Sensor parameter calibration method and device, computer program product and vehicle

Abstract

The embodiment of the invention provides a sensor parameter calibration method and device, a computer program product and a vehicle, and relates to the technical field of sensors, the method comprises the steps that target load information and relative motion data corresponding to the vehicle are acquired, the target load information is used for representing the load condition corresponding to at least part of the area in the vehicle, and the relative motion data is used for representing the load condition corresponding to at least part of the area in the vehicle; the relative motion data is used for representing position change information of the target sensor relative to the target object in the driving process of the vehicle; determining a target mapping relation corresponding to the target load according to the target load information; and determining a target parameter of the target sensor according to the relative motion data and the target mapping relation. The technical problem that it is difficult to balance the efficiency and correctness of sensor parameter calibration in the prior art is solved.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical field of sensors, and in particular, to a method and device for calibrating sensor parameters, a computer program product, and a vehicle. Background Art

[0002] The parameters of a sensor reflect the relative position and attitude relationship of the sensor in space. Affected by external factors (such as long-term use, collision, etc.), the parameters of the sensor may change, affecting the performance of the sensor. At this time, it is necessary to recalibrate the parameters of the sensor.

[0003] In the related art, manual operation is usually relied on. The sensor is placed in a standard calibration site, and a calibration board is used to recalibrate the sensor parameters. Although this method can ensure the correctness of the parameters, it consumes a large amount of labor costs and time costs. Although the method for automatically calibrating sensor parameters in the related art can reduce costs, the correctness of the calibrated parameters is low.

[0004] In view of the above problems, no effective solution has been proposed yet. Summary of the Invention

[0005] The embodiments of the present application provide a method and device for calibrating sensor parameters, a computer program product, and a vehicle, aiming to improve the problem that it is difficult to balance the efficiency and correctness of sensor parameter calibration in the related art.

[0006] According to one aspect of the embodiments of the present application, a method for calibrating sensor parameters is provided, including: obtaining target load information and relative motion data corresponding to a vehicle, where the target load information is used to characterize the load condition corresponding to at least part of the area in the vehicle, and the relative motion data is used to characterize the position change information of a target sensor relative to a target object during the driving process of the vehicle; determining a target mapping relationship corresponding to the target load according to the target load information, where the target mapping relationship characterizes the corresponding relationship between the calibrated motion data of the vehicle and the calibrated parameters of the target sensor, different target loads correspond to different target mapping relationships, and the calibrated motion data is used to characterize the position change information of the target sensor relative to a test calibration object during the process of a parameter calibration test; and determining the target parameters of the target sensor according to the relative motion data and the target mapping relationship.

[0007] The above-mentioned sensor parameter calibration method provided by the embodiments of the present application achieves the following technical effects: First, according to the load conditions corresponding to at least some areas in the vehicle, the target load information corresponding to the vehicle is obtained in real time to ensure the timeliness of the target load information. According to the position change information of the target sensor relative to the target object, relative motion data is obtained, providing data support for correctly calibrating the pose parameters of the sensor in the follow-up; by conducting parameter calibration tests on the target sensor under various target load conditions to determine the calibration motion data, establishing the corresponding relationship between the calibration motion data of the vehicle and the calibration parameters of the target sensor, obtaining the target mapping relationships corresponding to various target loads respectively. Furthermore, according to the target load information, the target mapping relationship corresponding to the target load is determined, and the target mapping relationship of the target sensor under the target load information can be determined more accurately; further, according to the relative motion data and the target mapping relationship, the target parameters of the target sensor are automatically determined, reducing the dependence on manual operations and improving the calibration efficiency of the sensor parameters. Thus, the present application determines the target mapping relationship corresponding to the target load according to the target load information, and according to the relative motion data and the target mapping relationship, achieves the purpose of automatically and more accurately calibrating the parameters of the target sensor, thereby realizing the technical effects of improving the efficiency and correctness of sensor parameter calibration, and further solving the technical problem in the related art that it is difficult to balance the efficiency and correctness of sensor parameter calibration.

[0008] According to another aspect of the embodiments of the present application, there is also provided a sensor parameter calibration device, including: an acquisition module, configured to acquire the target load information and relative motion data corresponding to the vehicle, where the target load information is used to characterize the load conditions corresponding to at least some areas in the vehicle, and the relative motion data is used to characterize the position change information of the target sensor relative to the target object during the driving process of the vehicle; a determination module, configured to determine the target mapping relationship according to the target load information and the calibration data set, where the calibration data set is obtained by conducting parameter calibration tests on the target sensor under various load conditions; a calibration module, configured to calibrate the parameters of the target sensor by using the relative motion data and the target mapping relationship.

[0009] According to another aspect of the embodiments of the present application, there is also provided a computer program product, including a computer program, where when the computer program is executed by a processor, it implements the sensor parameter calibration method in any one of the above.

[0010] According to another aspect of the embodiments of the present application, there is also provided a vehicle, including an in-vehicle processor and an in-vehicle memory, where the in-vehicle memory is configured to store a computer program; the in-vehicle processor is configured to execute the computer program stored on the memory to implement the sensor parameter calibration method in any one of the above. Description of the Drawings

[0011] The accompanying drawings described herein are used to provide a further understanding of the present application and form a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation to the present application. In the drawings:

[0012] Figure 1 is a flowchart of a sensor parameter calibration method provided by one embodiment of the present application;

[0013] Figure 2 is a schematic diagram of an optional sensor parameter calibration method provided by one embodiment of the present application;

[0014] Figure 3 is a schematic diagram of an optional verification of the pose parameters of a target sensor provided by one embodiment of the present application;

[0015] Figure 4 is a structural block diagram of a sensor parameter calibration device provided by one embodiment of the present application;

[0016] Figure 5 is a structural block diagram of a vehicle provided by one embodiment of the present application;

[0017] Figure 6 is a hardware structural block diagram of a computing terminal for implementing the sensor parameter calibration method provided by one embodiment of the present application;

[0018] Figure 7 is a structural block diagram of an electronic device provided by one embodiment of the present application. Detailed Embodiments

[0019] In order to make the technical problems, technical solutions and beneficial effects solved by the present application more clear and understandable, the present application will be further described in detail below in conjunction with embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0020] A sensor parameter calibration method provided by an embodiment of the present application includes: obtaining target load information and relative motion data corresponding to a vehicle, where the target load information is used to characterize the load condition corresponding to at least part of the vehicle, and the relative motion data is used to characterize the position change information of a target sensor relative to a target object during the driving process of the vehicle; determining a target mapping relationship corresponding to the target load according to the target load information, where the target mapping relationship characterizes the corresponding relationship between the calibrated motion data of the vehicle and the calibrated parameters of the target sensor, different target loads correspond to different target mapping relationships, and the calibrated motion data is used to characterize the position change information of the target sensor relative to a test calibration object during the parameter calibration test; and determining the target parameters of the target sensor according to the relative motion data and the target mapping relationship.

[0021] The above-mentioned sensor parameter calibration method provided by the embodiments of the present application achieves the following technical effects: First, according to the load conditions corresponding to at least some areas in the vehicle, the target load information corresponding to the vehicle is obtained in real time to ensure the timeliness of the target load information. According to the position change information of the target sensor relative to the target object, relative motion data is obtained, which provides data support for correctly calibrating the pose parameters of the sensor in the subsequent process; by conducting parameter calibration tests on the target sensor under various target load conditions to determine the calibration motion data, and establishing the corresponding relationship between the calibration motion data of the vehicle and the calibration parameters of the target sensor, the target mapping relationships corresponding to various target loads are obtained. Furthermore, according to the target load information, the target mapping relationship corresponding to the target load is determined, and the target mapping relationship of the target sensor under the target load information can be determined more accurately; further, according to the relative motion data and the target mapping relationship, the target parameters of the target sensor are automatically determined, reducing the dependence on manual operations and improving the calibration efficiency of the sensor parameters. Thus, the present application determines the target mapping relationship corresponding to the target load according to the target load information, and according to the relative motion data and the target mapping relationship, achieves the purpose of automatically and more accurately calibrating the parameters of the target sensor, thereby realizing the technical effects of improving the efficiency and correctness of sensor parameter calibration, and further solving the technical problem in the related art that it is difficult to balance the efficiency and correctness of sensor parameter calibration.

