Position detection method, vehicle control method, device, vehicle, medium and chip

Abstract

The disclosed embodiments provide a position detection method, a vehicle control method, a device, a vehicle, a medium and a chip, and relate to the field of autonomous driving technology. The position detection method includes: obtaining a set of perception data collected by multiple different types of sensors of a vehicle; for each perception data set, determining whether the matching detection result corresponding to the perception data set is a first target result according to the similarity of the perception data at different times in the perception data set; wherein the first target result is used to characterize that the similarity between the target position of the vehicle at the target time and the historical position of the vehicle at the historical time is greater than or equal to a first preset threshold; when the matching detection result corresponding to the first perception data set in multiple perception data sets is the first target result, if the matching detection result corresponding to the second perception data set is the second target result, then it is determined that the target position is the same as the historical position.

Description

Technical Field

[0001] The present disclosure relates to the field of autonomous driving technology, and more specifically, to a position detection method, a vehicle control method, a device, a vehicle, a medium, and a chip. Background Art

[0002] In autonomous driving technology, vehicles can collect environmental data through sensors, use Simultaneous Localization And Mapping (SLAM) technology to map the surrounding environment during movement, and realize the positioning of the vehicle itself based on the mapping results. However, due to errors in the mapping and positioning process, the accumulation of errors will lead to unreliable mapping and positioning results. Summary of the invention

[0003] In view of this, the embodiments of the present disclosure propose a new technical solution for position detection.

[0004] According to a first aspect of an embodiment of the present disclosure, a position detection method is provided, the method comprising:

[0005] Acquire a set of perception data respectively collected by a plurality of different types of sensors of the vehicle; wherein each of the perception data sets includes a plurality of perception data collected by a sensor of the same type at different times within a preset time;

[0006] For each perception data set, determining whether a matching detection result corresponding to the perception data set is a first target result according to the similarity of the perception data at different times in the perception data set; wherein the first target result is used to characterize that the similarity between the target position of the vehicle at the target time and the historical position of the vehicle at a historical time is greater than or equal to a first preset threshold, and the historical time is a historical time earlier than the target time within the preset time;

[0007] In the case where the matching detection result corresponding to the first perception data set among the multiple perception data sets is the first target result, determining whether the matching detection result corresponding to the second perception data set is the second target result; wherein the second perception data set is other perception data sets among the multiple perception data sets except the first perception data set, and the second target result is used to characterize that the similarity between the target position and the historical position is greater than or equal to a second preset threshold, and the second preset threshold is less than the first preset threshold;

[0008] When the matching detection result corresponding to the second perception data set is a second target result, it is determined that the target position is the same as the historical position.

[0009] Optionally, the multiple sensors of different types include image sensors and position sensors, and the position sensors include positioning navigation units and / or inertial measurement units IMU.

[0010] Optionally, the perception data set includes a location data set acquired based on the location sensor; and determining whether the matching detection result corresponding to the perception data set is the first target result according to the similarity of the perception data at different times in the perception data set includes:

[0011] If a historical location that satisfies a first preset location condition and that meets the target location is searched in the location data set, determining that the similarity between the target location and the historical location is greater than or equal to a first preset threshold, so as to determine that the matching detection result corresponding to the location data set is the first target result;

[0012] The first preset location condition includes that the distance between the target location and the historical location is less than or equal to a first preset distance threshold.

[0013] Optionally, the first preset location condition further includes at least one of the following:

[0014] The cumulative change in the heading angle of the vehicle during the process of traveling from the historical position to the target position is greater than or equal to a first preset angle threshold;

[0015] The cumulative change in slope of the vehicle during the process of traveling from the historical position to the target position is less than or equal to a first preset slope threshold.

[0016] Optionally, the first perception data set is a position data set acquired by the position sensor, and the second perception data set is an image data set acquired by the image sensor.

[0017] Optionally, the perception data set includes an image data set acquired based on the image sensor; and determining whether the matching detection result corresponding to the perception data set is the first target result according to the similarity of the perception data at different times in the perception data set includes:

[0018] Sampling the image data at different times in the image data set to obtain image features corresponding to the image data; wherein the image features include feature description vectors of all or part of the regions of the image data;

[0019] If historical moment image data whose image feature similarity with the target moment image data is greater than or equal to a first preset image similarity threshold is searched in the image data set, the matching detection result corresponding to the image data set is determined to be the first target result.

[0020] Optionally, the image sensor is a forward-looking camera of the vehicle.

[0021] Optionally, the first perception data set is an image data set acquired by the image sensor, and the second perception data set is a position data set acquired by the position sensor.

[0022] Optionally, the target moment is N consecutive moments, where N is a preset positive integer greater than 1.

[0023] According to a second aspect of an embodiment of the present disclosure, a vehicle control method is provided, the method comprising:

[0024] Acquire a set of perception data respectively collected by a plurality of different types of sensors of the vehicle; wherein each of the perception data sets includes a plurality of perception data collected by a sensor of the same type at different times within a preset time;

[0025] Controlling the vehicle to travel according to the set of perception data;

[0026] Among them, the multiple perception data sets are used to determine whether the target position of the vehicle at the target time is the same as the historical position of the vehicle at the historical time.

[0027] According to a third aspect of an embodiment of the present disclosure, an electronic device is provided, comprising a memory and a processor, wherein the memory is used to store computer instructions, and the processor is used to call the computer instructions from the memory to execute a method as described in any one of the first aspect and / or the second aspect.

[0028] According to a fourth aspect of an embodiment of the present disclosure, a vehicle is provided, comprising a memory and a processor, wherein the memory is used to store computer instructions, and the processor is used to call the computer instructions from the memory to execute a method as described in any one of the first aspect and / or the second aspect.

[0029] According to a fifth aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the method described in any one of the first aspect and / or the second aspect is implemented.

[0030] According to a sixth aspect of an embodiment of the present disclosure, a chip is provided, comprising a processing unit, wherein the processing unit is configured to execute a method as described in any one of the first aspect and / or the second aspect.

[0031] Based on the position detection method provided by the embodiment of the present disclosure, a set of perception data collected by multiple different types of sensors of the vehicle is obtained, and detection is performed separately according to the multiple perception data sets, which can avoid missed detection. Furthermore, if the matching detection result of the first perception data set is determined to be the first target result, it is determined whether the matching detection result of the second perception data set is the second target result. Only when the matching detection result corresponding to the second perception data set is the second target result, the target position is determined to be the same as the historical position. In this way, the respective advantages of different types of sensors of the vehicle can be fully utilized to achieve position consistency detection in various scenarios, avoid missed detection and false detection, and thus improve the robustness of position detection.

[0032] Other features and advantages of the present disclosure will become apparent from the following detailed description of exemplary embodiments of the present disclosure with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0034] Figure 1 It is a schematic diagram of an intelligent network connection system to which the method provided by the embodiment of the present disclosure can be applied.

[0035] Figure 2 is based on Figure 1 The illustrated embodiment provides a schematic diagram of a vehicle.

[0036] Figure 3 It is a schematic diagram of various vehicle driving route scenarios provided by an embodiment of the present disclosure.

[0037] Figure 4 It is a flow chart of a position detection method provided by an embodiment of the present disclosure.

[0038] Figure 5 It is a flow chart of a position detection method provided by an embodiment of the present disclosure.

[0039] Figure 6 It is a flow chart of a vehicle control method provided by an embodiment of the present disclosure.

