Vehicle way-finding positioning method and device and electronic equipment

Abstract

The invention provides a vehicle path-finding positioning method and device and electronic equipment, and relates to the technical field of auxiliary driving, and the method comprises the steps: determining a predicted pose of a vehicle according to the obtained inertial measurement data and wheel speed data of the vehicle; extracting a specific identifier from the obtained surround view image of the vehicle; projecting the specific identifier to a map coordinate system to obtain a map identifier; determining a matching pose according to the map identifier; fusing the predicted pose and the matched pose to obtain a fused pose; and based on an established way-finding map, the vehicle is positioned through the fusion pose, so that the positioning precision of the vehicle during automatic parking is improved, and the memory parking function of the vehicle is ensured to be realized.

Description

Technical Field

[0001] The present invention relates to the field of assisted driving technology, and in particular to a vehicle path finding and positioning method, device and electronic equipment. Background Art

[0002] As the demand for autonomous driving grows, major vehicle companies are engaged in fierce technological competition. Autonomous driving technology has become a hot area of ​​research for scientific researchers. Autonomous driving is divided into different levels according to the degree of automation. At present, L2-level autonomous driving technology has been widely used in the products of many manufacturers. And L3-level autonomous driving products are in the stage of gradual mass production. In the L3 level, an important function is memory parking (Home-zone Parking Assist, referred to as HPA). This function allows the vehicle to autonomously complete automatic parking operations from the parking lot entrance to the designated parking space, and from the parking space to the parking lot exit in a specific area without human intervention.

[0003] Memory parking is mainly divided into four sections: mapping and positioning, environmental perception, vehicle planning, and vehicle control. In some memory parking technology solutions, mapping and positioning rely on satellite signals. When the satellite signal is weak, the positioning effect is not ideal. Summary of the invention

[0004] The problem solved by the present invention is how to improve the positioning accuracy of a vehicle during automatic parking.

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

[0006] In a first aspect, the present invention provides a vehicle path finding and positioning method, comprising:

[0007] Determine the predicted position and posture of the vehicle based on the acquired inertial measurement data and wheel speed data of the vehicle;

[0008] Extracting a specific identifier from the acquired surround view image of the vehicle;

[0009] Projecting the specific identifier to a map coordinate system to obtain a map identifier;

[0010] Determining a matching pose according to the map identifier;

[0011] Update the pathfinding map according to the matching posture;

[0012] Fusing the predicted pose and the matched pose to obtain a fused pose;

[0013] Based on the established path-finding map, the vehicle is positioned through the fused posture.

[0014] Optionally, determining the predicted position and posture of the vehicle according to the acquired inertial measurement data and wheel speed data of the vehicle includes:

[0015] Determining an initial posture of the vehicle according to acceleration data and angular velocity data, wherein the inertial measurement data includes the acceleration data and the angular velocity data of the vehicle;

[0016] The initial posture is predicted and recursively deduced according to the acceleration data, the angular velocity data and the wheel speed data to obtain the predicted posture.

[0017] Optionally, the recursively predicting the initial posture according to the acceleration data, the angular velocity data and the wheel speed data to obtain the predicted posture comprises:

[0018] Determine whether there is a satellite signal;

[0019] When the satellite signal exists, determining the initial position of the vehicle through the satellite signal, and determining the predicted posture according to the initial position, the acceleration data, the angular velocity data and the wheel speed data;

[0020] When the satellite signal does not exist, the initial position is obtained by traversing the path-finding map, and the predicted posture is determined according to the initial position, the acceleration data, the angular velocity data and the wheel speed data.

[0021] Optionally, the specific identifier comprises a ground identifier fixed to the ground.

[0022] Optionally, projecting the specific identifier to a map coordinate system to obtain a map identifier includes:

[0023] Projecting the specific identifier in the pixel coordinate system to the vehicle coordinate system;

[0024] Determining a predicted position of the vehicle at the time of acquiring the surround view image according to the inertial measurement data and the wheel speed data;

[0025] According to the predicted position, the specific identifier located in the vehicle coordinate system is projected to the map coordinate system to obtain the map identifier.