[0022] Embodiment 1

[0023] The embodiments of the present application provide a sensor parameter calibration method. Please refer to Figure 1 , including the following steps:

[0024] S110: Obtain the target load information and relative motion data corresponding to the vehicle, where the target load information is used to characterize the load conditions corresponding to at least some areas in the vehicle, and the relative motion data is used to characterize the position change information of the target sensor relative to the target object during the driving process of the vehicle;

[0025] S120: Determine the target mapping relationship corresponding to the target load according to the target load information, where the target mapping relationship characterizes the corresponding relationship between the calibration motion data of the vehicle and the calibration parameters of the target sensor, and different target loads correspond to different target mapping relationships. The calibration motion data is used to characterize the position change information of the target sensor relative to the test calibration object during the process of the parameter calibration test;

[0026] S130: Determine the target parameters of the target sensor according to the relative motion data and the target mapping relationship.

[0027] The above-mentioned target load information can be obtained by using weight sensors, which can be installed in multiple load areas of the vehicle (such as the driver's seat area, passenger seat area, cargo hold area, etc.). The target load information can include, but is not limited to: the load of the driver's seat, the load of the passenger seat, and the load of the cargo hold. Select some areas from the multiple load areas in the vehicle, and use the weight sensors installed in each load area of this part of the area to measure the load corresponding to the load area, so as to obtain the target load information corresponding to some areas in the vehicle in real time.

[0028] It should be noted that multiple weight sensors can be installed in each load area among the above-mentioned multiple load areas. Each weight sensor among the multiple weight sensors installed in the load area is used to measure the load measurement information corresponding to the weight sensor, and the load measurement information corresponding to each weight sensor in the load area is weighted and averaged to obtain the target load information corresponding to the load area.

[0029] The above-mentioned target sensors can include, but are not limited to: vision sensors, laser sensors, radar sensors, thermal imaging sensors. The above-mentioned target object can be a reference object for calibrating sensor parameters. The target object can include, but is not limited to: traffic signs, traffic lights, road static obstacles (such as pedestrian guardrails, trees, street lights, etc.).

[0030] The above-mentioned relative motion data can at least include a distance change amount and an azimuth change amount. The distance change amount can be used to characterize the distance change from the target sensor to the target object. The azimuth change amount can be used to characterize the azimuth change from the target sensor to the target object. The relative motion data can also include, but is not limited to: relative speed change amount, relative acceleration change amount. During the driving process of the vehicle, the target sensor is used to automatically collect information multiple times to obtain multiple distance information of the target sensor relative to the target object and multiple azimuth information of the target sensor relative to the target object. Based on the multiple distance information and multiple azimuth information collected, the relative motion data is calculated.

[0031] The above-mentioned parameter calibration test can refer to the process of setting multiple load conditions (such as multiple loads of the driver's seat) during the vehicle production stage, and recording the calibration parameters of the target sensor and the position change information of the target sensor relative to the test calibration object under each load condition among the multiple load conditions. Through this parameter calibration test, the mapping relationship corresponding to each load condition among the multiple load conditions can be obtained. The above-mentioned load conditions can be set according to actual needs. The load conditions can at least include the load of the driver's seat. The load conditions can also include, but are not limited to: the load of the passenger seat, the load of the cargo hold.

[0032] The calibration data set obtained by performing parameter calibration tests on the target sensor under various load conditions is stored in the sensor calibration system. The sensor calibration system reads the target load information obtained in real time by the weight sensor, automatically selects the mapping relationship corresponding to the target load information from the calibration data set, and uses this mapping relationship as the target mapping relationship. The accuracy of the obtained target mapping relationship is relatively high. Further, using the relative motion data and the target mapping relationship, the target parameters of the target sensor are automatically determined, and thus the target sensor is automatically calibrated using the target parameters, improving the efficiency and correctness of calibrating the target sensor.

[0033] The sensor parameter calibration method provided in the embodiment of the present application achieves the following technical effects: First, according to the load conditions corresponding to at least some areas in the vehicle, the target load information corresponding to the vehicle is obtained in real time to ensure the timeliness of the target load information. According to the position change information of the target sensor relative to the target object, relative motion data is obtained, providing data support for correctly calibrating the pose parameters of the sensor in the follow-up. By performing parameter calibration tests on the target sensor under various target load conditions to determine the calibration motion data, the corresponding relationship between the calibration motion data of the vehicle and the calibration parameters of the target sensor is established, and the target mapping relationships corresponding to various target loads are obtained. Furthermore, according to the target load information, the target mapping relationship corresponding to the target load is determined, and the target mapping relationship of the target sensor under the target load information can be determined more accurately. Further, according to the relative motion data and the target mapping relationship, the target parameters of the target sensor are automatically determined, reducing the dependence on manual operations and improving the calibration efficiency of the sensor parameters. Thus, the present application determines the target mapping relationship corresponding to the target load according to the target load information, and according to the relative motion data and the target mapping relationship, achieves the purpose of automatically and more accurately calibrating the target sensor, thereby realizing the technical effect of improving the efficiency and correctness of sensor parameter calibration, and further solving the technical problem in the related art that it is difficult to balance the efficiency and correctness of sensor parameter calibration.

[0034] The sensor parameter calibration method provided in the embodiment of the present application determines the target mapping relationship corresponding to the target load according to the target load information, and uses the relative motion data and the target mapping relationship to automatically determine the target parameters of the target sensor for calibrating the target sensor, and can be widely applied to multiple preset application scenarios.

[0035] For example, in the intelligent driving scenario, a household car obtains the target load information and relative motion data through built-in load detection devices (such as seat weight sensors) and various vehicle-mounted sensors (such as cameras, radars, lidar, etc.), can accurately determine the target mapping relationship corresponding to the target load, automatically determine the target parameters of the target sensor, and thus automatically complete the parameter calibration of the target sensor.

[0036] For example, in the scenario of intelligent logistics distribution, intelligent driving trucks or driverless trucks use intelligent driving technology for automatic navigation and distribution. They require accurate load information and relative motion data to ensure that when loading goods of different weights, they can correctly determine the target parameters of the target sensor, ensure the sensor performance, and thus drive stably and accurately identify obstacles and destinations in the distribution route.

[0037] For example, in the scenario of UAV flight, when the pose parameters of the UAV sensor change, correspondingly, the target load information is used to characterize the load situation corresponding to at least part of the area in the UAV, and the relative motion data is used to characterize the position change information of the target sensor relative to the target object during the flight of the UAV. Obtain the target load information and relative motion data corresponding to the UAV, determine the target mapping relationship corresponding to the target load according to the target load information, and automatically determine the target parameters of the target sensor according to the relative motion data and the target mapping relationship to ensure the performance of the UAV sensor, so as to ensure that the UAV can fly smoothly.

[0038] The sensor parameter calibration method provided by the embodiments of the present application can be but is not limited to being applied to the above-listed application scenarios. With the continuous evolution of technology, the above method can also be applied to a wider range of scenarios, such as the application scenario of parameter calibration of intelligent devices in construction, the application scenario of parameter calibration of camera devices in security systems, etc. By efficiently calibrating the parameters of the device, it can support various advanced functions and applications of the device, improve the calibration efficiency of the device parameters, reduce the maintenance cost of the device, and thus improve the overall performance of the device.

[0039] Optionally, in the above step S110, obtaining the target load information includes the following steps:

[0040] S111: In response to the power-on event of the vehicle, use the driver's seat weight sensor installed in the vehicle to obtain the target load information, where the target load information is used to characterize the load corresponding to the driver's seat area in the vehicle.

[0041] The above power-on event can be used to activate the driver's seat weight sensor. This power-on event can be triggered during the vehicle's power-on process. When the start button is pressed, the vehicle's power system is activated, triggering the vehicle's power-on event. Furthermore, in response to the vehicle's power-on event, the vehicle's electronic control system is activated to control the operation of the vehicle controller unit and in-vehicle sensors (especially the driver's seat weight sensor installed in the vehicle).