[0040] Figure 7 It is a structural schematic diagram of an electronic device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0041] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that the relative arrangement of components and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present disclosure unless otherwise specifically stated.

[0042] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the present disclosure, its application, or uses.

[0043] Technologies, methods and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the above-mentioned technologies, methods and equipment should be considered as part of the specification.

[0044] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0045] It should be noted that like reference numerals and letters refer to similar items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0046] First, the application scenarios of the embodiments of the present disclosure are described.

[0047] Figure 1 1 is a schematic diagram of an intelligent network system 100 to which the method provided by the embodiment of the present disclosure can be applied. Figure 1 As shown, the intelligent network connection system 100 may include: a vehicle 101, a server 102, and a user terminal 103.

[0048] In some examples, vehicle 101 may be a vehicle with an autonomous driving function. Autonomous driving is also known as unmanned driving or intelligent driving. Vehicles with autonomous driving functions can achieve driving tasks such as environmental perception, decision planning, and control execution. The level of autonomous driving can refer to the automotive intelligence classification standards formulated by the Society of Automotive Engineers (SAE). For example, L0 is manual driving, L1 is assisted driving, L2 is partial autonomous driving, L3 is conditional autonomous driving, L4 is highly autonomous driving, and L5 is fully autonomous driving. The above classification method for autonomous driving levels is only used as an example, and the disclosed embodiments do not limit the classification standards and levels of autonomous driving.

[0049] In some examples, the server 102 may be a single server or a distributed server cluster consisting of multiple servers, and its deployment method may include a local server or a cloud server. The server 102 may communicate with the vehicle 101 and / or the user terminal 103 based on a communication network, and provide various services for the vehicle 101 and / or the user terminal 103. For example, the server may receive the perception data sent by the vehicle, and provide the vehicle with services such as high-precision maps, data analysis, and decision-making planning. For another example, the server may receive query instructions or control instructions sent by the user terminal, and provide corresponding services to the user.

[0050] In some examples, the user terminal 103 may be any form of electronic device that provides services to users, such as a personal computer, a laptop, a smart tablet, a smart phone, a smart wearable device, etc. The user may interact with the vehicle or server through the human-computer interaction terminal configured in the vehicle 101, or may interact with the vehicle or server through the user terminal 103, for example, querying the status and / or parameters of the vehicle through the user terminal, or controlling the vehicle to perform set tasks and / or modify configuration parameters, etc.; wherein the user terminal runs an application based on the intelligent network connection system to achieve interaction with the vehicle or server, and the application may be a local application, a web application, a small program, etc., which is not limited here.

[0051] In some examples, the above-mentioned application running on the user terminal can provide authentication or authorization services for the user. The user who is successfully authenticated and granted the corresponding authority can query and / or control the vehicle within the scope of the granted authority.

[0052] The vehicle 101, the server 102 and the user terminal 103 can communicate with each other through the communication link provided by the communication network 104. The communication network 104 may include one or more networks of any type, for example, the communication network 104 may include the Internet, a local area network (LAN), a wide area network (WAN), a virtual private network (VPN), a public switched telephone network (PSTN), a satellite communication network, Wi-Fi, 2G, 3G, 4G, 5G, 6G, NB-IoT, eMTC, infrared, Bluetooth, NFC and other networks providing communication, or a combination of the above multiple networks. The communication networks between the vehicle 101 and the server 102, between the user terminal 103 and the server 102, and between the user terminal 103 and the vehicle 101 may be the same or different.

[0053] It should be noted that Figure 1The structure of the intelligent network connection system 100 shown in the figure is only illustrative. The intelligent network connection system in the embodiment of the present disclosure is not limited to the above structure, and may include more or fewer devices as needed, and may also combine or split the devices. For example, the intelligent network connection system may not include a user terminal and / or a server; for another example, the user terminal and the server may be deployed together.

[0054] Figure 2 is based on Figure 1 The illustrated embodiment provides a schematic diagram of a vehicle 101. Figure 2 As shown, the vehicle 101 may include a perception component 1011, a computing platform 1012, an execution component 1013, etc. The perception component 1011, the computing platform 1012, and the execution component 1013 may be connected via a bus or other means.

[0055] In some examples, the perception component 1011 can be used to collect information about the vehicle itself or outside. The perception component 1011 can include at least one of a visual sensor unit, a radar, a positioning and navigation unit, an inertial measurement unit (IMU), or other sensor units, wherein the visual sensor unit can include one or more cameras, and the radar can include at least one of a laser radar, a millimeter wave radar, an ultrasonic radar, or other radars. The positioning and navigation unit can be used to receive signals from a global navigation satellite system (GNSS) to achieve precise positioning of the vehicle. The global navigation satellite system can include one or more of a global positioning system (GPS), a BeiDou navigation satellite system (BDS), and a GLONASS satellite navigation and positioning system (GLONASS).

[0056] In some examples, the computing platform 1012 may include a device with computing capabilities, which is used to process the perception information collected by the perception component 1011 to obtain control information, and send corresponding control instructions to the execution component 1013, so that the execution component 1013 performs corresponding actions, thereby realizing control of the vehicle 101. For example, the computing platform 1012 can perform positioning and mapping (Simultaneous Localization and Mapping, SLAM), path planning, behavior decision-making and other behaviors on the vehicle, thereby realizing autonomous control of the vehicle. The computing platform 1012 may include at least one processor and at least one memory, and each processor may execute instructions stored in the memory individually or collectively to implement the method provided in the embodiment of the present disclosure. The processor in the embodiment of the present disclosure may include a central processing unit (CPU), a graphic processing unit (GPU), a neural-network processing unit (NPU), a tensor processing unit (TPU), a data processing unit (DPU), a digital signal processor (DSP), a field programmable gate array (FPGA), a system on chip (SOC), an application specific integrated circuit (ASIC), a microcontroller unit (MCU) or at least one of other processors. The memory can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic memory, a flash memory, a disk or an optical disk. In addition to storing instructions, the memory can also store data, such as high-precision maps, path information, vehicle location, direction, speed and other data. The data stored in the memory can be retrieved and used by the processor.

[0057] In some examples, the vehicle's computing platform can independently perform computing tasks or communicate with a server to complete computing tasks. For example, the vehicle's computing platform can cooperate with the server to complete corresponding computing tasks.

[0058] The computing platform 1012 may be arranged in the vehicle 101, and part or all of the computing platform 1012 may also be arranged in a server corresponding to the vehicle. For example, some functions of the computing platform 1012 with higher real-time requirements may be arranged in the vehicle, and other functions with lower real-time requirements may be arranged in the server corresponding to the vehicle.

[0059] In some examples, the execution component 1013 is used to execute corresponding actions based on the control of the computing platform 1012, so that the vehicle 101 completes the moving task. The execution component 1013 may include, for example, a power component, a brake component, a transmission component, a steering component, and the like.

[0060] It should be noted that Figure 2 The structure of the vehicle 101 shown in the figure is only illustrative. The vehicle in the embodiment of the present disclosure is not limited to the above structure, and may include more or fewer components as needed, and the devices may be combined or split. For example, the vehicle may not include the above computing platform. For another example, the vehicle may also include a communication component, an interface component, a multimedia component, an input component, an output component, etc.

[0061] In autonomous driving technology, vehicles can collect environmental data through sensors, use SLAM technology to map the surrounding environment during movement, and realize the positioning of the vehicle itself based on the mapping results. However, due to errors in the mapping and positioning process, the accumulation of errors will lead to unreliable mapping and positioning results.