[0026] Optionally, determining a matching posture according to the map identifier includes:

[0027] Interpolating the grid probabilities of adjacent grids of the map identifier to obtain a continuous grid probability distribution;

[0028] constructing a cost function based on the map identifier and the continuous grid probability distribution;

[0029] The matching posture is determined according to the cost function, wherein the matching posture includes a posture corresponding to a minimum value when the cost function converges.

[0030] Optionally, before positioning the vehicle by using the fused posture based on the established path-finding map, the method further includes:

[0031] According to the matching posture, projecting the specific identifier in the vehicle coordinate system to the map coordinate system to obtain an updated identifier;

[0032] Based on the map coordinate system, the map identifier and the updated identifier at the same coordinate point are subjected to binary Bayesian filtering to obtain the updated path-finding map.

[0033] Optionally, fusing the predicted pose and the matched pose to obtain a fused pose includes:

[0034] Determine a posture transformation value of the vehicle between an acquisition time and a fusion time, and convert the matching posture into an equivalent posture at the fusion time according to the posture transformation value, wherein the acquisition time includes a time when the surround view image is acquired, and the equivalent posture is used to be equivalent to the matching posture at the fusion time;

[0035] The predicted posture and the equivalent posture are fused through a Kalman filter to obtain the fused posture.

[0036] In a second aspect, the present invention provides a vehicle path finding and positioning device, comprising:

[0037] A prediction module, used to determine the predicted position and posture of the vehicle based on the acquired inertial measurement data and wheel speed data of the vehicle;

[0038] An extraction module, used to extract a specific identifier from the acquired surround view image of the vehicle;

[0039] A projection module, used for projecting the specific identifier to a map coordinate system to obtain a map identifier;

[0040] A matching module, used for determining a matching pose according to the map identifier;

[0041] A fusion module, used for fusing the predicted pose and the matched pose to obtain a fused pose;

[0042] A pathfinding module is used to locate the vehicle through the fused posture based on an established pathfinding map.

[0043] In a third aspect, the present invention provides an electronic device, including a memory and a processor;

[0044] The memory is used to store computer programs;

[0045] The processor is used to implement the vehicle path finding and positioning method as described in the first aspect when executing the computer program.

[0046] The beneficial effects of the vehicle path finding and positioning of the present invention are:

[0047] In view of the working environment of the vehicle during pathfinding and positioning, the predicted posture of the vehicle is recursively determined through inertial measurement data and wheel speed data, and the predicted posture is used as the prior prediction of the fused posture; on the other hand, specific identifiers are extracted through surround images to avoid problems such as being easily affected by traditional camera images. The specific identifier is projected to the map coordinate system as the projection of the prior image semantics on the map, and then the observed posture is determined, that is, the matching posture that matches the identifier; the matching posture can be fused with the predicted posture to obtain a fused posture corrected based on the predicted value and the observed value; the positioning and pathfinding of the vehicle can be realized based on the pathfinding map and the fused posture, thereby improving the positioning accuracy of the vehicle and ensuring the realization of the vehicle's memory parking function. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 A schematic diagram of a process flow of a vehicle path finding and positioning method according to an embodiment of the present invention;

[0049] Figure 2 This is a schematic diagram of a detailed process of step S100 of the vehicle path finding and positioning method according to an embodiment of the present invention;

[0050] Figure 3 This is a schematic diagram of a detailed process of step S300 of the vehicle path finding and positioning method according to an embodiment of the present invention;

[0051] Figure 4 This is a schematic diagram of a detailed process of step S500 of the vehicle path finding and positioning method according to an embodiment of the present invention;

[0052] Figure 5 A flowchart of a vehicle path finding and positioning method according to an embodiment of the present invention;

[0053] Figure 6 2 is an exemplary diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0054] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. Although certain embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be interpreted as being limited to the embodiments described herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present invention. It should be understood that the drawings and embodiments of the present invention are only for exemplary purposes and are not intended to limit the scope of protection of the present invention.