[0042] The above driver seat weight sensor can be assembled under the driver's seat of the vehicle or in the support structure. The driver seat weight sensor can be used to measure the load in the driver seat area in real time. The driver seat weight sensor can include, but is not limited to, a pressure sensor, a strain gauge sensor, or a capacitive sensor. In particular, when a pressure sensor is selected as the driver seat weight sensor, the driver seat weight sensor is used to measure the main driver load according to the pressure change of the seat when the driver sits in the driver seat area, so as to obtain the target load information in real time.

[0043] It should be noted that when the driverless function is in the on state, the above target load information can be 0 (that is, the driver seat area is in an unmanned state). In response to the driver seat weight sensor not detecting a pressure change in the seat and the sensor calibration system detecting that the driverless function is in the on state, it is determined that the target load information is 0.

[0044] In an exemplary application scenario, Figure 2 is a schematic diagram of an optional sensor parameter calibration method provided by an embodiment of the present application. As Figure 2 shown, during the vehicle power-on process, the vehicle power-on event is triggered. In response to the vehicle power-on event, the driver seat weight sensor assembled in the vehicle is used to measure the main driver load corresponding to the driver seat area, and the target load information is obtained in real time.

[0045] The above optional embodiment of the present application can achieve the following technical effects: In response to the vehicle power-on event, the driver seat weight sensor assembled in the vehicle can be used to obtain the target load information in real time, ensure the timeliness of the target load information, improve the accuracy of the target load information, and provide accurate data support for obtaining the correct target mapping relationship subsequently.

[0046] Optionally, in the above step S110, obtaining the relative motion data includes the following steps:

[0047] S112: In response to the vehicle being in the driving mode and meeting the target calibration conditions, the target sensor is used to obtain the relative motion data, where the target calibration conditions include: the road slope corresponding to the driving road area where the vehicle is located meets the first preset condition, the road elements in the driving road area meet the second preset condition, and the cockpit riding state of the vehicle meets the third preset condition.

[0048] The above driving mode can represent that the vehicle is in a driving state. When the vehicle is in the driving mode, this status information will be synchronized to the sensor calibration system as one of the conditions for judging whether to start the calibration process, ensuring that the calibration is carried out when the vehicle is driving normally, and avoiding the target sensor from obtaining a large amount of invalid relative motion data, consuming computing resources, and causing resource waste.

[0049] The above target calibration conditions can assist in determining whether to start the calibration process. The above driving road area can be the section where the vehicle is currently driving. The driving road area can include various area information (such as road slope information, road element information, road type information, etc.). The above area information can be obtained through an in-vehicle vision sensor in combination with a deep learning algorithm. The area information can be monitored in real time through the in-vehicle sensor. By obtaining the area information of the driving road area, the system can determine whether the target calibration conditions are met, so as to make an intelligent decision on the timing of starting the sensor calibration process and ensure that the sensor parameters can be effectively calibrated.

[0050] The above road slope can also be obtained by real-time monitoring with an in-vehicle tilt sensor. The road slope can also be obtained by combining the positioning system carried by the vehicle with high-precision map data. The above first preset condition can be the condition that the road slope corresponding to the driving road area needs to meet. The first preset condition can be used to represent whether the road slope is greater than the flat threshold, and the flat threshold can be preset in advance. When the road slope corresponding to the driving road area where the vehicle is located is not greater than the flat threshold, it is determined that the driving road area where the vehicle is located is in a "flat and open state". At this time, the road slope corresponding to the driving road area where the vehicle is located meets the first preset condition.

[0051] The above road elements can include but are not limited to: traffic signs, traffic lights, road markings (such as lane dividers, stop lines, zebra crossings, guiding arrows, etc.), road static obstacles, road dynamic obstacles (such as pedestrians, surrounding vehicles, etc.). The above second preset condition can be used to represent whether the target road elements are included in the driving road area. In particular, when the target road elements are set as traffic signs and / or traffic lights, the road image in the driving road area is obtained through the in-vehicle vision sensor, and the deep learning algorithm is used to analyze the road image to determine whether the target road elements (i.e., traffic signs and / or traffic lights) are included in the road image. If the target road elements are included in the road image, it is determined that the road elements in the driving road area meet the second preset condition.

[0052] The above cockpit riding state can be determined by using the driver's seat weight sensor and the passenger seat weight sensor, and the passenger seat weight sensor can be used to measure the load in the passenger seat area in real time. The cockpit riding state can include, but is not limited to: only the driver's seat loaded state, only the passenger seat loaded state, the driver's seat loaded and the passenger seat loaded state, the driver's seat unloaded and the passenger seat unloaded state. When the driver's seat weight sensor detects the main driver load corresponding to the driver's seat area and the passenger seat weight sensor does not detect the load corresponding to the passenger seat area, it is determined that the cockpit riding state is in the only driver's seat loaded state; when the driver's seat weight sensor does not detect the main driver load corresponding to the driver's seat area and the passenger seat weight sensor detects the load corresponding to the passenger seat area, it is determined that the cockpit riding state is in the only passenger seat loaded state. Correspondingly, it can be determined when the cockpit riding state is in the state where both the driver's seat and the passenger seat are loaded, and when the cockpit riding state is in the state where both the driver's seat and the passenger seat are unloaded.

[0053] The above third preset condition can be used to characterize whether the cockpit riding state of the vehicle meets the target riding state. In particular, when the target riding state is set to the only passenger seat loaded state, the driver's seat area is detected by using the driver's seat weight sensor, and the passenger seat area is detected by using the passenger seat weight sensor. In response to the cockpit riding state being in the only driver's seat loaded state, it is determined that the cockpit riding state of the vehicle meets the third preset condition.

[0054] In an exemplary application scenario, still as Figure 2 shown, the target road element is set to a traffic sign and / or a traffic signal light, and the target riding state is set to the only passenger seat loaded state. The road image in the driving road area is obtained by using a vision sensor, and the road image is analyzed and judged by using a deep learning algorithm. When the road slope is not greater than the flat threshold, the road image contains the target road element, and the cockpit riding state is in the only driver's seat loaded state, it is determined that the target calibration condition is met. In response to the vehicle being in the driving mode and meeting the target calibration condition, the sensor calibration system starts an automatic calibration process and obtains relative motion data by using the target sensor.

[0055] The above optional embodiments of the present application can achieve the following technical effects: By judging whether the vehicle is in the driving mode, it is ensured that the sensor parameter calibration is carried out when the vehicle is driving normally. By setting the target calibration condition, the environment conducive to sensor parameter calibration can be more accurately screened out, avoiding the situation of introducing parameter calibration errors due to complex environment, resulting in the inability to obtain accurate relative motion data, ensuring that the target sensor can obtain more accurate relative motion data, and laying a data foundation for determining the target parameters of the target sensor in the future.

[0056] Optionally, in the above step S112, obtaining relative motion data using the target sensor includes the following steps:

[0057] S1121: Select an object from multiple candidate road elements in a preset area in front of the driving road;

[0058] S1122: Sense the object using the target sensor to obtain relative motion data.

[0059] The above object can be selected according to the detection range threshold of the target sensor. The above detection range threshold can be used to characterize the effective detection range of the sensor. The detection range threshold can be determined according to the physical characteristics and algorithm performance of the target sensor. The detection range threshold can also be stored in the internal memory in advance. The detection range threshold can include a minimum detection distance value and a maximum detection distance value. The minimum detection distance value can be used to represent the minimum distance limit for the sensor to detect the object, and the maximum detection distance value can be used to represent the maximum distance limit for the sensor to detect the object. By obtaining the detection range threshold of the target sensor, it is possible to ensure that the target sensor provides accurate and stable detection data within this range, and avoid the situation where the object cannot be effectively sensed due to exceeding the detection range of the sensor.

[0060] The above candidate road elements can include traffic signs (such as traffic indication signs, traffic warning signs, traffic information signs, etc.), traffic lights (such as traffic lights, pedestrian icon lights, etc.).