[0062] To eliminate such errors, the loop detection function can be introduced. Loop detection, also known as closed loop detection or position consistency detection, can associate current data with historical data by identifying the locations that the vehicle has reached before, so as to optimize the vehicle's mapping and positioning process based on historical data, thereby improving the accuracy and robustness of mapping and positioning. Therefore, loop detection is of great significance to the positioning and mapping of autonomous vehicles.

[0063] In related technologies, traditional image features or deep learning features can be extracted to construct a visual feature container, and then the features of the current frame are similar to the historical features in the container. If the similarity exceeds a certain threshold, it is considered that a loop is detected. However, various types of routes may appear during vehicle driving. Figure 3 is a schematic diagram of various vehicle driving route scenarios provided by the embodiments of the present disclosure, such as Figure 3As shown, when a vehicle is driving in a parking lot, there may be a variety of vehicle driving routes, thus forming a variety of loop scenarios, including same-direction loop 3A, reverse loop 3B, and cross loop 3C (also known as point loop). Among them: the same-direction loop means that the vehicle passes through a certain place or path again in the same direction as before during driving, that is, the direction of movement of the vehicle is consistent with the direction when it passed the same place before. The reverse loop means that the vehicle passes through a certain place or path in the opposite direction of the previous direction during driving, that is, the direction of movement of the vehicle is opposite to the direction when it passed the same place before. The cross loop involves the vehicle switching from one path to another perpendicular path at an intersection or similar environment, that is, the direction of movement of the vehicle differs from the direction when it passed the same place before by a certain angle, such as 90 degrees or more. However, the current loop detection scheme is difficult to adapt to various loop scenarios, and loop missed detection and false detection are prone to occur. In addition, the amount of calculation is large, more resources are required, and it cannot be detected in real time on the vehicle side.

[0064] In response to the problems of related technologies, the embodiments of the present disclosure provide a position detection method, which organically combines the perception data collected by multiple different types of sensors of the vehicle, fully utilizes the respective advantages of different sensors, and thus can realize real-time detection of various loop closure scenarios, greatly improving the robustness of the loop closure detection scheme.

[0065] Figure 4 is a flow chart of a position detection method provided by an embodiment of the present disclosure. The position detection method can be Figure 1 The vehicle and / or server are executed as shown. Figure 4 As shown, the position detection method of this embodiment may include the following steps S410 to S440.

[0066] Step S410, obtaining a set of perception data collected by multiple different types of sensors of the vehicle.

[0067] Each set of perception data may include a plurality of perception data collected by sensors of the same type at different times within a preset time.

[0068] In some examples, the multiple different types of sensors may include image sensors and position sensors, and the multiple perception data sets may include image data sets collected by the image sensors and position data sets collected by the position sensors. In this way, the image data sets collected by the image sensors and the position data sets collected by the position sensors of the vehicle may be obtained in this step.

[0069] In some examples, the preset time may be any continuous time period, such as a continuous time period including a target time and a historical time period, and the target time period may be the current time period or the target time period for position detection. For example, the preset time period may be a time period from vehicle startup to the current time period; the preset time may also be a preset time period starting from the current time period, such as the current time period and 30 minutes before, or the current time period and three hours before.

[0070] In some examples, multiple different types of sensors of a vehicle may periodically collect corresponding perception data respectively, and the above perception data set may be a set obtained by arranging the perception data collected multiple times in chronological order. Optionally, the perception data collected by different types of sensors may be aligned according to timestamps (such as collection time), and the perception data collected at different times may be perception data corresponding to different timestamps. For example, based on a preset baseline timing or taking the collection time sequence of one type of sensor as the baseline timing (that is, the baseline timing includes multiple collection times), the perception data collected by other types of sensors may be aligned to the baseline timing, so that the perception data of different types of sensors may be aligned to the same timestamp sequence for joint processing. Among them, the timestamp alignment method may be to set the aligned collection time of the perception data to the time closest to the actual collection time of the perception data on the baseline timing, so that the perception data aligned to the same timestamp of the baseline timing may have the same or similar collection time, for example, the time difference of the collection time is less than or equal to a specific time threshold. It should be noted that the collection period of the perception data collected by different types of sensors may be the same or different, and this embodiment does not limit this.

[0071] Step S420: for each perception data set, determine whether the matching detection result corresponding to the perception data set is the first target result according to the similarity of the perception data at different times in the perception data set.

[0072] The first target result can be used to characterize that the similarity between the target position of the vehicle at the target moment and the historical position of the vehicle at the historical moment is greater than or equal to a first preset threshold, and the historical moment is a historical moment earlier than the target moment within a preset time. It should be noted that the similarity can be the degree of proximity or matching between the two geographical locations (the target position and the historical position) in a specific feature, and the specific feature can be a geographical location coordinate and / or a visual image feature.

[0073] In some examples, the target moment may be the current moment, such as the moment when the vehicle last collected sensory data. Optionally, the time difference between the historical moment and the target moment may be greater than a preset time threshold (e.g., 1 second or 2 seconds). For example, the historical moment may be searched for in multiple moments before the target moment that are 2 seconds apart.

[0074] In some examples, the target moment may be N consecutive moments, where N is a pre-set positive integer greater than 1, for example, N may be equal to 10. Similarly, the historical moment may also be N consecutive moments. In this way, the matching detection results of N consecutive frames of perception data are all the first target results, and then the matching detection result corresponding to the perception data set is determined to be the first target result, thereby achieving continuity detection and further improving the robustness of position matching detection.

[0075] Step S430, when the matching detection result corresponding to the first perception data set among the multiple perception data sets is the first target result, determine whether the matching detection result corresponding to the second perception data set is the second target result.

[0076] The second perception data set may be a perception data set other than the first perception data set in the plurality of perception data sets, and the second target result may be used to characterize that the similarity between the target location and the historical location is greater than or equal to a second preset threshold. The second preset threshold may be the same as or different from the first preset threshold, for example, the second preset threshold may be less than the first preset threshold to provide a looser matching detection condition.

[0077] In some examples, the first perception data set and the second perception data set are acquired by sensors of different types. For example, the first perception data set is an image data set acquired by an image sensor, and the second perception data set is a position data set acquired by a position sensor. For another example, the first perception data set is a position data set acquired by a position sensor, and the second perception data set is an image data set acquired by an image sensor.

[0078] Step S440, when the matching detection result corresponding to the second perception data set is the second target result, it is determined that the target position is the same as the historical position.

[0079] In some examples, the target position is the same as the historical position, that is, the target position of the vehicle is a historical position that has been reached before, which indicates that the vehicle has looped. In this way, the current data of the vehicle's current position can be associated and optimized based on the historical data corresponding to the historical position (such as position information, environmental information, vehicle posture, planned path, speed and acceleration control, etc.) to improve the robustness of vehicle positioning and control.

[0080] By adopting the method of steps S410 to S440, a new position detection method is provided, including: obtaining a set of perception data collected by multiple different types of sensors of the vehicle in step S410, detecting each perception data set in step S420, thereby avoiding missed detection, and further, if it is determined that the matching detection result of the first perception data set is the first target result, then determining whether the matching detection result of the second perception data set is the second target result in step S430, the first perception data set and the second perception data set can be collected by different types of sensors, and the detection conditions of the first target result and the second target result can be different, that is, cross-validation is performed in step S430, thereby avoiding false detection, and finally determining that the target position is the same as the historical position when the matching detection result corresponding to the second perception data set is the second target result. In this way, the respective advantages of different types of sensors of the vehicle can be fully utilized to realize position consistency detection in various scenarios, avoid missed detection and false detection, and thus improve the robustness of position detection.