[0055] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0056] The term "including" and its variations used in this document are open inclusions, that is, "including but not limited to"; the term "based on" means "based at least in part on"; the term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one other embodiment"; the term "some embodiments" means "at least some embodiments"; the term "optionally" means "optional embodiments". The relevant definitions of other terms will be given in the following description. It should be noted that the concepts of "first", "second", etc. mentioned in the present invention are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0057] It should be noted that the modifications of "one" and "plurality" mentioned in the present invention are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0058] The names of the messages or information exchanged between multiple devices in the embodiments of the present invention are only used for illustrative purposes, and are not used to limit the scope of these messages or information.

[0059] This embodiment provides a vehicle path finding and positioning method, device and electronic equipment.

[0060] like Figure 1 As shown, a vehicle path finding and positioning method provided by an embodiment of the present invention includes:

[0061] Step S100, determining the predicted position and posture of the vehicle based on the acquired inertial measurement data and wheel speed data of the vehicle.

[0062] In one embodiment, when the vehicle is in the home-zone parking assist (HPA) state, the automatic driving system can autonomously complete all instructions for the vehicle to automatically park in and out of the parking lot from the parking lot entrance to the parking lot, or from the parking lot to the parking lot exit, in a specific parking space in a specific area without human intervention. The working environment of the vehicle may be in an underground parking lot, and the satellite signal cannot pass through the building. In the underground parking lot, the vehicle cannot be positioned by satellite positioning. When positioning the vehicle, it is necessary to reduce the dependence on satellite signals. Therefore, in an embodiment of the present invention, inertial measurement is performed by an inertial measurement unit (IMU) in the vehicle to obtain inertial measurement data of the vehicle during driving; the wheel speed data of the vehicle during driving is obtained by a sensor set on the chassis of the vehicle; the inertial measurement data and the wheel speed data are fused, and the vehicle posture is recursively deduced to obtain the predicted posture of the vehicle. The predicted posture is used as a priori posture during posture fusion, providing an initial guess about the system state for the Kalman filter as a basis for further optimization and correction.

[0063] Step S200: extracting a specific identifier from the acquired surround view image of the vehicle.

[0064] In one embodiment, since a single camera can only obtain the size ratio of the imaged object, it is impossible to obtain the scale information of the object according to the image, nor is it possible to directly obtain the precise distance or depth information of the object in the scene. In the case of only one image, it is difficult to accurately estimate the motion trajectory or speed of the object. On the other hand, when a traditional single camera extracts feature points, it is easily affected by the environment, such as light intensity, light reflection, etc., resulting in unstable image positioning and low precision. In order to overcome the above problems, in an embodiment of the present invention, a surround image is obtained by a surround camera, and image semantics are extracted by the surround image, that is, specific identifiers in the scene are extracted. The surround camera is fixedly installed on the vehicle, and the installation posture of the surround camera is known. Therefore, the height of the surround camera from the ground is also known. Through the pinhole imaging principle, the distance of the object relative to the camera is known, and the real length and width of the ground object in the surround image can be determined, and then the marker to be identified is identified by the image recognition algorithm, and the marker is identified and segmented, that is, the specific identifier and its location are obtained to ensure the accuracy of image recognition. By using image recognition, the specific identifier of interest is extracted instead of the entire image, which can reduce the amount of calculation and improve processing efficiency.

[0065] Step S300: Project the specific identifier to a map coordinate system to obtain a map identifier.