[0061] In an exemplary application scenario, still as Figure 2 shown, traffic signs and traffic lights are set as candidate road elements. According to the detection range threshold, select some road elements within the detection range threshold from multiple candidate road elements in the preset area in front of the driving road area, and select road elements with a distance from the current vehicle position not less than the target distance (such as 150 m) as the object from these road elements. The above target distance can be configured according to the maximum detection distance value in the detection range threshold. Further, sense the object multiple times using the target sensor to obtain multiple sets of position change information of the target sensor relative to the object, and calculate the relative motion data using these multiple sets of position change information. The above position change information can be calculated based on the current sensing data and the previous sensing data. The above current sensing data can be the sensing data obtained by currently sensing the object using the target sensor. Correspondingly, relative to the current sensing data, the sensing data obtained by the target sensor when sensing the object for the previous time is the previous sensing data.

[0062] It should be noted that after obtaining the relative motion data each time, the sensing data obtained by using the target sensor to sense the target during the process of obtaining the relative motion data this time can be cleared to avoid interference with the process of obtaining the relative motion data next time.

[0063] The above optional embodiments of the present application can achieve the following technical effects: By obtaining the detection range threshold of the target sensor, and according to this detection range threshold, determining the partial road elements within the detection range threshold of the target sensor from multiple candidate road elements, and then automatically selecting the target for calibrating the sensor parameters from this partial road elements. Compared with the related art solution of placing the sensor in a standard calibration site to complete parameter calibration, the present application is not limited to a special site, can realize the automation of target selection, and improve the intelligence of the sensor parameter calibration method. Further, using the target sensor to sense the target and obtain relative motion data can avoid the situation that the target sensor performs invalid sensing on the target due to the target exceeding the detection range threshold of the target sensor, ensuring the validity of the obtained relative motion data.

[0064] Optionally, in the above step S1122, using the target sensor to sense the target to obtain relative motion data includes the following steps:

[0065] S11221: Using the target sensor to collect multiple groups of position parameters corresponding to the target at multiple sensing time points, where the multiple sensing time points are determined according to the target driving range of the vehicle and a preset sensing time interval, and the target driving range is determined according to the detection range threshold;

[0066] S11222: Calculating the average change amount of the multiple groups of position parameters to obtain relative motion data.

[0067] The above target driving range can be used to represent the effective driving area for collecting the position parameters corresponding to the target. The starting point of the target driving range can be the current position of the vehicle. The end point of the target driving range can be determined according to the distance from the vehicle to the target (denoted as M1) and the minimum detection distance value of the detection range threshold (denoted as M2). The distance difference obtained by subtracting the minimum detection distance value from the distance from the vehicle to the target (i.e., M1 - M2) is determined as the end point of the target driving range.

[0068] The above sensing time interval can be the time period for the target sensor to collect the position parameters corresponding to the target. This sensing time interval can be used to represent the collection frequency of the position parameters. This sensing time interval can be configured according to the actual situation.

[0069] The above-mentioned perception time points can be used to represent the time points for collecting position parameters. Each perception time point can correspond to a set of position parameters among multiple sets of position parameters. For example, at a certain perception time point, the collected position parameters are determined as the position parameters corresponding to that perception time point. When the vehicle is driving within the target driving range, according to the preset perception time interval, the perception time points are determined, and at this perception time point, a set of position parameters corresponding to the target object is collected using the target sensor.

[0070] The above-mentioned position parameters can at least include a distance parameter and an azimuth angle parameter. The above-mentioned distance parameter can represent the distance from the target sensor to the target object. The above-mentioned azimuth angle parameter can represent the azimuth angle from the target sensor to the target object. The above-mentioned position parameters can also include, but are not limited to, a relative speed parameter, the lateral position coordinate of the target object, and the longitudinal position coordinate of the target object.

[0071] The above-mentioned mean value of the change amount can include the mean value of the distance parameter and the mean value of the azimuth angle parameter. This mean value of the change amount can be used to characterize the average displacement change during the relative movement between the target object and the vehicle. This mean value of the change amount can be calculated through the following process: Calculate the difference between each set of parameters in the multiple sets of continuously collected position parameters and the previous set of parameters to obtain a set of change amounts (including the distance change amount and the azimuth angle change amount), calculate the average value of the obtained multiple sets of change amounts, and determine the average value of the multiple sets of distance change amounts and the average value of the multiple sets of azimuth angle change amounts as the relative movement data. By calculating the mean value of the change amount for multiple sets of position parameters, part of the random error can be eliminated, ensuring that the obtained relative movement data is more stable and accurate, improving the accuracy of the target parameters, and thus improving the accuracy of sensor parameter calibration.

[0072] In an exemplary application scenario, still as Figure 2 shown, the above-mentioned perception time interval is preset to 60 seconds. The above-mentioned position parameters include a distance parameter (denoted as, L) and an azimuth angle parameter (denoted as, θ). During the process of the vehicle driving within the target driving range, the target sensor collects a set of position parameters corresponding to the target object every 60 seconds and transmits them to the sensor calibration system, obtaining multiple sets of position parameters corresponding to the target object.

[0073] For example, during the process of the vehicle driving within the target driving range, multiple perception time points are determined according to the preset perception time interval. Suppose there are n perception time points (denoted as, t1, t2... t n ), for any perception time point t j , a set of position parameters {Lj, θj} corresponding to the target object is collected using the target sensor, and this set of position parameters is transmitted to the sensor calibration system. Similarly, for the n perception time points, the target sensor collects a total of n sets of position parameters, denoted as {L1, θ1}, {L2, θ2},..., {Ln, θn} respectively.

[0074] Further, starting from the second set of position parameters, the difference between each set of the n sets of position parameters and the previous set of parameters is calculated, that is, {L2, θ2} - {L1, θ1}, {L3, θ3} - {L2, θ2}, ……, {Ln, θn} - {Ln-1, θn-1}, which are sequentially denoted as {△L1, △θ1}, {△L2, △θ2} …… {△Ln-1, △θn-1}, and then the mean value of these change amounts is calculated to obtain relative motion data, denoted as {△La, △θa}.

[0075] The above optional embodiments of the present application can achieve the following technical effects: by using the target driving range and the preset sensing time interval to determine multiple sensing time points, automatically collecting multiple sets of position parameters corresponding to the target object at multiple sensing time points by using the target sensor, and then calculating the mean value of the change amounts of the multiple sets of position parameters, it is possible to avoid the poor stability of the relative motion data caused by random errors. Compared with calculating the relative motion data by using only two sets of position parameters, the embodiments of the present application can improve the accuracy of the automatically obtained relative motion data and provide more accurate data support for determining the target parameters of the target sensor in the future.

[0076] Optionally, in the above sensor parameter calibration method, the following steps may further be included:

[0077] S121: According to the target load information and multiple load conditions, select a target mapping relationship from multiple candidate mapping relationships, where the multiple candidate mapping relationships are used to represent the corresponding relationship between the preset calibration parameters of the target sensor and the calibration motion data under multiple load conditions, and the calibration motion data is used to represent the position change information of the target sensor relative to the test calibration object during the parameter calibration test.

[0078] The above test calibration object may be a reference object used to calibrate the parameters of the target sensor during the parameter calibration test. The above calibration motion data may include a calibration distance change amount and a calibration azimuth angle change amount. The above candidate mapping relationship may be obtained by using a deep learning model, and the deep learning model may include, but is not limited to: a fully connected neural network model, a convolutional neural network model, a recurrent neural network model, a long short-term memory network model, a generative adversarial network model, and a deep residual network model. The above preset calibration parameters may represent the initial pose parameters of the target sensor under each load condition during the parameter calibration test. The above preset calibration parameters may be determined by the physical characteristics of the target sensor. The above preset calibration parameters may include multiple candidate parameter components.

[0079] The above target mapping relationship can be selected using multiple load conditions and an error range, and the error range can be set according to the actual situation. The sensor calibration system compares the target load information with multiple load conditions to determine whether the target load information is within the error range of a certain load condition. If the judgment result is "the target load information is within the error range of a certain load condition", then the candidate mapping relationship corresponding to this load condition is selected as the target mapping relationship.

[0080] The above judgment process can be as follows: If the difference between the target load information and the load information corresponding to a certain load condition does not exceed the error range, then it is determined that the judgment result is "the target load information is within the error range of a certain load condition", and the candidate mapping relationship corresponding to this load condition is selected as the target mapping relationship.