[0081] In some embodiments of the present disclosure, the above-mentioned multiple different types of sensors may include an image sensor and a position sensor, and the multiple perception data sets may include an image data set collected by the image sensor and a position data set collected by the position sensor.

[0082] In one implementation, the position sensor may include a positioning navigation unit and / or an inertial measurement unit (IMU).

[0083] In one implementation, the position sensor may include a positioning and navigation unit, which may be used to receive signals from a global navigation satellite system GNSS to achieve precise positioning of the vehicle and record the absolute position data of the vehicle (e.g., global trajectory points based on longitude and latitude coordinates). Among them, GNSS may include one or more of GPS, GLONASS, and Beidou satellite navigation systems. The position data obtained based on the positioning and navigation unit may be displayed or stored in the form of (x1, y1, z1), where x1 may be the longitude of the vehicle's position, y1 may be the latitude of the vehicle's position, and z1 may be the vehicle's heading angle (yaw), which may also be referred to as a direction angle or yaw angle, for example, which may characterize the rotation angle of the vehicle's forward direction of motion relative to a reference direction, and the reference direction may be geographic north.

[0084] In another implementation, the position sensor may include an inertial measurement unit (IMU). The IMU can be used to locate the vehicle and record the relative position data of the vehicle (such as local track points, also called odometer track points). For example, the IMU may include an accelerometer and a gyroscope, which are used to measure the acceleration and angular velocity of the vehicle, achieve relative positioning of the vehicle, and obtain multiple odometer track points starting from a reference point, which may be the initial position point of the vehicle. Optionally, the odometer track of the vehicle can be determined based on the IMU and the chassis information of the vehicle. For example, the acceleration and angular velocity of the IMU strategy and the chassis information of the vehicle can be input into the odometer of the vehicle, and the track points output by the odometer are used as the odometer track of the vehicle. The chassis information may include at least one of the wheel diameter or radius, body size, speed, steering angle, and other information of the vehicle. The position data obtained based on the inertial measurement unit can be displayed or stored in the form of (x2, y2, z2), where x2 can be the x-axis coordinate of the vehicle position based on the vehicle coordinate system or the global coordinate system, y2 can be the y-axis coordinate of the vehicle position based on the vehicle coordinate system or the global coordinate system, and z can be the heading angle (yaw) of the vehicle, which can also be called the direction angle or yaw angle, for example, which can represent the rotation angle of the vehicle's moving direction relative to a reference direction, and the reference direction can be the x-axis or y-axis direction.

[0085] In another implementation, the position sensor may include a positioning and navigation unit and an inertial measurement unit (IMU). In this way, accurate positioning of the vehicle in different scenarios can be achieved. For example, in outdoor scenarios, the positioning and navigation unit can receive GNSS signals that meet preset signal conditions to achieve global positioning of the vehicle. In indoor scenarios (such as indoor parking lots), if the GNSS signal becomes weak, the positioning and navigation unit cannot achieve accurate vehicle positioning, and the vehicle can be positioned through the IMU. In this way, through the cooperation of the positioning and navigation unit and the IMU, the vehicle can be accurately positioned in various scenarios.

[0086] In one implementation, the image sensor may include a camera of the vehicle, and the image data set may be a set of multiple frames of environment images captured by the camera. For example, the image sensor is a front-view camera of the vehicle, and the image data set includes multiple frames of front-view environment images captured by the front-view camera; for another example, the image sensor is a rear-view camera of the vehicle, and the image data set includes multiple frames of rear-view environment images captured by the rear-view camera; for another example, the image sensor is a camera of multiple different viewing angles of the vehicle, and the image data set includes panoramic scanning images captured by cameras of different viewing angles.

[0087] In another implementation, the image sensor may include a laser radar of the vehicle, and the image data set may be a set of multi-frame point cloud images obtained based on laser radar detection.

[0088] In another implementation, the image sensor may include a camera and a lidar of the vehicle, and the image data set may be a fused image obtained by fusing an environment image taken by the camera with a point cloud image detected by the lidar.

[0089] In one implementation, the image sensor of the vehicle may include only a front-view camera. In the related art, if only the image output by the front-view camera is used for position consistency detection, due to the problem of camera viewing angle, Figure 3 The reverse loop and point loop scenes shown in the figure have a high probability of missed detection. However, in this embodiment, the perception data obtained by the two types of sensors, the front-view camera and the position sensor, respectively detect the consistency of the target position and the historical position, which can avoid missed detection and improve the accuracy of position detection. Furthermore, the two position sensors used, the positioning navigation unit and the inertial measurement unit, are both standard sensors for vehicles, and there is no need to add additional hardware, which reduces the cost of vehicle position detection.

[0090] For example, a vehicle may be equipped with a positioning navigation unit, an inertial measurement unit, and a forward-looking camera, which may output global position data, local position data, and image data, respectively, thereby implementing the position detection method of the embodiment of the present disclosure without the need for additional sensors.

[0091] In one implementation, the vehicle's image sensor may include a panoramic camera, which acquires a panoramic image. The panoramic image is processed to obtain a feature map based on a bird's-eye view, such as a BEV (Bird's Eye View) feature map. Based on the feature map, a feature descriptor in the form of BEV can be extracted to perform image comparison and determine whether the target position and the historical position are the same, thereby further improving the robustness of the detection.

[0092] The perception data set in the embodiments of the present disclosure may include an image data set obtained based on an image sensor and / or a position data set obtained based on a position sensor. For different perception data sets, the implementation method of determining whether the matching detection result corresponding to the perception data set is the first target result in the above step S402 may be different.

[0093] In some embodiments of the present disclosure, the above-mentioned perception data set may include a location data set acquired based on a location sensor, and whether the matching detection result corresponding to the perception data set (i.e., the location data set) is the first target result may be determined in the following manner:

[0094] If a historical location that meets the first preset location condition with the target location is searched in the location data set, it is determined that the similarity between the target location and the historical location is greater than or equal to a first preset threshold, so as to determine that the matching detection result corresponding to the location data set is the first target result.

[0095] The first preset location condition may include at least one of the following conditions 1 to 4:

[0096] Condition 1: The distance between the target location and the historical location is less than or equal to a first preset distance threshold.

[0097] The first preset distance threshold may be any preset value. For example, the first preset distance threshold may be 1 meter, 3 meters or 5 meters, which may be preset according to the accuracy requirement of the position detection. Optionally, the distance between the target position and the historical position may be a distance obtained by Euclidean distance, straight-line distance or other calculation methods.

[0098] Condition 2: The cumulative change in the heading angle of the vehicle during the process of traveling from the historical position to the target position is greater than or equal to the first preset angle threshold.

[0099] The heading angle may also be referred to as the direction angle, and the first preset angle threshold may be any preset value. For example, the first preset angle threshold may be 90 degrees, 180 degrees, or 270 degrees. The accumulated change in the heading angle of condition 2 may avoid misdetection caused by vehicle reversing, adapt to various scenarios, and further improve the robustness of position detection in different scenarios.