[0066] In one embodiment, after a specific identifier is extracted through image recognition, since the specific identifier is in an image-based coordinate system, it cannot be used directly to match the vehicle's posture. It is necessary to project the specific identifier to the map coordinate system where the routing map is located, that is, to project the specific identifier from the surround image to the routing map as a map identifier, and determine the posture relationship between the vehicle and the routing map through the map identifier.

[0067] Step S400: determining a matching pose according to the map identifier.

[0068] In one embodiment, the image acquired by the surround-view camera is determined, and the position of the vehicle relative to the path-finding map needs to be matched through the determined image. Therefore, the specific identifier is projected to the map coordinate system as a map identifier, and the current position of the vehicle in the path-finding map is determined based on the map identifier. The position with the highest probability is used as the matching position. The matching position is used as the observation position in the position fusion, used to provide external information of the vehicle's position, and used to correct the state of the predicted position, reduce the estimation error, and improve the real-time adaptability of the vehicle during memory parking.

[0069] Step S500, fusing the predicted pose and the matched pose to obtain a fused pose.

[0070] In one embodiment, the state guess of the vehicle posture is provided by predicting the posture, and the posture estimate is corrected based on the difference between the posture and the actual observed matching posture, so that the vehicle maintains a robust result in a changing environment.

[0071] Step S600: positioning the vehicle through the fused posture based on the established path-finding map.

[0072] Based on the path-finding map and fused posture, the vehicle can be positioned and memory-parked.

[0073] In this embodiment, the predicted posture of the vehicle is determined recursively by inertial measurement data and wheel speed data in view of the working environment of the vehicle during pathfinding and positioning, and the predicted posture is used as the prior prediction of the fused posture; on the other hand, specific identifiers are extracted through surround images to avoid problems such as being easily affected by traditional camera images. The specific identifier is projected to the map coordinate system as the projection of the prior image semantics on the map, and then the observed posture, that is, the matching posture that matches the identifier, is determined; the matching posture can correct and update the pathfinding map on the one hand, and on the other hand, it can be fused with the predicted posture to obtain a fused posture corrected based on the predicted value and the observed value; the positioning and pathfinding of the vehicle can be realized based on the pathfinding map and the fused posture, thereby improving the positioning and mapping accuracy of the vehicle and ensuring the realization of the vehicle's memory parking function.

[0074] Alternatively, if Figure 2As shown, the method of determining the predicted posture of the vehicle based on the acquired inertial measurement data and wheel speed data of the vehicle includes:

[0075] Step S110, determining an initial posture of the vehicle according to the acceleration data and the angular velocity data, wherein the inertial measurement data includes the acceleration data and the angular velocity data of the vehicle.

[0076] Step S120, recursively predicting the initial posture according to the acceleration data, the angular velocity data and the wheel speed data to obtain the predicted posture.

[0077] In one embodiment, the inertial measurement data is measured by a three-axis accelerometer and a three-axis gyroscope installed in the vehicle. Before determining the predicted posture of the vehicle based on the inertial measurement data and the discussion data, the inertial measurement data is initialized, the angle of rotation of the vehicle around its horizontal axis (Pitch angle) and the angle of rotation of the vehicle around its longitudinal axis (Roll angle) are calculated by gravity acceleration, and the gyroscope and accelerometer zero bias are calibrated. When in use, the initial posture of the vehicle, i.e., the initial posture, is determined by the inertial measurement data, and is fused with the wheel speed data obtained from the chassis to iteratively deduce the posture of the vehicle in the future, i.e., the predicted posture.

[0078] Optionally, the recursively predicting the initial posture according to the acceleration data, the angular velocity data and the wheel speed data to obtain the predicted posture comprises:

[0079] Determine whether there is a satellite signal.

[0080] When the satellite signal exists, the initial position of the vehicle is determined by the satellite signal, and the predicted posture is determined according to the initial position, the acceleration data, the angular velocity data and the wheel speed data.