[0081] In an exemplary application scenario, still as Figure 2 shown, during the parameter calibration test, under a certain load condition (e.g., the driver's load is i kilograms), multiple groups of calibration position parameters corresponding to the test calibration object are collected using the target sensor. The above calibration position parameters can include calibration distance parameters and calibration azimuth angle parameters. Based on multiple groups of calibration position parameters, the calibration motion data corresponding to this load condition is calculated and denoted as {△L, △θ}. Further, using a deep learning model, a corresponding relationship between the preset calibration parameters (denoted as {X, Y, Z, α, β, γ}) and the calibration motion data under this load condition is established to obtain the candidate mapping relationship corresponding to this load condition (denoted as {P i}), and the preset calibration parameters under this load condition and the candidate mapping relationship corresponding to this load condition are stored in the non-volatile memory (Non-Volatile Memory, abbreviated as NVM) of the vehicle. Thus, the relational expression {X, Y, Z, α, β, γ}{P i} = {△L, △θ} can be obtained. For example, when the driver's load is 50 kg, the preset calibration parameters under the load condition of the driver's load being 50 kg are {X 50 , Y 50 , Z 50 , α 50 , β 50 , γ 50}, and the calibration motion data under the load condition of the driver's load being 50 kg is {△L 50 , △θ 50}, and the relational expression {X 50 , Y 50 , Z 50 , α 50 , β 50 , γ 50}{P 50} = {△L50 , △θ 50}。

[0082] Still in the above application scenario, further, for multiple load conditions, candidate mapping relationships corresponding to each load condition in the multiple load conditions can be obtained respectively. In particular, taking the driver's load as the load condition, multiple load conditions with the driver's load being 50 kg, 55 kg, 60 kg, 65 kg, 70 kg, 75 kg, and 80 kg are selected respectively. These multiple load conditions are only examples, and the present application does not limit the specific load conditions. A corresponding table of the driver's load and the candidate mapping relationships can be obtained, as shown in Table 1.

[0083] Table 1

[0084]

[0085] Still in the above application scenario, according to the target load information obtained in real time, the sensor calibration system compares the target load information with multiple load conditions. If the target load information is within the error range of a certain load condition, the candidate mapping relationship corresponding to that load condition is selected as the target mapping relationship. For example, the error range can be set to -2.5 kg to +2.5 kg. The error range corresponding to the load condition of "the driver's load is 50 kg" can be (the driver's load is 47.5 kg to 52.5 kg). That is to say, assuming the target load information is 49 kg, it is determined that the target load information is within the error range of the load condition of "the driver's load is 50 kg", and P 50 is selected as the target mapping relationship.

[0086] The above optional embodiments of the present application can achieve the following technical effects: By determining the candidate mapping relationships corresponding to each load condition in multiple load conditions during the parameter calibration experiment, it is ensured that the system can automatically select the corresponding target mapping relationship from multiple candidate mapping relationships according to the target load information obtained in real time, improving the selection accuracy of the target mapping relationship.

[0087] Optionally, the above sensor parameter calibration method further includes:

[0088] S140: Compare the parameter values of the target parameter and the preset calibration parameter corresponding to the target mapping relationship to obtain a comparison result;

[0089] S150: In response to the comparison result satisfying the target rewriting condition, calibrate the target sensor with the target parameter and generate a calibration success feedback message;

[0090] S160: In response to the comparison result not satisfying the target rewriting condition, generate a no-calibration required feedback message.

[0091] The above-mentioned preset calibration parameters are stored in the non-volatile memory of the vehicle, and the non-volatile memory may include, but is not limited to: Flash Memory, Phase Change Memory (PCM for short), Magneto-Resistive Random Access Memory (MRAM for short). The above-mentioned target parameters can characterize the pose parameters of the current calculated target sensor under the target load information. The target parameters may include a plurality of target parameter components respectively corresponding to a plurality of candidate parameter components.

[0092] The sensor calibration system obtains the preset calibration parameters corresponding to the target mapping relationship from the non-volatile memory, and uses the obtained relative motion data to calculate the target parameters corresponding to the target sensor according to the target mapping relationship. Further, a comparative analysis is performed on the preset calibration parameters and the target parameters, and the parameters of the target sensor are automatically calibrated according to the analysis result.

[0093] The above comparison result may be a parameter difference value. The comparison result can be used to characterize the difference between the preset calibration parameters and the target parameters. The above target rewriting condition is used to assist in determining whether parameter calibration of the target sensor is required. The target rewriting condition can be adjusted according to user requirements. The above calibration success feedback information can be used to characterize that the parameter calibration of the target sensor is successful. The above no-calibration feedback information can be used to characterize that there is no need to calibrate the pose parameters of the target sensor.

[0094] In an exemplary application scenario, still as Figure 2 shown, the target load information is "the main driver's load is 50 kg", and the determined target mapping relationship is P 50 , and the preset calibration parameters {X, Y, Z, α, β, γ} corresponding to the target mapping relationship are obtained. Further, using the obtained relative motion data {△La, △θa}, according to the target mapping relationship P 50 the target parameters corresponding to the target sensor are calculated, denoted as {X a , Y a , Z a , α a , β a , γ a}.

[0095] Still in the above application scenario, still as Figure 2 shown, the sensor calibration system compares and analyzes each of the multiple candidate parameter components included in the preset calibration parameters with the multiple target parameter components respectively corresponding to the multiple candidate parameter components included in the target parameters to obtain a comparison result. For example, the preset calibration parameters are {X, Y, Z, α, β, γ}, and the target parameters are {Xa ,Y a ,Z a ,α a ,β a ,γ a}, and sequentially compare and analyze X with X a ,Y with Y a ,Z with Z a ,α with α a ,β with β a ,γ with γ a ,to obtain the comparison result.

[0096] Still in the above application scenario, determine whether the comparison result meets the target rewriting condition. If the comparison result meets the target rewriting condition, the sensor calibration system sends an update parameter instruction to the target sensor and sends a calibration feedback instruction to the target sensor according to a preset period. The target sensor updates the pose parameters using the target parameters according to the update parameter instruction, and generates a calibration success feedback message after the update is successful. In response to the target sensor receiving the calibration feedback instruction, the calibration success feedback message is transmitted back to the sensor calibration system. If the comparison result does not meet the target rewriting condition, the sensor calibration system generates a no-calibration required feedback message, and the target sensor does not need to perform parameter calibration. The above update parameter instruction includes at least the target parameters.

[0097] It should be noted that still in the above application scenario, in response to the sensor calibration system successfully receiving the calibration success feedback message, the sensor calibration system can also read the pose parameters stored in the target sensor, and determine whether the read pose parameters are consistent with the target parameters. If the judgment result indicates that the read pose parameters are not consistent with the target parameters, the sensor calibration system resends the update parameter instruction to the target sensor. If the judgment result indicates that the read pose parameters are consistent with the target parameters, the sensor calibration system determines that the calibration is completed.

[0098] In an exemplary application scenario, the above preset calibration parameters can be the pose parameters pre-calibrated for the target sensor before the vehicle leaves the factory, and can include multiple candidate parameter components under the calibration conditions. The above calibration conditions can be determined by the calibration motion data and the preset load information during the vehicle test process. The sensor calibration system sequentially compares and analyzes the multiple candidate parameter components included in the preset calibration parameters with the multiple target parameter components respectively corresponding to the multiple candidate parameter components included in the target parameters to obtain the comparison result.

[0099] It should be noted that in another exemplary application scenario, multiple sets of pre-calibrated data can be stored in the NVM of the vehicle. For example, multiple sets of data corresponding to multiple load values, where each load value corresponds to a set of pose parameters. After the vehicle is powered on, according to the real-time load value of the vehicle, a set of pose parameters corresponding to the real-time load value is selected from the above multiple sets of data as the above-mentioned preset calibration parameters. In addition, the real-time load value may not be the same as any of the multiple load values. In this case, the load value closest to the real-time load value is determined from the multiple load values, and a set of pose parameters corresponding to the closest load value is determined as the above-mentioned preset calibration parameters.