[0100] Condition three: the cumulative change in slope when the vehicle travels from the historical position to the target position is less than or equal to the first preset slope threshold.

[0101] The first preset slope threshold may be any preset value. For example, the first preset slope threshold may be 1 degree, 5 degrees, or 10 degrees. The accumulated slope change in condition 3 may be used to adapt to multi-storey parking lot scenarios, thereby preventing locations at the same longitude and latitude in different layers from being mistakenly identified as the same location, and further improving the robustness of location detection in different scenarios.

[0102] In some examples, the slope of the vehicle can be detected by the inertial measurement unit of the vehicle, and the change in the slope can be accumulated. For example, the acceleration and angular velocity of the vehicle can be measured by the inertial measurement unit, and the slope of the vehicle can be estimated by the acceleration and angular velocity. It should be noted that the detection and statistics of the cumulative change in the slope can also refer to the implementation methods in the relevant technology, and the embodiments of the present disclosure are not limited to this.

[0103] Condition 4: The height difference between the target position and the historical position is less than or equal to the first preset height threshold.

[0104] Among them, the first preset height threshold can be any preset value. For example, the first preset height threshold can be 1 meter, 1.5 meters or 3 meters, etc. For another example, the first preset height threshold can be set to different values ​​according to the loop length, and the loop length can be the mileage of the vehicle passing the same location twice. Through the height difference in condition four, scenes such as multi-storey parking lots and overpasses can be adapted to avoid locations with different heights but the same longitude and latitude being mistakenly identified as the same location, further improving the robustness of location detection in different scenarios.

[0105] In some examples, the height of the vehicle may be detected by an inertial measurement unit of the vehicle, or may be detected by a positioning navigation unit.

[0106] Optionally, the height may be an elevation, ie, the height of the vehicle relative to sea level, and the height difference may be an elevation difference.

[0107] The first preset location condition may be the above-mentioned condition 1, or a combination of condition 1 and conditions 2, 3, and 4. For example, the first preset location condition may include only condition 1; for another example, the first preset location condition may include condition 1 and condition 2; for another example, the first preset location condition may include condition 1 and condition 3; for another example, the first preset location condition may include condition 1 and condition 4; for another example, the first preset location condition may include condition 1, condition 2, and condition 3; for another example, the first preset location condition may include condition 1, condition 2, and condition 4; for another example, the first preset location condition may include condition 1, condition 2, condition 3, and condition 4, and the combination thereof is not limited to the above-mentioned combination.

[0108] In some examples, a search may be performed in the location data set based on a nearest neighbor search algorithm to determine whether there is a historical location that satisfies the first preset location condition between the target location and the location.

[0109] For example, the location data at multiple different moments can be sampled (e.g., storing one location data every 0.2 m), and stored in container T1 as historical location data. For the location data at any target moment (e.g., the current moment), it is possible to search in container T1 based on the nearest neighbor search whether there is a historical location that meets the above-mentioned first preset location condition, for example, the Euclidean distance between the historical location and the current target location is less than a certain first preset distance threshold (e.g., 3.0 m), and the cumulative change in the heading angle of the vehicle between the historical location and the current location is greater than or equal to the first preset angle threshold (e.g., 180 degrees); if there is a historical location that meets the above-mentioned first preset location condition, it can be determined that the similarity between the target location and the historical location is greater than or equal to the first preset threshold, that is, it can be determined that the matching detection result corresponding to the location data set is the first target result; conversely, if there is no historical location that meets the above-mentioned first preset location condition, it can be determined that the matching detection result corresponding to the location data set is not the first target result.

[0110] The above nearest neighbor search can be performed based on a k-dimensional tree (Kdtree) or other search algorithms. For example, the location data (x, y, yaw) at multiple different times, where (x, y) is the plane coordinate of the location point, and yaw is the heading angle corresponding to the location point, can be used for the nearest neighbor search. (x, y) of the location data is constructed by recursively dividing the two-dimensional space into smaller areas based on Kdtree. For example, in the two-dimensional space, for each node of Kdtree, a dimension (for example, x or y) can be selected, and then the location points are divided into two groups according to the median of the dimension: one group is on the left side of the median, and the other group is on the right side, and a Kdtree containing a root node and a leaf node is constructed. In this way, when searching for the nearest neighbor, starting from the root node of Kdtree, the query point is compared with the value of the current node on the split dimension, and then the left subtree or the right subtree is searched recursively, so that the historical position closest to the target position can be quickly found. Furthermore, if the historical location satisfies the first preset location condition, it can be determined that the similarity between the target location and the historical location is greater than or equal to the first preset threshold, that is, the matching detection result corresponding to the location data set can be determined as the first target result.

[0111] In some examples, based on a nearest neighbor search algorithm, the historical location that is closest to the target location at the target moment can be determined from the location data at different moments contained in the location data set; when the target location and the historical location satisfy the above-mentioned first preset location condition, it is determined that the similarity between the target location and the historical location is greater than or equal to the first preset threshold, so as to determine that the matching detection result corresponding to the location data set is the first target result.

[0112] It should be noted that the position data may be the position data of the vehicle detected by a positioning navigation unit and / or an inertial measurement unit of the vehicle.

[0113] In some examples, if it is determined that the matching detection result corresponding to the position data set is the first target result, the first perception data set in the above step S430 is the position data set obtained by the position sensor, and the second perception data set is the image data set obtained by the image sensor.

[0114] In this way, after a preliminary position detection is performed based on the position sensor, that is, when the matching detection result of the position data set is the first target result, cross-validation is performed based on the image sensor, that is, when the matching detection result of the image data set is the second target result, it is determined that the target position is the same as the historical position, which can avoid missed detection and false detection and improve the robustness of position consistency detection.

[0115] In some embodiments of the present disclosure, the above-mentioned perception data set may include an image data set acquired based on an image sensor, and whether the matching detection result corresponding to the perception data set (i.e., the image data set) is the first target result may be determined by the following steps S4201 to S4202:

[0116] Step S4201, sampling the image data at different times in the image data set to obtain the image features corresponding to each image data.

[0117] The image feature may include a feature description vector of the entire or partial region of the image data.

[0118] In some examples, key points (such as corner points, edges or center points, etc.) of image data can be extracted based on a preset feature extraction algorithm, and a feature description vector corresponding to the image data is generated as its image feature. The key point can be a key point of the entire area of ​​the image or a key point of a partial area. The preset feature extraction algorithm can be a scale-invariant feature transform (SIFT), a histogram of oriented gradients (HOG), or other image feature extraction algorithms based on deep neural networks.

[0119] In other examples, a scene global feature descriptor (i.e., the above-mentioned image feature) is obtained based on a visual place recognition (Visual Place Recognition) model. For example, the image feature can be a 128-dimensional global feature descriptor vector, which can achieve lightweight detection, thereby concisely and comprehensively representing the features of the image for comparison between images.

[0120] Step S4202: If historical moment image data having an image feature similarity greater than or equal to a first preset image similarity threshold with the target moment image data is searched in the image data set, the matching detection result corresponding to the image data set is determined to be the first target result.

[0121] The first preset image similarity threshold may be any preset similarity threshold. Optionally, the first preset image similarity threshold may be a value greater than or equal to half of the theoretical maximum similarity. For example, the similarity between images may be a value between 0 and 1, and the first preset image similarity threshold may be any preset value greater than or equal to 0.5, such as 0.5, 0.8 or 0.9.