[0081] When the satellite signal does not exist, the initial position is obtained by traversing the path-finding map, and the predicted posture is determined according to the initial position, the acceleration data, the angular velocity data and the wheel speed data.

[0082] In some embodiments, the vehicle is in an underground garage when performing memory parking, and it is difficult to receive satellite signals; in other embodiments, the vehicle is in a garage or open-air parking lot where satellite signals can be received when performing memory parking, and it is necessary to consider the positioning solutions for the vehicle when performing memory parking in different environments.

[0083] Specifically, in the process of obtaining the initial posture, it can be determined whether there is a receivable satellite signal. When there is a satellite signal, the initial position can be determined by using the satellite signal, and then the predicted posture is determined by recursive prediction based on the initial position through the acceleration data, angular velocity data and wheel speed data. When there is no satellite signal, the initial position is found by traversing the path-finding map, and the predicted posture is determined by recursive prediction based on the initial position through the acceleration data, angular velocity data and wheel speed data.

[0084] Optionally, the specific identifier comprises a ground identifier fixed to the ground.

[0085] Specifically, in order to avoid the situation where the image acquired by a single camera is easily affected, the surround view camera is used to acquire surround view images, which can determine the true length and width of ground objects. The features that are fixed to the ground and can be used as identifiers are used as ground identifiers for pose matching, which can obtain more accurate matching poses and are not affected by environmental factors. By constraining the information to the ground, when the surround view camera external parameters are known, more accurate geometric information can be obtained through surround view images, which has higher positioning accuracy.

[0086] Alternatively, if Figure 3 As shown, projecting the specific identifier to a map coordinate system to obtain a map identifier includes:

[0087] Step S310: projecting the specific identifier in the pixel coordinate system to the vehicle coordinate system.

[0088] Step S320: determining the predicted position of the vehicle at the time of acquiring the surround view image according to the inertial measurement data and the wheel speed data.

[0089] Step S330: According to the predicted position, project the specific identifier in the vehicle coordinate system to the map coordinate system to obtain the map identifier.

[0090] In one embodiment, since the specific identifier is still in the pixel coordinate system with the surround image as the reference system, when matching the posture, the coordinate system needs to be unified, and the specific identifier is projected to the ground under the vehicle coordinate system, wherein the vehicle coordinate system is defined as a coordinate system relative to the center of the rear axle of the vehicle, and the coordinate system is used to transform the specific identifier. According to the predicted position, the specific identifier is further projected from the vehicle coordinate system to the map coordinate system as a map identifier, so that the map identifier can match the posture of the vehicle in the map coordinate system.

[0091] Optionally, determining a matching posture according to the map identifier includes:

[0092] The grid probabilities of adjacent grids of the map identifier are interpolated to obtain a continuous grid probability distribution.

[0093] A cost function is constructed based on the map identifier and the continuation grid probability distribution.

[0094] The matching posture is determined according to the cost function, wherein the matching posture includes a posture corresponding to a minimum value when the cost function converges.

[0095] In one embodiment, a suitable method is selected to ensure the smoothness of the interpolation process, so that the probability of the grid around the map identifier projected into the map coordinate system is continuously distributed, and a cost function based on the map identifier and the grid probability distribution is constructed. The accuracy of the matching posture is evaluated by the cost function. When the cost function is larger, the probability of the accurate matching posture is smaller; conversely, when the cost function is smaller, the probability of the accurate matching posture is greater. Based on the cost function, the value of the maximum probability that all map identifiers are consistent with the mapping of the same type of points is calculated. Through iterative calculation, it converges to the observed posture corresponding to the local maximum probability, that is, the matching posture.

[0096] Since vehicle motion is a continuous process, the vehicle's posture at the next moment is predicted based on the posture at the previous moment through inertial measurement data and wheel speed. After obtaining the predicted prior posture, the specific identifier is projected to the vehicle coordinate system, and the specific identifier in the vehicle coordinate system is further projected to the map coordinate system through the predicted vehicle posture. The cost function with a loss function can be established through the projection of the prior image semantics on the map, and the local optimal value can be solved by iteration, thereby ensuring that the accurate matching posture can be calculated even when the local image is occluded or the image is not correct.

