the technical effect achieved by adopting this technical solution is: by judging whether the length of the observation data priority queue reaches the time threshold, the delay of the second sensor data can be determined, which can speed up the judgment efficiency of the data.

[0014] In an example of the present invention, after correcting and fusing the oldest data in the prediction data queue according to the second sensing data, the positioning method further includes: removing the oldest data and removing useless prediction data queues in the prediction data queue.

[0015] Compared with the existing technology, the technical effect achieved by adopting this technical solution is: by fusing and correcting the prediction data queue, the latest fused correction data is obtained, and the oldest data and the useless data queue in the prediction data queue are removed, and the fused correction data is transmitted to the current moment, thereby realizing real-time accurate positioning and status identification of the vehicle.

[0016] On the other hand, the present invention also provides a positioning device for a vehicle, wherein the vehicle is provided with a first sensor and a second sensor, and the positioning device includes: a filtering module, the filtering module is used to use a Kalman filter to filter the first sensor data from the first sensor to obtain and record a prediction data queue for the positioning of the vehicle; a correction module, the correction module is used to correct and fuse the prediction data queue based on the second sensor data from the second sensor and the time delay degree of the second sensor data to obtain a positioning result for the vehicle.

[0017] In one example of the present invention, the correction module is specifically used to: determine whether there is a time delay in the second sensor data; if there is a time delay in the second sensor data, correct and fuse the oldest data in the prediction data queue based on the second sensor data; if there is no time delay in the second sensor data, correct and fuse the latest data in the prediction data queue based on the second sensor data.

[0018] In one example of the present invention, to determine whether there is a time delay in the second sensor data, the positioning device includes: a judgment module, wherein the judgment module is used to add the second sensor data to an observation data priority queue when the second sensor generates the second sensor data; compare the length of the observation data priority queue with a delay time threshold to obtain a comparison result; and determine whether there is a time delay in the second sensor data based on the comparison result.

[0019] In one example of the present invention, based on the comparison result, it is determined whether the second sensor data has a time delay, and the judgment module is specifically used to: when the length of the observation data priority queue reaches the delay time threshold, determine that the second sensor data has a time delay; when the length of the observation data priority queue does not reach the delay time threshold, determine that the second sensor data does not have a time delay.

[0020] In one embodiment of the present invention, after correcting and fusing the oldest data in the prediction data queue according to the second sensing data, the positioning device further includes: a data removal module, which is used to remove the oldest data and remove useless prediction data queues in the prediction data queue.

[0021] Compared with the existing technology, the technical effect achieved by adopting this technical solution is: the positioning device can achieve the technical effect corresponding to any technical solution in the above-mentioned positioning method, which will not be repeated here.

[0022] In summary, after adopting the technical solution of the present invention, the following technical effects can be achieved:

[0023] (1) Filtering the first sensor data of the first sensor through the Kalman filter, and recording the prediction data queue, and adopting the method of coordinating the first sensor and the second sensor, using the second sensor data of the second sensor to fuse and correct the prediction data queue, obtain the latest prediction data, and achieve the effect of precise positioning;

[0024] (2) Comparing the length of the observation data priority queue with the time threshold to determine whether the second sensor data has reached the delay can speed up the determination of whether the data is delayed and make an effective judgment on the second sensor data;

[0025] (3) Two different processing methods are used for the case where there is a time delay in the second sensor data: for the second sensor data with a time delay, the oldest data in the prediction data queue is corrected and fused according to the second sensor data; for the second sensor data without a time delay, the latest data in the prediction data queue is corrected and fused according to the second sensor data; the use of two different correction methods can improve the processing efficiency of the correction and fusion of the prediction data queue. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 A schematic flow chart of a method for positioning a vehicle provided in Example 1 of the present invention.

[0027] Figure 2 for Figure 1 Schematic diagram of the specific steps in . DETAILED DESCRIPTION

[0028] To make the above-mentioned objects, features and advantages of the present invention more clearly understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0029] [First embodiment]

[0030] See also Figure 1 , which is a flow chart of a positioning method for a vehicle provided by the first embodiment of the present invention.