[0100] The above optional embodiments of the present application can achieve the following technical effects: The target parameters are automatically calculated using the relative motion data and the target mapping relationship. By comparing and analyzing the preset calibration parameters and the target parameters, a comparison result is obtained. Furthermore, by setting the target rewrite condition, the sensor calibration system can automatically determine whether parameter calibration of the target sensor is required, enhancing the intelligence of the sensor parameter calibration method. In addition, by setting the target rewrite condition according to the actual situation, it is possible to support the user to flexibly select the timing of calibrating the pose parameters of the target sensor, meet the personalized needs of the user, and improve the accuracy of sensor parameter calibration. Compared with the sensor parameter calibration methods in the related art, this solution does not need to rely on complex calibration tools and professional sites, and can automatically calibrate the sensor parameters during the vehicle driving process, improving the calibration efficiency of the sensor parameters and reducing the cost of sensor parameter calibration.

[0101] Optionally, the preset calibration parameters include multiple first parameter value components, the target parameters include multiple second parameter value components respectively corresponding to the multiple first parameter value components, the comparison result includes multiple difference components, and the multiple difference components are obtained by subtracting the multiple first parameter value components from the multiple second parameter value components. In the above step S150, the comparison result satisfying the target rewrite condition includes:

[0102] S151: At least one target difference component among the multiple difference components is greater than the target threshold component corresponding to the target difference component among the multiple threshold components corresponding to the target rewrite condition.

[0103] The above first parameter value components may include a candidate first distance component corresponding to the target sensor under the x-axis of the spatial coordinate, a candidate second distance component corresponding to the target sensor under the y-axis of the spatial coordinate, a candidate third distance component corresponding to the target sensor under the z-axis of the spatial coordinate, a candidate pitch angle corresponding to the target sensor, a candidate roll angle corresponding to the target sensor, and a candidate yaw angle corresponding to the target sensor.

[0104] The second parameter value component described above may include a target first distance component from the target sensor to the target object along the x-axis of the spatial coordinate, a target second distance component from the target sensor to the target object along the y-axis of the spatial coordinate, a target third distance component from the target sensor to the target object along the z-axis of the spatial coordinate, a target pitch angle of the target sensor to the target object, a target roll angle of the target sensor to the target object, and a target yaw angle of the target sensor to the target object. The difference component described above may include a first distance component difference, a second distance component difference, a third distance component difference, a pitch angle difference, a roll angle difference, and a yaw angle difference.

[0105] Subtract a certain first parameter value component in the multiple first parameter value components included in the preset calibration parameter from the second parameter value component corresponding to this first parameter value component in the target parameter in sequence to obtain the difference component corresponding to this first parameter value component. Similarly, the difference components corresponding to each first parameter value component can be obtained to obtain multiple difference components. Determine in sequence whether each difference component in the multiple difference components is greater than the threshold component corresponding to this difference component in the multiple threshold components corresponding to the target rewriting condition. If at least one target difference component is greater than the target threshold component corresponding to this target difference component in the multiple threshold components corresponding to the target rewriting condition, it is determined that the comparison result meets the target rewriting condition.

[0106] In an exemplary application scenario, still as Figure 2 shown, the preset calibration parameter is {X, Y, Z, α, β, γ}, and the target parameter is {X a , Y a , Z a , α a , β a , γ a}, and the multiple threshold components corresponding to the target rewriting condition are {δX, δY, δZ, δα, δβ, δγ}. For the first parameter value component X, compare and analyze X with X a to obtain the difference component △X corresponding to this first parameter value component X. Compare and analyze △X with δX. If △X is greater than δX, it is determined that the analysis result corresponding to this first parameter value component X is "the difference component is greater than the corresponding threshold component". If △X is not greater than δX, it is determined that the analysis result corresponding to this first parameter value component X is "the difference component is not greater than the corresponding threshold component". Similarly, the difference components corresponding to each first parameter value component can be compared and analyzed with the threshold components corresponding to this first parameter value component to obtain corresponding multiple analysis results. If multiple analysis results are all "the difference component is not greater than the corresponding threshold component", it is determined that the comparison result does not meet the target rewriting condition; if at least one of the multiple analysis results is "the difference component is greater than the corresponding threshold component", it is determined that the comparison result meets the target rewriting condition.

[0107] The above optional embodiments of the present application can achieve the following technical effects: By successively subtracting each first parameter value component from the corresponding second parameter value component, the difference component corresponding to each first parameter value component is obtained. Further, by comparing and analyzing each difference component with the corresponding threshold component, if there is a difference component greater than the corresponding threshold component, the pose parameters of the target sensor are calibrated, which can more accurately identify the timing of calibrating the pose parameters of the target sensor and more timely calibrate the parameters of the target sensor.

[0108] Optionally, the above sensor parameter calibration method further includes the following steps:

[0109] S170: In response to the vehicle being equipped with a calibrated sensor, obtain the first position information and the second position information corresponding to the target object. Among them, the first position information includes the first distance parameter and the first azimuth parameter of the target sensor relative to the target object, and the second position information includes the second distance parameter and the second azimuth parameter of the calibrated sensor relative to the target object;

[0110] S180: Perform subtraction calculation on the first distance parameter and the second distance parameter to obtain a first difference;

[0111] S190: Perform subtraction calculation on the first azimuth parameter and the second azimuth parameter to obtain a second difference;

[0112] S200: In response to the first difference being greater than the preset distance accuracy redundancy value and / or the second difference being greater than the preset azimuth accuracy redundancy value, when it is detected that the vehicle is in the driving mode and meets the target calibration conditions, re-determine the target parameters of the target sensor.

[0113] The above calibrated sensor can be a sensor with correct pose parameters. The calibrated sensor can include, but is not limited to: vision sensors (such as on-vehicle cameras), lidar sensors, millimeter-wave radar sensors, and thermal imaging sensors. The above first distance parameter can represent the distance from the target sensor to the target object, and the above first azimuth parameter can represent the azimuth of the target sensor to the target object. The above second distance parameter can represent the distance from the calibrated sensor to the target object, and the above second azimuth parameter can represent the azimuth of the calibrated sensor to the target object. The above first difference and the above second difference can be used to assist in judging whether the pose parameters of the target sensor are correct.

[0114] The above distance accuracy redundancy value and the above azimuth angle accuracy redundancy value can represent the maximum deviation value that the pose parameters of the target sensor can allow. The above distance accuracy redundancy value can be a preset distance error range. The above azimuth angle accuracy redundancy value can be a preset azimuth angle error range. The above distance accuracy redundancy value and the above azimuth angle accuracy redundancy value can be used to assist in determining whether it is necessary to re-determine the target parameters of the target sensor.

[0115] In an exemplary application scenario, Figure 3 is a schematic diagram of an optional method for verifying the pose parameters of a target sensor provided in an embodiment of the present application. As Figure 3 shown, the lidar sensor installed in the vehicle is selected as the calibrated sensor. The sensor calibration system selects a target object, and uses the target sensor to collect the first distance parameter and the first azimuth angle parameter of the target sensor relative to the target object, denoted as {L bd , θ bd}. The calibrated sensor is used to collect the second distance parameter and the second azimuth angle parameter of the calibrated sensor relative to the target object, denoted as {L cz , θ cz}. The first difference is calculated by subtracting the second distance parameter from the first distance parameter, denoted as △L yz , that is, △L yz = L bd - L bd . The second difference is calculated by subtracting the second azimuth angle parameter from the first azimuth angle parameter, denoted as △θ yz , that is, △θ yz = θ bd - θ cz . Further, it is determined whether the first difference is greater than the preset distance accuracy redundancy value (denoted as △L ry ), and it is determined whether the second difference is greater than the preset azimuth angle accuracy redundancy value (denoted as △θ ry ). In response to the first difference being greater than the preset distance accuracy redundancy value and / or the second difference being greater than the preset azimuth angle accuracy redundancy value, when it is detected that the vehicle is in the driving mode and meets the target calibration conditions, the target parameters of the target sensor are re-determined. In response to the first difference not being greater than the preset distance accuracy redundancy value and the second difference not being greater than the preset azimuth angle accuracy redundancy value, the sensor calibration system considers that the parameter calibration is correct and passes the pose parameter verification of the target sensor.