[0097] Alternatively, if Figure 4 As shown, before positioning the vehicle through the fusion posture based on the established path-finding map, the method further includes:

[0098] Step S510: projecting the specific identifier in the vehicle coordinate system to the map coordinate system according to the matching posture to obtain an updated identifier.

[0099] Step S520: Based on the map coordinate system, the map identifier and the updated identifier at the same coordinate point are subjected to binary Bayesian filtering to obtain the updated path-finding map.

[0100] In the embodiment of the present invention, the so-called updated pathfinding map includes two processes of establishing a pathfinding map from scratch and updating a pathfinding map from old to new. For example, when the vehicle memorizes a new route, the predicted posture of the vehicle is determined according to the acquired inertial measurement data and wheel speed data of the vehicle; a specific identifier is extracted from the acquired surround view image of the vehicle; the specific identifier is projected to the map coordinate system to obtain a map identifier; a matching posture is determined according to the map identifier; and then the pathfinding map is determined according to the matching posture. When the vehicle updates the memorized route, after obtaining the matching posture through the above steps, the observed posture obtained by the surround view image is determined, and the observed posture is introduced and compared with the semantic points in the previously established pathfinding map, that is, compared with the map identifier in the pathfinding map, and the probability value of the grid where the consistent semantic points are located is increased; the probability value of the grid where the inconsistent semantic points are located is reduced. In the process of building and updating the map, the update of the pathfinding map in the memorized parking path can be completed through steps S100-S400.

[0101] In one embodiment, after determining the matching posture, the specific identifier of the ground in the vehicle coordinate system is re-projected back to the map coordinate system through the matching posture, and the map semantics at the same coordinate point are compared. When the two are consistent, the probability value of the grid in which they are located is updated and increased. When the two are inconsistent, the probability value of the grid in which they are located is reduced to obtain an updated pathfinding map.

[0102] Specifically, the extracted specific identifier is converted to the vehicle coordinate system relative to the center of the rear axle of the vehicle using the projection relationship. Then, the same specific identifier in the same area on the map is subjected to a binary Bayesian update. The grid where the current specific identifier is located will have its probability value increased through Bayesian filtering. Similarly, if the grid with the same specific identifier in this area is not in this coordinate system, its probability value will be reduced through Bayesian filtering. Through the above algorithm, when noise appears in the pathfinding map, it will be eliminated or its probability of existence will be reduced during multiple iterations, so that the final pathfinding map retains the grid coordinates of the same specific identifier that appears multiple times.

[0103] Alternatively, if Figure 5 As shown, the fusing of the predicted pose and the matched pose to obtain the fused pose includes:

[0104] Determine a posture transformation value of the vehicle between an acquisition time and a fusion time, and convert the matching posture into an equivalent posture at the fusion time according to the posture transformation value, wherein the acquisition time includes a time when the surround view image is acquired, and the equivalent posture is used to be equivalent to the matching posture at the fusion time;

[0105] The predicted posture and the equivalent posture are fused through a Kalman filter to obtain the fused posture.

[0106] Since the acquisition and feature extraction of surround view images takes a certain amount of time, the matching pose obtained is earlier than the current fusion pose. It is necessary to first calculate the pose transformation value from the surround view image acquisition time to the current time (fusion time), and then convert the pose at the surround view image acquisition time to the current time, making it equivalent to the observation value at the current time. The predicted position at the surround view image acquisition time is determined by inertial measurement data and wheel speed data, that is, it is equivalent to the pose at the current time. The predicted pose and the matching pose equivalent to the current time are fused by the Kalman filter, that is, the predicted pose and the observed pose are fused to obtain the a posteriori fusion pose, and the positioning result data is used to realize the memory parking of the vehicle based on the path finding map.