[0122] In some examples, the moment with the highest image feature similarity to the image data at the target moment can be obtained from the image data at multiple moments as the historical moment. If the similarity between the image features of the image data at the historical moment and the image features of the image data at the target moment is greater than or equal to a first preset image similarity threshold, the matching detection result corresponding to the image data set is determined to be the first target result.

[0123] For example, the image features corresponding to the image data at multiple different moments can be sampled (such as storing a vector every 0.2m) and stored in container T2; for the image data at any target moment (such as the current moment), the feature similarity can be calculated from the image features in container T2. If the similarity is greater than a first preset image similarity threshold, the matching detection result of the target position in the perception data set can be determined as the first target result.

[0124] In this way, the image features can be represented concisely and comprehensively based on the feature description vector, thereby realizing lightweight similarity matching between images, reducing the amount of computational complexity of image data matching, and improving the matching detection efficiency of image data, thereby improving the real-time performance of position detection, making it easier to deploy and implement the position detection method on the vehicle side.

[0125] In some examples, the image sensor in this embodiment can be a front-view camera of the vehicle. Thus, compared with the use of a panoramic camera, the use of a front-view camera can further reduce the amount of image processing operations, further improve the real-time performance of position detection, and be lower in cost.

[0126] In some examples, if it is determined that the matching detection result corresponding to the image data set is the first target result, the first perception data set in the above step S430 is the image data set obtained by the image sensor, and the second perception data set is the position data set obtained by the position sensor.

[0127] In this way, after performing preliminary position detection based on the image sensor, that is, when the matching detection result of the image data set is the first target result, cross-validation is then performed based on the position sensor, that is, after determining that the matching detection result of the position data set is the second target result, it is determined that the target position is the same as the historical position, which can avoid missed detection and false detection and improve the robustness of position consistency detection.

[0128] In some examples, the manner of determining whether the matching detection result of the location data set is the second target result may be similar to the manner of determining whether the matching detection result of the location data set is the first target result, but the corresponding thresholds may be the same or different.

[0129] For example, if the first perception data set is an image data set acquired by an image sensor, and the second perception data set is a position data set acquired by a position sensor, then it can be determined whether the matching detection result corresponding to the second perception data set (i.e., the position data set) is the second target result in the following manner:

[0130] If a historical location that meets the second preset location condition with the target location is searched in the location data set, it is determined that the similarity between the target location and the historical location is greater than or equal to the second preset threshold, so as to determine that the matching detection result corresponding to the location data set is the second target result.

[0131] The second preset location condition may include at least one of the following:

[0132] The distance between the target location and the historical location is less than or equal to a second preset distance threshold, and the second preset distance threshold may be greater than or equal to the first preset distance threshold;

[0133] The cumulative change in the heading angle of the vehicle during the process of traveling from the historical position to the target position is greater than or equal to a second preset angle threshold, and the second preset angle threshold may be the same as or different from the first preset angle threshold;

[0134] The cumulative change in slope during the process of the vehicle traveling from the historical position to the target position is less than or equal to a second preset slope threshold, wherein the second preset slope threshold may be the same as or different from the first preset slope threshold.

[0135] In this way, when other conditions remain unchanged, if the second preset distance threshold is greater than the first preset distance threshold, for example, the first preset distance threshold is 3 meters and the second preset distance threshold is 4 meters, then in the cross-validation, more relaxed conditions can be used for the position data set collected by the position sensor, and the position consistency detection can also be accurately realized in combination with the image data collected by the image sensor, thereby improving the robustness of the position consistency detection.

[0136] In other examples, the manner of determining whether the matching detection result of the image data set is the second target result may be similar to the manner of determining whether the matching detection result of the image data set is the first target result, but the corresponding thresholds may be the same or different.

[0137] For example, if the first perception data set is a position data set acquired by a position sensor, and the second perception data set is an image data set acquired by an image sensor, it can be determined whether the matching detection result corresponding to the second perception data set (i.e., the image data set) is the second target result in the following manner:

[0138] The image data at different moments in the image data set are sampled to obtain image features corresponding to the image data; if image data at a historical moment whose image feature similarity with the image data at the target moment is greater than or equal to a second preset image similarity threshold value is searched in the image data set, the matching detection result corresponding to the image data set is determined to be the first target result. The second preset image similarity threshold value may be less than or equal to the first preset image similarity threshold value.

[0139] In this way, when other conditions remain unchanged, if the second preset image similarity threshold is greater than the first preset image similarity threshold, for example, the first preset image similarity threshold is 0.8 and the second preset image similarity threshold is 0.6, then in the cross-validation, more relaxed conditions can be used for the image data set collected by the image sensor, and the position consistency detection can also be accurately realized in combination with the position data collected by the position sensor, thereby improving the robustness of the position consistency detection.

[0140] The above-mentioned position detection method can utilize lightweight position information (trajectory) and visual feature descriptor information of the image for position detection, requires low computing power, and can improve the real-time performance of position detection; utilizes the vehicle's standard sensors (such as positioning navigation unit and / or inertial measurement unit), without the need for additional hardware, thereby reducing the cost of loop detection for the vehicle; and supports a richer range of loop closure scene types, thereby improving the robustness of detection.

[0141] In some examples, the position detection method provided by the embodiments of the present disclosure can be executed by the vehicle side due to its higher real-time performance and lower cost. For example, it can be applied to loop detection in vehicle-side mapping, positioning, or other business processes of autonomous driving.

[0142] Figure 5 is a flow chart of a position detection method provided by an embodiment of the present disclosure. The position detection method can be Figure 1 The vehicle and / or server are executed as shown. Figure 5 As shown, the position detection method of this embodiment may include the following steps S510 to S560.

[0143] Step S510, obtaining a set of perception data collected by multiple different types of sensors of the vehicle.

[0144] The multiple sensors of different types may include a positioning navigation unit, an inertial measurement unit, and an image sensor, and the perception data set may include the following:

[0145] Based on the first position data set collected by the positioning and navigation unit, for example, the absolute position data (i.e., global trajectory points) at multiple moments collected by the positioning and navigation unit within a preset time can be obtained;

[0146] A second position data set collected by an inertial measurement unit, for example, relative position data (i.e., local trajectory points) at multiple moments collected by the inertial measurement unit within a preset time;

[0147] The image data set collected by the image sensor may, for example, be based on image data collected by the image sensor at multiple moments within a preset time.

[0148] Step S520: Based on the first position data set collected by the positioning and navigation unit, determine whether the matching detection result of the first position data set is a first target result.

[0149] Among them, the first target result can be used to characterize that the similarity between the target position of the vehicle at the target moment and the historical position of the vehicle at the historical moment is greater than or equal to a first preset threshold, and the historical moment is a historical moment earlier than the target moment within the preset time.

[0150] In some examples, the specific method of determining whether the matching detection result of the first location data set is the first target result can refer to the description of determining whether the matching detection result corresponding to the location data set is the first target result in the aforementioned embodiment of the present disclosure, and will not be repeated here.

[0151] Step S530: Based on the second position data set collected by the inertial measurement unit, determine whether the matching detection result of the second position data set is the first target result.

[0152] In some examples, the specific method of determining whether the matching detection result of the second location data set is the first target result can refer to the description of determining whether the matching detection result corresponding to the location data set is the first target result in the aforementioned embodiment of the present disclosure, and will not be repeated here.