[0116] The above optional embodiments of the present application can achieve the following technical effects: By using the calibrated sensor to verify the pose parameters of the target sensor, the correctness of the pose parameters of the target sensor can be ensured. In addition, in the case where the pose parameters of the target sensor fail to pass the verification, the process of determining the target parameters of the target sensor can be automatically restarted, enhancing the intelligence of the sensor parameter calibration method.

[0117] Embodiment 2

[0118] The embodiment of the present application also provides a sensor parameter calibration device 40. Please refer to Figure 4 , which includes: an acquisition module 410 for acquiring the target load information and relative motion data corresponding to the vehicle, where the target load information is used to characterize the load condition corresponding to at least part of the vehicle area, and the relative motion data is used to characterize the position change information of the target sensor relative to the target object during the driving of the vehicle; a determination module 420 for determining the target mapping relationship corresponding to the target load according to the target load information, where the target mapping relationship represents the corresponding relationship between the calibrated motion data of the vehicle and the calibrated parameters of the target sensor, different target loads correspond to different target mapping relationships, and the calibrated motion data is used to characterize the position change information of the target sensor relative to the test calibration object during the parameter calibration test; a calibration module 430 for determining the target parameters of the target sensor according to the relative motion data and the target mapping relationship.

[0119] The above sensor parameter calibration device provided by the embodiment of the present application achieves the following technical effects: Through the acquisition module, according to the load condition corresponding to at least part of the vehicle area, the target load information corresponding to the vehicle is acquired in real time to ensure the timeliness of the target load information, and the relative motion data is acquired according to the position change information of the target sensor relative to the target object; through the determination module, by using the mapping relationships respectively corresponding to various load information obtained by performing parameter calibration tests on the target sensor under various load conditions, the target mapping relationship of the target sensor under the target load information can be determined more accurately; further, through the determination module, according to the relative motion data and the target mapping relationship, the target parameters of the target sensor are automatically determined, reducing the dependence on manual operations, improving the efficiency and correctness of the calibration of the sensor parameter calibration device, so that the pose parameters of the target sensor can be calibrated more efficiently and accurately by using this sensor parameter calibration device.

[0120] Embodiment 3

[0121] The embodiment of the present application also provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, it can implement the above-mentioned sensor parameter calibration method.

[0122] Optionally, the above computer program product may provide a sensor parameter calibration service based on the above sensor parameter calibration method.

[0123] Optionally, in this embodiment, the above computer program product may be a set of instructions and codes pre-written according to the above sensor parameter calibration method. The computer program product can run on various different computer platforms, including personal computers, servers, mobile devices, etc.

[0124] Optionally, in this embodiment, the instructions and codes corresponding to the computer program product are used to implement the following method steps: obtaining target load information and relative motion data corresponding to a vehicle, where the target load information is used to characterize the load condition corresponding to at least part of the vehicle area, and the relative motion data is used to characterize the position change information of the target sensor relative to the target during the driving process of the vehicle; determining a target mapping relationship corresponding to the target load according to the target load information, where the target mapping relationship characterizes the corresponding relationship between the calibrated motion data of the vehicle and the calibrated parameters of the target sensor, different target loads correspond to different target mapping relationships, and the calibrated motion data is used to characterize the position change information of the target sensor relative to the test calibration object during the parameter calibration test; determining the target parameters of the target sensor according to the relative motion data and the target mapping relationship.

[0125] The above computer program product provided by the embodiments of the present application achieves the following technical effects: using a processor to execute a computer program for implementing the sensor parameter calibration method of any one of the above, obtaining target load information corresponding to the vehicle in real time according to the load condition corresponding to at least part of the vehicle area to ensure the timeliness of the target load information, and obtaining relative motion data according to the position change information of the target sensor relative to the target; using the mapping relationships respectively corresponding under the conditions of multiple load information obtained by performing parameter calibration tests on the target sensor under multiple load conditions, the target mapping relationship of the target sensor under the target load information can be determined more accurately; further, according to the relative motion data and the target mapping relationship, automatically determining the target parameters of the target sensor, reducing the dependence on manual operations, so that the parameter calibration of the target sensor is automatically completed during the driving process of the vehicle by using this computer program product, improving the efficiency and correctness of sensor parameter calibration.

[0126] Embodiment 4

[0127] The embodiments of the present application further provide a vehicle 50. Please refer to Figure 5 , including an in-vehicle memory 510 and an in-vehicle processor 520. Among them, the in-vehicle memory 510 is used to store a computer program; the in-vehicle processor 520 is used to execute the computer program stored on the memory to implement the sensor parameter calibration method of any embodiment.

[0128] The vehicle provided by the embodiment of the present application achieves the following technical effects: storing the computer program corresponding to the sensor parameter calibration method for implementing any one of the above in the vehicle-mounted memory, and using the vehicle-mounted processor to execute the computer program stored on the memory, obtaining the target load information corresponding to the vehicle in real time according to the load conditions corresponding to at least some areas in the vehicle to ensure the timeliness of the target load information, and obtaining relative motion data according to the position change information of the target sensor relative to the target object; being able to more accurately determine the target mapping relationship of the target sensor under the target load information by using the mapping relationships respectively corresponding to various load information obtained from the parameter calibration tests of the target sensor under various load conditions; further, according to the relative motion data and the target mapping relationship, being able to automatically determine the target parameters of the target sensor during the driving of the vehicle, thereby automatically calibrating the parameters of the target sensor, reducing the dependence on manual operations, improving the efficiency and correctness of sensor parameter calibration, enhancing the performance of vehicle-mounted sensors, ensuring the normal operation of vehicle-mounted sensors, reducing the vehicle maintenance cost, and enhancing the overall performance of the vehicle.

[0129] Those of ordinary skill in the art can understand that, similarly, the above vehicle can also be a computing terminal. Please refer to Figure 6 As shown, the computing terminal 60 (such as, a computer terminal, a mobile intelligent terminal, a vehicle terminal, or a cloud computing virtual terminal, etc.) may include: one or more processors 602 (such as, may include processors 602a, 602b,..., 602n), a memory 604 for storing data, and a transmission device 606 for implementing communication functions, wherein the processor 602 may include, but is not limited to, processing components such as a microcontroller unit (MCU) or a field programmable gate array (FPGA).

[0130] The above computing terminal 60 may further include: a display, an input / output interface, a universal serial bus (USB) port (this USB port may be one of the ports of the computer bus, not shown in the figure), a network interface (not shown in the figure), a power supply (not shown in the figure), and a camera (not shown in the figure).

[0131] It should be noted that one or more processors 602 and / or other data processing circuits in the above computing terminal 60 may be embodied in whole or in part as software, hardware, firmware, or any arbitrary combination thereof. In addition, the data processing circuit may be a single independent processing module, or may be wholly or partially incorporated into any one of the other elements in the computing terminal 60 (or mobile device).

[0132] The memory 604 can be used to store software programs and modules of application software, such as program instructions and data storage devices corresponding to the path planning method in the embodiments of the present application. The processor 602 executes various functional applications and data processing by running the software programs and modules stored in the memory 604, that is, implements the above-mentioned path planning method. The memory 604 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memories. In some instances, the memory 604 may further include a memory remotely provided relative to the processor 602, and these remote memories can be connected to the vehicle terminal 60 through a network. Examples of the above-mentioned network include but are not limited to the Internet, intranet, local area network, mobile communication network, and combinations thereof.

[0133] The transmission device 606 is used to receive or send data via a network. Specific examples of the above-mentioned network may include a wireless network provided by a communication provider of the vehicle terminal 60. In one instance, the transmission device 606 includes a network adapter (Network Interface Controller, abbreviated as NIC) and a network interface, and the network adapter can be connected to other network devices through a base station so as to communicate with the Internet. The transmission device 606 can perform data communication in a wired and / or wireless network connection manner. In one instance, the transmission device 606 can be a radio frequency (RF) module, and this RF module is used to communicate with the Internet wirelessly.

[0134] The input / output interface can be connected to the input / output devices corresponding to the computing terminal 60 to implement input / output functions. The above-mentioned input / output devices may include but are not limited to: cursor control devices, keyboards, displays, etc. The above-mentioned input / output devices can be built into the computing terminal 60 or external external devices of the computing terminal 60.