[0107] An embodiment of the present invention provides a vehicle path finding and positioning device, comprising:

[0108] A prediction module, used to determine the predicted position and posture of the vehicle based on the acquired inertial measurement data and wheel speed data of the vehicle;

[0109] An extraction module, used to extract a specific identifier from the acquired surround view image of the vehicle;

[0110] A projection module, used for projecting the specific identifier to a map coordinate system to obtain a map identifier;

[0111] A matching module, used for determining a matching posture according to the map identifier;

[0112] A fusion module, used for fusing the predicted pose and the matched pose to obtain a fused pose;

[0113] A pathfinding module is used to locate the vehicle through the fused posture based on an established pathfinding map.

[0114] like Figure 6 As shown, an electronic device 600 provided by an embodiment of the present invention includes a memory 610 and a processor 620; the memory 610 is used to store a computer program; the processor 620 is used to implement the vehicle path finding and positioning method as described above when executing the computer program.

[0115] In other words, an electronic device 600 includes a memory 610 and a processor 620 coupled to the memory 610; the memory 610 is configured to store a computer program; and the processor 620 is configured to perform the following operations when executing the computer program:

[0116] Determine the predicted position and posture of the vehicle based on the acquired inertial measurement data and wheel speed data of the vehicle;

[0117] Extracting a specific identifier from the acquired surround view image of the vehicle;

[0118] Projecting the specific identifier to a map coordinate system to obtain a map identifier;

[0119] Determining a matching pose according to the map identifier;

[0120] Update the pathfinding map according to the matching posture;

[0121] Fusing the predicted pose and the matched pose to obtain a fused pose;

[0122] Based on the established path-finding map, the vehicle is positioned through the fused posture.

[0123] An embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the vehicle path finding and positioning method as described above is implemented.

[0124] In other words, a non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor performs the following operations:

[0125] Determine the predicted position and posture of the vehicle based on the acquired inertial measurement data and wheel speed data of the vehicle;

[0126] Extracting a specific identifier from the acquired surround view image of the vehicle;

[0127] Projecting the specific identifier to a map coordinate system to obtain a map identifier;

[0128] Determining a matching pose according to the map identifier;

[0129] Update the pathfinding map according to the matching posture;

[0130] Fusing the predicted pose and the matched pose to obtain a fused pose;

[0131] Based on the established path-finding map, the vehicle is positioned through the fused posture.

[0132] An electronic device 600 that can be used as a server or client of the present invention will now be described, which is an example of a hardware device that can be applied to various aspects of the present invention. The electronic device 600 is intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device 600 can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0133] The electronic device 600 includes a computing unit, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) or a computer program loaded from a storage unit into a random access memory (RAM). In the RAM, various programs and data required for the operation of the device can also be stored. The computing unit, ROM, and RAM are connected to each other via a bus. An input / output (I / O) interface is also connected to the bus.

[0134] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc. In the present application, the unit described as a separate component may or may not be physically separated, and the component displayed as a unit may or may not be a physical unit, that is, it may be located in one place, or it may be distributed on multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment of the present invention. In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0135] Although the present invention is disclosed as above, the protection scope of the present invention is not limited thereto. Those skilled in the art may make various changes and modifications without departing from the spirit and scope of the present invention, and these changes and modifications will fall within the protection scope of the present invention.

Claims

1. A vehicle path finding and positioning method, characterized in that: include: Determine the predicted position and posture of the vehicle based on the acquired inertial measurement data and wheel speed data of the vehicle; Extracting a specific identifier from the acquired surround view image of the vehicle; Projecting the specific identifier to a map coordinate system to obtain a map identifier; Determining a matching pose according to the map identifier; Fusing the predicted pose and the matched pose to obtain a fused pose; Based on the established path-finding map, the vehicle is positioned through the fused posture.