[0031] In a specific embodiment, a vehicle is provided with a first sensor and a second sensor, and the positioning method includes: using a Kalman filter to filter the first sensor data from the first sensor to obtain and record a prediction data queue for the positioning of the vehicle; based on the second sensor data from the second sensor and the time delay degree of the second sensor data, the prediction data queue is corrected and fused to obtain a positioning result for the vehicle.

[0032] Specifically, a vehicle is equipped with a first sensor and a second sensor. Furthermore, the first sensor may be an IMU sensor, which can meet real-time prediction requirements and ensure accuracy in a short timeframe, while also offering short processing times and acceptable latency. The second sensor may be of various types, such as vision, laser, GNSS, or wheel speedometers, and there may be multiple second sensors. The positioning method includes: after collecting first sensor data, the first sensor performs filtering processing using a Kalman filter to generate a prediction data queue. The first sensor data is then corrected and fused based on the time delay of the second sensor data to obtain a positioning result. Kalman filtering technology is widely used and utilizes a decay function model as a matrix. The exponentiation of the matrix represents a recursive function, ultimately resulting in a logarithmic difference equation. The optimal solution of the difference equation can be predicted using the covariance. Using a Kalman filter to filter the first sensor data and fusing and correcting the first sensor data with the second sensor data allows for more accurate positioning of the vehicle and identification of its status.

[0033] In today's society, vehicles, such as cars, trucks, high-speed trains, and ships, have become indispensable tools in daily life. A first sensor and a second sensor are installed on these vehicles. These two sensors work together to more accurately locate the vehicle. For example, when a truck is moving on a highway, the onboard IMU sensor collects first sensor data. After collecting this first sensor data, it is filtered using a Kalman filter to generate a prediction data queue. This first sensor data is then fused and corrected based on the time delay of the second sensor data obtained by the second sensor. The final corrected sensor data is output as the most recent, accurate data, used to determine the truck's location.

[0034] In a specific embodiment, the prediction data queue is corrected and fused based on the second sensor data from the second sensor and the degree of time delay of the second sensor data, including: determining whether the second sensor data has a time delay; if the second sensor data has a time delay, correcting and fusing the oldest data in the prediction data queue based on the second sensor data; if the second sensor data does not have a time delay, correcting and fusing the latest data in the prediction data queue based on the second sensor data.

[0035] Specifically, before correcting and fusing the second sensor data of the second sensor, it is necessary to first determine whether there is a time delay in the second sensor data. There are two different processing methods for the two different results obtained from the judgment. Among them, when there is a time delay in the second sensor data, the oldest data in the prediction data queue is corrected and fused according to the second sensor data. Because there may be multiple second sensors, each sensor has a different time delay, which will produce a time error, it is necessary to use the oldest data in the same timestamp to correct and fuse the prediction data queue; when there is no time delay in the second sensor data, the latest data in the timestamp can be used to correct and fuse the prediction data queue. Using two different fusion correction methods for whether different second sensor data have delays can improve the efficiency of the correction and fusion processing of the prediction data queue and orderly judge and process the second sensor data.

[0036] In a specific embodiment, determining whether there is a time delay in the second sensor data includes: when the second sensor generates the second sensor data, adding the second sensor data to an observation data priority queue; comparing the length of the observation data priority queue with a delay time threshold to obtain a comparison result; and determining whether there is a time delay in the second sensor data based on the comparison result.

[0037] Specifically, to determine whether the second sensor data is delayed, the second sensor data is added to the observation data priority queue and the length of the observation data priority queue is compared with the delay time threshold to obtain a result. Based on the comparison result, whether the length of the observation data priority queue reaches the time threshold required for fusion is determined to determine whether the second sensor data is delayed. This standard for data delay determination is unified, allowing for effective judgment of the second sensor data.