[0135] Those of ordinary skill in the art can understand that Figure 6 The structure of the shown computing terminal 60 is only schematic and does not impose strict limitations on the structure of the above-mentioned computing terminal 60. For example, the computing terminal 60 may further include more or fewer components than those shown in Figure 6 or the computing terminal 60 may have components of different categories from those shown in Figure 6 shown.

[0136] Embodiment Five

[0137] The embodiments of the present application further provide an electronic device, 70, please refer to Figure 7, including a memory 710 and a processor 720. Among them, the memory 710 is used to store computer programs; the processor 720 is used to execute the programs stored on the memory to implement the sensor parameter calibration method introduced in any embodiment of the present application.

[0138] The above electronic device provided by the embodiments of the present application achieves the following technical effects: By using the processor to execute the computer program for implementing the sensor parameter calibration method of any one of the above, according to the load conditions corresponding to at least some areas in the vehicle, the target load information corresponding to the vehicle is obtained in real time to ensure the timeliness of the target load information. According to the position change information of the target sensor relative to the target object, relative motion data is obtained; By using the mapping relationships respectively corresponding to various load information obtained from parameter calibration tests of the target sensor under various load conditions, the target mapping relationship of the target sensor under the target load information can be determined more accurately; Further, according to the relative motion data and the target mapping relationship, the target parameters of the target sensor are automatically determined, reducing the dependence on manual operations, and automatically completing the parameter calibration of the target sensor during the vehicle driving process, improving the efficiency and correctness of sensor parameter calibration.

[0139] Those of ordinary skill in the art can understand that Figure 7 The structure shown is only for illustration, and the electronic device can also be a terminal device such as a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a handheld computer, and a Mobile Internet Device (MID for short). Figure 7 It does not limit the structure of the above electronic device. For example, the electronic device 70 may further include more or fewer components than those shown in Figure 7 (such as a network interface, a display device, etc.), or have a different configuration from that shown in Figure 7 shown.

[0140] The serial numbers of the above embodiments of the present application are only for description and do not represent the advantages and disadvantages of the embodiments.

[0141] In the above embodiments of the present application, the descriptions of each embodiment have their own focuses. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0142] In the present application, "a plurality of" means two or more.

[0143] In this application, unless otherwise clearly defined, the terms "install", "connect", and "couple" shall be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, and it can be the communication inside two components. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0144] The terms "first", "second", "third", "fourth", etc. (if any) in this application are used to distinguish similar objects and do not necessarily describe a specific order or sequence.

[0145] The term "and / or" in this application is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this application generally indicates that the related objects before and after are in an "or" relationship.

[0146] If there is no special instruction, all steps of this application can be carried out in sequence or randomly. For example, the method includes steps A and B, which means that the method can include steps A and B carried out in sequence, or steps B and A carried out in sequence. For example, it is mentioned that the method may further include step C, which means that step C can be added to the method in any order. For example, the method can include steps A, B, and C, or steps A, C, and B, or steps C, A, and B, etc.

[0147] The above are only the preferred embodiments of this application and are not intended to limit this application. Any modifications, equivalent replacements, and improvements made within the spirit and principles of this application shall be included in the protection scope of this application.

Claims

1. A method for calibrating sensor parameters, characterized in that, Including: Obtain the target load information and relative motion data corresponding to the vehicle, where the target load information is used to characterize the load condition corresponding to at least part of the vehicle, and the relative motion data is used to characterize the position change information of the target sensor relative to the target during the driving process of the vehicle; Determine the target mapping relationship corresponding to the target load according to the target load information, where the target mapping relationship characterizes the corresponding relationship between the calibrated motion data of the vehicle and the calibrated parameters of the target sensor, and different target loads correspond to different target mapping relationships. The calibrated motion data is used to characterize the position change information of the target sensor relative to the test calibration object during the parameter calibration test; Determine the target parameters of the target sensor according to the relative motion data and the target mapping relationship.

2. The sensor parameter calibration method according to claim 1, characterized in that Obtaining the target load information includes: In response to the power-on event of the vehicle, use the driver's seat weight sensor installed in the vehicle to obtain the target load information, where the target load information is used to characterize the load corresponding to the driver's seat area in the vehicle.

3. The sensor parameter calibration method according to claim 1, wherein Obtaining the relative motion data includes: In response to the vehicle being in the driving mode and meeting the target calibration conditions, use the target sensor to obtain the relative motion data, where the target calibration conditions include: the road slope corresponding to the driving road area where the vehicle is located meets the first preset condition, the road elements in the driving road area meet the second preset condition, and the cockpit riding state of the vehicle meets the third preset condition.

4. The sensor parameter calibration method according to claim 3, wherein Using the target sensor to obtain the relative motion data includes: Select the target from multiple candidate road elements in the preset area in front of the driving road; Use the target sensor to sense the target to obtain the relative motion data.

5. The sensor parameter calibration method according to claim 4, wherein Using the target sensor to sense the target to obtain the relative motion data includes: Use the target sensor to collect multiple groups of position parameters corresponding to the target at multiple sensing time points, where the multiple sensing time points are determined according to the target driving range of the vehicle and the preset sensing time interval, and the target driving range is determined according to the detection range threshold; Calculate the average value of the change amounts of the multiple groups of position parameters to obtain the relative motion data.

6. The sensor parameter calibration method according to claim 1, wherein The sensor parameter calibration method further includes: Compare the target parameters with the preset calibration parameters to obtain a comparison result; In response to the comparison result meeting the target rewriting condition, use the target parameters to calibrate the target sensor and generate a calibration success feedback message; In response to the comparison result not meeting the target rewriting condition, generate a feedback message indicating that calibration is not required.

7. The sensor parameter calibration method according to claim 6, characterized in that, The preset calibration parameters include multiple first parameter value components, the target parameters include multiple second parameter value components corresponding to the multiple first parameter value components respectively, the comparison result includes multiple difference components, and the multiple difference components are obtained by subtracting the multiple second parameter value components from the multiple first parameter value components. The comparison result meeting the target rewriting condition includes: At least one target difference component among the multiple difference components is greater than the target threshold component corresponding to the target difference component among the multiple threshold components corresponding to the target rewriting condition.

8. The sensor parameter calibration method according to any one of claims 1 to 7, characterized in that, The method for calibrating the sensor parameters further includes: In response to the vehicle being equipped with a calibrated sensor, acquiring first position information and second position information corresponding to the target object, where the first position information includes a first distance parameter and a first azimuth parameter of the target sensor relative to the target object, and the second position information includes a second distance parameter and a second azimuth parameter of the calibrated sensor relative to the target object; Performing a subtraction calculation on the first distance parameter and the second distance parameter to obtain a first difference; Performing a subtraction calculation on the first azimuth parameter and the second azimuth parameter to obtain a second difference; In response to the first difference being greater than a preset distance accuracy redundancy value and / or the second difference being greater than a preset azimuth accuracy redundancy value, when it is detected that the vehicle is in a driving mode and meets the target calibration condition, re-determining the target parameter of the target sensor.

9. A sensor parameter calibration device, characterized in that, including: an acquisition module, configured to acquire target load information and relative motion data corresponding to the vehicle, where the target load information is used to characterize the load condition corresponding to at least part of the vehicle, and the relative motion data is used to characterize the position change information of the target sensor relative to the target object during the driving process of the vehicle; a determination module, configured to determine a target mapping relationship corresponding to the target load according to the target load information, where the target mapping relationship characterizes the correspondence between the calibrated motion data of the vehicle and the calibrated parameters of the target sensor, and different target loads correspond to different target mapping relationships, and the calibrated motion data is used to characterize the position change information of the target sensor relative to the test calibration object during the parameter calibration test; a calibration module, configured to determine the target parameter of the target sensor according to the relative motion data and the target mapping relationship.

10. A vehicle, characterized in that, including an in-vehicle processor and an in-vehicle memory, where the in-vehicle memory is used to store a computer program; the in-vehicle processor is configured to execute the computer program stored on the memory to implement the method for calibrating the sensor parameters according to any one of claims 1 to 8.

Images