2. The vehicle path finding and positioning method according to claim 1, characterized in that: Determining the predicted position and posture of the vehicle according to the acquired inertial measurement data and wheel speed data of the vehicle includes: Determining an initial posture of the vehicle according to acceleration data and angular velocity data, wherein the inertial measurement data includes the acceleration data and the angular velocity data of the vehicle; The initial posture is predicted and recursively deduced according to the acceleration data, the angular velocity data and the wheel speed data to obtain the predicted posture.

3. The vehicle path finding and positioning method according to claim 2, characterized in that: The step of recursively predicting the initial posture according to the acceleration data, the angular velocity data, and the wheel speed data to obtain the predicted posture comprises: Determine whether there is a satellite signal; When the satellite signal exists, determining the initial position of the vehicle through the satellite signal, and determining the predicted posture according to the initial position, the acceleration data, the angular velocity data and the wheel speed data; When the satellite signal does not exist, the initial position is obtained by traversing the path-finding map, and the predicted posture is determined according to the initial position, the acceleration data, the angular velocity data and the wheel speed data.

4. The vehicle path finding and positioning method according to claim 1, characterized in that: The specific identifier includes a ground identifier fixed to the ground.

5. The vehicle path finding and positioning method according to any one of claims 1 to 4, characterized in that: The projecting the specific identifier to a map coordinate system to obtain a map identifier comprises: Projecting the specific identifier in the pixel coordinate system to the vehicle coordinate system; Determining a predicted position of the vehicle at the time of acquiring the surround view image according to the inertial measurement data and the wheel speed data; According to the predicted position, the specific identifier located in the vehicle coordinate system is projected to the map coordinate system to obtain the map identifier.

6. The vehicle path finding and positioning method according to any one of claims 1 to 4, characterized in that: Determining the matching posture according to the map identifier comprises: Interpolating the grid probabilities of adjacent grids of the map identifier to obtain a continuous grid probability distribution; constructing a cost function based on the map identifier and the continuous grid probability distribution; The matching posture is determined according to the cost function, wherein the matching posture includes a posture corresponding to a minimum value when the cost function converges.

7. The vehicle path finding and positioning method according to any one of claims 1 to 4, characterized in that: Before positioning the vehicle by using the fused posture based on the established path-finding map, the method further includes: According to the matching posture, projecting the specific identifier in the vehicle coordinate system to the map coordinate system to obtain an updated identifier; Based on the map coordinate system, the map identifier and the updated identifier at the same coordinate point are subjected to binary Bayesian filtering to obtain the updated path-finding map.

8. The vehicle path finding and positioning method according to any one of claims 1 to 4, characterized in that: The fusing the predicted pose and the matched pose to obtain a fused pose comprises: Determine a posture transformation value of the vehicle between an acquisition time and a fusion time, and convert the matching posture into an equivalent posture at the fusion time according to the posture transformation value, wherein the acquisition time includes a time when the surround view image is acquired, and the equivalent posture is used to be equivalent to the matching posture at the fusion time; The predicted posture and the equivalent posture are fused through a Kalman filter to obtain the fused posture.

9. A vehicle path finding and positioning device, characterized in that: include: A prediction module, used to determine the predicted position and posture of the vehicle based on the acquired inertial measurement data and wheel speed data of the vehicle; An extraction module, used to extract a specific identifier from the acquired surround view image of the vehicle; A projection module, used for projecting the specific identifier to a map coordinate system to obtain a map identifier; A matching module, used for determining a matching pose according to the map identifier; A fusion module, used for fusing the predicted pose and the matched pose to obtain a fused pose; A pathfinding module is used to locate the vehicle through the fused posture based on an established pathfinding map.

10. An electronic device, characterized in that: including memory and processor; The memory is used to store computer programs; The processor is used to implement the vehicle path finding and positioning method as described in any one of claims 1 to 8 when executing the computer program.

Images