[0038] In a specific embodiment, based on the comparison result, it is determined whether the second sensor data has a time delay, including: when the length of the observation data priority queue reaches the delay time threshold, it is determined that the second sensor data has a time delay; when the length of the observation data priority queue does not reach the delay time threshold, it is determined that the second sensor data does not have a time delay.

[0039] Specifically, the steps for determining the second sensor data are as follows: when the length of the observation data priority queue reaches a delay time threshold, the second sensor data is determined to be delayed; when the length of the observation data priority queue does not reach the delay time threshold, the second sensor data is determined to be free of time delay. By determining whether the length of the observation data priority queue reaches the time threshold to determine whether the second sensor data is delayed, the efficiency of data determination can be improved.

[0040] In a specific embodiment, after correcting and fusing the oldest data in the prediction data queue according to the second sensing data, the positioning method further includes: removing the oldest data and removing useless prediction data queues in the prediction data queue.

[0041] Specifically, after the predicted data queue is corrected and fused based on the second sensor data from the second sensor, the oldest data is removed, and useless predicted data queues are removed from the predicted data queue. The latest fused and corrected data is retained and transmitted to the current moment. By fusing and correcting the predicted data queue, the latest fused and corrected data is obtained, and the oldest data and useless data queues from the predicted data queue are removed. The fused and corrected data is transmitted to the current moment, achieving real-time, accurate positioning and status identification of the vehicle.

[0042] [Second embodiment]

[0043] A second embodiment of the present invention further provides a positioning device for a vehicle. The vehicle is provided with a first sensor and a second sensor. The positioning device includes a filtering module, a correction module, a judgment module, and a data removal module.

[0044] In a specific embodiment, the filtering module is configured to use a Kalman filter to filter the first sensing data from the first sensor to obtain and record a prediction data queue for positioning of the vehicle;

[0045] In a specific embodiment, the correction module is used to correct and fuse the prediction data queue according to the second sensing data from the second sensor and the time delay degree of the second sensing data to obtain a positioning result for the vehicle.

[0046] The correction module is specifically used to: determine whether there is a time delay in the second sensor data; if there is a time delay in the second sensor data, correct and fuse the oldest data in the prediction data queue according to the second sensor data; if there is no time delay in the second sensor data, correct and fuse the latest data in the prediction data queue according to the second sensor data.

[0047] In a specific embodiment, the judgment module is used to add the second sensor data to the observation data priority queue when the second sensor generates the second sensor data; compare the length of the observation data priority queue with the delay time threshold to obtain a comparison result; and determine whether there is a time delay in the second sensor data based on the comparison result.

[0048] The judgment module is specifically used to: when the length of the observation data priority queue reaches the delay time threshold, judge that there is a time delay in the second sensor data; when the length of the observation data priority queue does not reach the delay time threshold, judge that there is no time delay in the second sensor data.

[0049] In a specific embodiment, after correcting and fusing the oldest data in the prediction data queue according to the second sensor data, the positioning device further includes: a data removal module, the data removal module is used to remove the oldest data and remove useless prediction data queues in the prediction data queue

[0050] Specifically, a positioning device for a vehicle provided by an embodiment of the present invention can implement the positioning method of any embodiment of the present invention, and thus also has all the beneficial effects of the positioning method of any embodiment of the present invention.

[0051] Specifically, the vehicle's primary sensor, such as an IMU, collects primary sensor data. After the primary sensor data is collected, the filtering module uses a Kalman filter to filter the data, obtaining and recording a prediction data queue for the vehicle's positioning. Secondly, the correction module corrects and fuses the prediction data queue based on secondary sensor data obtained from secondary sensors, such as vision, laser, GNSS, and wheel speedometers, as well as the time delay of the secondary sensor data, to obtain a positioning result for the vehicle.

[0052] Furthermore, the correction module can be used to determine whether there is a delay in the second sensor data. For the second sensor data with a delay, the oldest data in the prediction data queue is used to correct and fuse the second sensor data; for the second sensor data without a delay, the latest data in the prediction data queue is used to correct and fuse the second sensor data.