[0153] Step S540: Based on the image data set collected by the image sensor, determine whether the matching detection result of the image data set is a first target result.

[0154] In some examples, the specific method of determining whether the matching detection result of the image data set is the first target result can be referred to the description in the aforementioned embodiments of the present disclosure, and will not be repeated here.

[0155] Step S550: If the matching detection result of any one of the first position data set, the second position data set, and the image data set is the first target result, cross-validation is performed based on other data sets.

[0156] The cross-validation method may be to determine whether the matching detection result corresponding to other data sets is the second target result.

[0157] In some examples, the information synchronization of the data sets can be performed first, and the perception data in the above three data sets can be synchronized to the track points of the second position data set according to the timestamp to obtain at least one position point pair, each of which can include the target position at the target moment and the historical position at the historical moment, and the target position and the historical position form a position point pair. Cross-validation can be performed based on the position point pairs obtained after synchronization, that is, determining whether the matching detection results corresponding to other data sets are the second target results.

[0158] In some examples, the specific method of determining whether the matching detection results corresponding to other data sets are the second target results can refer to the description of determining whether the matching detection results corresponding to the second perception data set are the second target results in the aforementioned embodiment of the present disclosure, and will not be repeated here.

[0159] Step S560: If the cross-validation passes, it is determined that the target location is the same as the historical location.

[0160] Among them, cross-validation can characterize the matching detection results corresponding to other data sets as the second target results.

[0161] By adopting the above method, three types of perception data sets, namely, a first position data set collected by a positioning and navigation unit (for example, global trajectory points), a second position data set collected by an inertial measurement unit (for example, local trajectory points), and an image data set collected by an image sensor, are respectively used for position consistency detection, so that missed detection can be avoided. Further cross-validation is performed based on the three types of perception data sets, and false detection can be encoded, thereby realizing the detection of various loop closure scenarios (such as underground garages without GNSS signals, reverse loops, cross loops, etc.), greatly improving the robustness of the loop detection algorithm, and because the existing sensors on the autonomous driving vehicle are used, the position detection method can be implemented at a low cost.

[0162] Figure 6 is a flow chart of a vehicle control method provided by an embodiment of the present disclosure. The vehicle control method may be Figure 1 The vehicle and / or server are executed as shown. Figure 6 As shown, the position detection method of this embodiment may include:

[0163] Step S610, obtaining a set of perception data collected by multiple different types of sensors of the vehicle.

[0164] Each perception data set includes a plurality of perception data collected by sensors of the same type at different times within a preset time.

[0165] Step S620: Control vehicle driving according to the perception data set.

[0166] Among them, multiple perception data sets can be used to determine whether the target position of the vehicle at the target time is the same as the historical position of the vehicle at the historical time.

[0167] By way of example, for each perception data set, it can be determined whether the matching detection result corresponding to the perception data set is the first target result based on the similarity of the perception data at different moments in the perception data set; wherein the first target result is used to characterize that the similarity between the target position of the vehicle at the target moment and the historical position of the vehicle at the historical moment is greater than or equal to a first preset threshold, and the historical moment is a historical moment that is earlier than the target moment within a preset time; when the matching detection result corresponding to the first perception data set among multiple perception data sets is the first target result, it is determined whether the matching detection result corresponding to the second perception data set is the second target result; wherein the second perception data set is the other perception data set among the multiple perception data sets except the first perception data set, and the second target result is used to characterize that the similarity between the target position and the historical position is greater than or equal to a second preset threshold, and the second preset threshold is less than the first preset threshold; when the matching detection result corresponding to the second perception data set is the second target result, it is determined that the target position is the same as the historical position.

[0168] It should be noted that the specific implementation method of determining whether the target position of the vehicle at the target moment is the same as the historical position of the vehicle at the historical moment based on multiple perception data sets can be referred to the description in the aforementioned embodiments of the present disclosure, and will not be repeated here.

[0169] In some examples, if it is determined based on multiple perception data sets that the target position of the vehicle at the target moment is the same as the historical position of the vehicle at the historical moment, it can be determined that the vehicle has a loop, that is, the target position is a historical position that the vehicle has reached before. Based on the historical data of the historical position, the accumulated errors in the vehicle's driving can be corrected and the accuracy of positioning can be improved to control the vehicle's driving. Furthermore, this method can also improve the consistency of local maps and global maps and effectively adapt to environmental changes.

[0170] In some examples, when the target position is the same as the historical position, the vehicle's posture can be calibrated according to the historical position, and the vehicle can be controlled according to the calibrated posture. For example, the vehicle's posture at the historical position can be used to correct the vehicle's current posture drift, which can refer to the phenomenon that the vehicle's positioning information gradually deviates from the actual position over time due to factors such as sensor errors and environmental changes. If the posture drift is not corrected, the vehicle's positioning will become increasingly inaccurate, affecting the safety and reliability of autonomous driving.

[0171] In some examples, the above-mentioned multiple different types of sensors may be standard intelligent driving sensors on the vehicle side, for example, may include image sensors (forward-looking cameras or panoramic cameras), position sensors (positioning navigation units and / or inertial measurement units IMU), etc.

[0172] In this way, the standard intelligent driving sensors on the vehicle side can be used to detect loop information in real time on the vehicle side and correct posture drift immediately without adding additional hardware configuration. This means that loop detection can be achieved in real time at low cost, thereby improving the safety and reliability of autonomous driving.

[0173] By adopting the above method, the respective advantages of different types of vehicle sensors can be fully utilized to achieve position consistency detection in various scenarios, avoid missed detection and false detection, improve the robustness of position detection, and thus improve the robustness of vehicle control.

[0174] Figure 7 Schematic diagram of the structure of an electronic device provided by an embodiment of the present disclosure. Figure 7 As shown, the electronic device 1000 may include a memory 1010 and a processor 1020. The memory 1010 may be used to store computer instructions, and the processor 1020 may be used to call computer instructions from the memory 1010 to execute all or part of the steps of any method in the aforementioned embodiments of the present disclosure. The processor may be one or more, and the one or more processors may execute instructions individually or together. The memory may also be one or more, and the one or more memories may store the above-mentioned computer instructions individually or together. Optionally, the electronic device may be Figure 1 Servers and / or vehicles in.

[0175] The embodiment of the present disclosure also provides a vehicle, which may include a memory and a processor, wherein the memory may be used to store computer instructions, and the processor may be used to call computer instructions from the memory to execute all or part of the steps of any method in the aforementioned embodiment of the present disclosure. The processor may be one or more, and the one or more processors may execute instructions individually or collectively, and the memory may also be one or more, and the one or more memories may store the above-mentioned computer instructions individually or collectively.

[0176] The vehicle in the foregoing embodiments of the present disclosure may be an electric vehicle, a hybrid vehicle, a fuel cell vehicle, or other types of vehicles. The vehicle may be an autonomous vehicle or a non-autonomous vehicle. For example, the vehicle provided in this embodiment may be Figure 1 or Figure 2 Vehicle shown.

[0177] The embodiments of the present disclosure also provide a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, any method in the aforementioned embodiments of the present disclosure is implemented. Optionally, the computer-readable storage medium may be a non-transitory storage medium, but is not limited thereto, and may also be a transient storage medium.