[0053] Furthermore, the judgment module is used to determine whether there is a time delay in the second sensor data. The judgment module is specifically used to: when the length of the observation data priority queue reaches the delay time threshold, determine that there is a time delay in the second sensor data; when the length of the observation data priority queue does not reach the delay time threshold, determine that there is no time delay in the second sensor data.

[0054] Furthermore, after correcting and fusing the oldest data in the prediction data queue according to the second sensing data, the data removal module removes the oldest data and removes useless prediction data queues in the prediction data queue.

[0055] After the above steps are completed, the fused correction data is transmitted to the current moment to achieve real-time accurate positioning and status identification of the vehicle.

[0056] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for positioning a vehicle, characterized in that: The vehicle is provided with a first sensor and a second sensor, and the positioning method includes: Using a Kalman filter, filtering the first sensing data from the first sensor to obtain and record a prediction data queue for the positioning of the vehicle; performing correction and fusion on the predicted data queue according to the second sensing data from the second sensor and the time delay of the second sensing data to obtain a positioning result for the vehicle; In a case where the second sensor generates the second sensing data, adding the second sensing data to an observation data priority queue; Comparing the length of the observation data priority queue with a delay time threshold to obtain a comparison result; determining, based on the comparison result, whether there is a time delay in the second sensing data; In the case where there is a time delay in the second sensor data, correcting and fusing the oldest data in the prediction data queue according to the second sensor data; In the case that there is no time delay in the second sensing data, the latest data in the prediction data queue is corrected and fused according to the second sensing data.

2. The positioning method according to claim 1, wherein: The determining, based on the comparison result, whether the second sensing data has a time delay includes: When the length of the observation data priority queue reaches the delay time threshold, determining that there is a time delay in the second sensor data; When the length of the observation data priority queue does not reach the delay time threshold, it is determined that there is no time delay in the second sensing data.

3. The positioning method according to claim 1, wherein: After correcting and fusing the oldest data in the prediction data queue based on the second sensing data, the positioning method further includes: The oldest data is removed, and useless prediction data queues in the prediction data queue are removed.

4. A positioning device for a vehicle, characterized in that: The vehicle is provided with a first sensor and a second sensor, and the positioning device comprises: a filtering module configured to filter the first sensing data from the first sensor using a Kalman filter to obtain and record a prediction data queue for positioning the vehicle; A correction module is used to correct and fuse the predicted data queue according to the second sensor data from the second sensor and the time delay degree of the second sensor data to obtain a positioning result for the vehicle.

5. The positioning device according to claim 4, characterized in that The correction module is specifically used for: determining whether there is a time delay in the second sensing data; In the case where there is a time delay in the second sensor data, correcting and fusing the oldest data in the prediction data queue according to the second sensor data; In the case that there is no time delay in the second sensing data, the latest data in the prediction data queue is corrected and fused according to the second sensing data.

6. The positioning device according to claim 5, characterized in that The determining whether the second sensing data has a time delay, the positioning device comprising: a judgment module, configured to add the second sensor data to an observation data priority queue when the second sensor generates the second sensor data; Comparing the length of the observation data priority queue with a delay time threshold to obtain a comparison result; According to the comparison result, it is determined whether the second sensing data has a time delay.

7. The positioning device according to claim 6, characterized in that The determining module is specifically configured to determine whether there is a time delay in the second sensing data based on the comparison result: When the length of the observation data priority queue reaches the delay time threshold, determining that there is a time delay in the second sensor data; When the length of the observation data priority queue does not reach the delay time threshold, it is determined that there is no time delay in the second sensing data.

8. The positioning device according to claim 5, characterized in that After correcting and fusing the oldest data in the prediction data queue according to the second sensing data, the positioning device further includes: A data removal module is used to remove the oldest data and remove useless prediction data queues in the prediction data queue.

Images