[0178] The present disclosure also provides a chip, which may include a processing unit, which may be used to execute all or part of the steps of any method in the foregoing embodiments of the present disclosure. The chip may be a chip in the form of an application-specific integrated chip ASIC, a system on chip SOC, a field programmable gate array FPGA, etc., which is not limited in the present embodiment. Optionally, the chip may also include a storage unit, which may be used to store computer instructions, and the processing unit may be used to call computer instructions from the storage unit to execute all or part of the steps of any method in the foregoing embodiments of the present disclosure.

[0179] The embodiments of the present disclosure further provide a computer program product, which may include a computer program. When the computer program is executed by a processor, any method in the aforementioned embodiments of the present disclosure may be implemented.

[0180] The present disclosure may be a system, a method and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement any method in the foregoing embodiments of the present disclosure.

[0181] A computer-readable storage medium may be a tangible device that can hold and store instructions used by an instruction execution device. A computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples of computer-readable storage media (a non-exhaustive list) include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination of the foregoing. As used herein, a computer-readable storage medium is not to be interpreted as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through a wire.

[0182] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.

[0183] The computer program instructions for performing the operation of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, which may include object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. Computer-readable program instructions may be executed entirely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be personalized by utilizing the state information of the computer-readable program instructions, and the electronic circuit may execute the computer-readable program instructions, thereby realizing various aspects of the present disclosure.

[0184] Various aspects of the present disclosure are described herein with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer-readable program instructions.

[0185] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so as to produce a machine, so that when these instructions are executed by the processor of the computer or other programmable data processing device, a device for implementing the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions make the computer, programmable data processing device, and / or other equipment work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0186] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operating steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0187] The flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the systems, methods and computer program products according to multiple embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of an instruction, and the module, a program segment or a part of an instruction contains one or more executable instructions for implementing the specified logical function. In some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions. It should be noted that implementation by hardware, implementation by software, and implementation by a combination of software and hardware are all equivalent.

[0188] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The choice of terms used herein is intended to best explain the principles of the embodiments, practical applications, or technical improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein. The scope of the present disclosure is defined by the appended claims.

Claims

1. A position detection method, characterized in that: The method comprises: Acquire a set of perception data respectively collected by a plurality of different types of sensors of the vehicle; wherein each of the perception data sets includes a plurality of perception data collected by a sensor of the same type at different times within a preset time; For each perception data set, determining whether a matching detection result corresponding to the perception data set is a first target result according to the similarity of the perception data at different times in the perception data set; wherein the first target result is used to characterize that the similarity between the target position of the vehicle at the target time and the historical position of the vehicle at a historical time is greater than or equal to a first preset threshold, and the historical time is a historical time earlier than the target time within the preset time; In the case where the matching detection result corresponding to the first perception data set among the multiple perception data sets is the first target result, determining whether the matching detection result corresponding to the second perception data set is the second target result; wherein the second perception data set is other perception data sets among the multiple perception data sets except the first perception data set, and the second target result is used to characterize that the similarity between the target position and the historical position is greater than or equal to a second preset threshold, and the second preset threshold is less than the first preset threshold; When the matching detection result corresponding to the second perception data set is a second target result, it is determined that the target position is the same as the historical position.

2. The method according to claim 1, characterized in that: The plurality of sensors of different types include image sensors and position sensors, and the position sensors include positioning navigation units and / or inertial measurement units (IMUs).

3. The method according to claim 2, characterized in that The perception data set includes a location data set acquired based on the location sensor; and determining whether a matching detection result corresponding to the perception data set is a first target result according to similarities of perception data at different times in the perception data set includes: If a historical location that satisfies a first preset location condition and that meets the target location is searched in the location data set, determining that the similarity between the target location and the historical location is greater than or equal to a first preset threshold, so as to determine that the matching detection result corresponding to the location data set is the first target result; The first preset location condition includes that the distance between the target location and the historical location is less than or equal to a first preset distance threshold.

4. The method according to claim 3, characterized in that The first preset location condition also includes at least one of the following: The cumulative change in the heading angle of the vehicle during the process of traveling from the historical position to the target position is greater than or equal to a first preset angle threshold; The cumulative change in slope during the process of the vehicle traveling from the historical position to the target position is less than or equal to a first preset slope threshold; The height difference between the target position and the historical position is less than or equal to a first preset height threshold.

5. The method according to claim 3, characterized in that: The first perception data set is a position data set acquired by the position sensor, and the second perception data set is an image data set acquired by the image sensor.

6. The method according to claim 2, characterized in that The perception data set includes an image data set acquired based on the image sensor; and determining whether a matching detection result corresponding to the perception data set is a first target result according to similarities of perception data at different times in the perception data set includes: Sampling the image data at different times in the image data set to obtain image features corresponding to the image data; wherein the image features include feature description vectors of all or part of the regions of the image data; If historical moment image data whose image feature similarity with the target moment image data is greater than or equal to a first preset image similarity threshold is searched in the image data set, the matching detection result corresponding to the image data set is determined to be the first target result.

7. The method according to claim 6, characterized in that The image sensor is a forward-looking camera of the vehicle.

8. The method according to claim 6, characterized in that The first perception data set is an image data set acquired by the image sensor, and the second perception data set is a position data set acquired by the position sensor.

9. The method according to any one of claims 1 to 8, characterized in that The target time is N consecutive time points, where N is a preset positive integer greater than 1.

10. A vehicle control method, characterized in that: The method comprises: Acquire a set of perception data respectively collected by a plurality of different types of sensors of the vehicle; wherein each of the perception data sets includes a plurality of perception data collected by a sensor of the same type at different times within a preset time; Controlling the vehicle to travel according to the set of perception data; wherein the plurality of perception data sets are used to determine whether a target position of a vehicle at a target time is the same as a historical position of the vehicle at a historical time; The method of determining whether the target position of the vehicle at the target time is the same as the historical position of the vehicle at the historical time includes: For each perception data set, determining whether a matching detection result corresponding to the perception data set is a first target result according to the similarity of the perception data at different times in the perception data set; wherein the first target result is used to characterize that the similarity between the target position of the vehicle at the target time and the historical position of the vehicle at a historical time is greater than or equal to a first preset threshold, and the historical time is a historical time earlier than the target time within the preset time; In the case where the matching detection result corresponding to the first perception data set among the multiple perception data sets is the first target result, determining whether the matching detection result corresponding to the second perception data set is the second target result; wherein the second perception data set is other perception data sets among the multiple perception data sets except the first perception data set, and the second target result is used to characterize that the similarity between the target position and the historical position is greater than or equal to a second preset threshold, and the second preset threshold is less than the first preset threshold; When the matching detection result corresponding to the second perception data set is a second target result, it is determined that the target position is the same as the historical position.

11. The method according to claim 10, characterized in that The controlling the vehicle to travel comprises: When the target position is the same as the historical position, calibrating the position and posture of the vehicle according to the historical position; The vehicle is controlled to travel according to the calibrated posture.

12. An electronic device, characterized in that: The invention comprises a memory and a processor, wherein the memory is used to store computer instructions, and the processor is used to call the computer instructions from the memory to execute: the method according to any one of claims 1 to 11.

13. A vehicle, characterized in that: The invention comprises a memory and a processor, wherein the memory is used to store computer instructions, and the processor is used to call the computer instructions from the memory to execute: the method according to any one of claims 1 to 11.

14. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, the computer program implements: the method according to any one of claims 1 to 11.

15. A chip, characterized in that: The chip comprises: the chip comprises a processing unit, and the processing unit is used to execute the method as claimed in any one of claims 1 to 11.

Images