Automatic vehicle parking system key parameter verification method based on miniature sand table experiment platform
Patent Information
- Application Number
- CN202510212279.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-25
- Publication Date
- 2025-06-13
Smart Images
Figure CN120141860A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of vehicle testing, and in particular to a method for verifying key parameters of a vehicle automatic parking system based on a miniature sandbox experimental platform. Background Art
[0002] In the development and deployment of the Automated Parking System (APS), verifying the performance of its key parameters is a very important step. The existing key parameter test and verification methods of the automatic parking system mainly include simulation verification methods and real vehicle verification methods. Among them, simulation verification refers to the system test and verification by simulating the working process of the automatic parking system and the interaction with the environment in a computer environment, while the environment and sensor models in the simulation platform may not fully reflect the complexity of the real world. Real vehicle testing verifies the performance of the automatic parking system in the actual road environment. The automatic parking system is installed on a real vehicle and actual tests are carried out in various scenarios, but there are problems of high cost, high risk and limitations.
[0003] Therefore, how to solve the problems of high cost, high safety risk and limitations of existing key parameter verification methods for vehicle automatic parking systems is an important issue that needs to be urgently addressed in the field of vehicle testing. Summary of the invention
[0004] The present invention provides a method for verifying key parameters of a vehicle automatic parking system based on a miniature sandbox experimental platform, so as to overcome the defects of the existing method for verifying key parameters of a vehicle automatic parking system, such as high cost, high safety risk and limitations, and realize accurate, safe and efficient verification of key parameters of the vehicle automatic parking system.
[0005] In a first aspect, the present invention provides a method for verifying key parameters of a vehicle automatic parking system based on a miniature sandbox experimental platform, wherein the miniature sandbox experimental platform includes a miniature vehicle and a sandbox platform, and is connected to the vehicle automatic parking system through a wireless communication network; the method includes: determining key parameters to be verified of the vehicle automatic parking system and vehicle control instructions; wherein the key parameters to be verified include any one of communication delay, communication packet loss rate, image resolution and perception accuracy; based on the miniature sandbox experimental platform, according to the key parameters to be verified and the vehicle control instructions, controlling the miniature vehicle to perform simulated driving in the sandbox platform, and collecting actual driving data of the miniature vehicle during the simulated driving process; based on the actual driving data and the pre-stored simulated driving data, obtaining the verification results of the key parameters to be verified; wherein the actual driving data includes the left side distance between the left side of the vehicle body of the miniature vehicle and the left side of the garage edge line, and the simulated driving data includes the simulated left side distance between the left side of the vehicle body of the miniature vehicle and the left side of the garage edge line.
[0006] Further, the determining of the key parameters to be verified and the vehicle control instructions of the vehicle automatic parking system includes: obtaining the current operating state information and environmental state information of the scaled-down vehicle in the scaled-down sand table experiment platform; generating vehicle control instructions according to the current operating state information and environmental state information of the scaled-down vehicle; wherein, the current operating state information includes pose information and speed information; the environmental state information includes traffic signal state information, street lamp state information, and lifting rod state information; and the vehicle control instructions include the desired turning angle and desired speed of the scaled-down vehicle.
[0007] Further, the scaled-down sand table experiment platform includes one or more scaled-down vehicles, and the scaled-down vehicles are distinguished by different color block designs; correspondingly, the step of obtaining the current operating state information of the scaled-down vehicle in the scaled-down sand table experiment platform specifically includes: obtaining the global image of the sand table platform and the scaled-down vehicle; performing color space conversion on the global image to obtain an HSV image; performing binarization according to the H value of the pixel points in the HSV image to obtain a binarized image; performing convex polygon fitting on the binarized image and calculating the minimum bounding rectangle of the convex polygon; in the case where the aspect ratio of the minimum bounding rectangle is within a set interval and the area of the minimum bounding rectangle is greater than a set area, taking the minimum bounding rectangle as the color block area on the top of the scaled-down vehicle to obtain the pose information of the scaled-down vehicle, and the pose information includes the position of the scaled-down vehicle.
[0008] Further, the step of obtaining the current operating state information of the scaled-down vehicle in the scaled-down sand table experiment platform specifically includes: obtaining the position of the scaled-down vehicle in adjacent frame global images; performing differential processing on the positions of the scaled-down vehicle in adjacent frame global images to obtain the change in the center point coordinates of the scaled-down vehicle; and obtaining the speed information of the scaled-down vehicle according to the change in the center point coordinates of the scaled-down vehicle.
[0009] Further, the obtaining of the verification result of the key parameters to be verified based on the actual driving data and the pre-stored simulated driving data includes: in the case where the difference between the left distance and the simulated left distance is less than or equal to a set threshold, determining the verification result of the key parameters to be verified as qualified; in the case where the difference between the left distance and the simulated left distance is greater than the set threshold, determining the verification result of the key parameters to be verified as unqualified.
[0010] Further, it further includes: in the case where the left distance is greater than a first set threshold, sending a first prompt message of successful vehicle storage to the scaled-down vehicle; in the case where the left distance is greater than zero and less than the first set threshold, sending a second prompt message of vehicle storage pressing the line to the scaled-down vehicle; and in the case where the left distance is greater than a second set threshold and less than zero, sending a third prompt message of vehicle not stored in the warehouse to the scaled-down vehicle.
[0011] Further, the sand table platform includes structured roads, various roadside devices, and a workstation. Among them, the structured roads are used to provide a simulated driving scenario for the scaled-down vehicles; the roadside devices are arranged on the structured roads and at least include street lights, traffic lights, parking lot lifting rods, and cameras with global coverage; the workstation is connected to both the roadside devices and the scaled-down vehicles simultaneously, and is used to adjust the start-stop states of various roadside devices, read the global images captured by the cameras, and send the expected turning angle and expected vehicle speed to the scaled-down vehicles according to the read global images for the scaled-down vehicles to complete the response.
[0012] In a second aspect, the present invention also provides a key parameter verification device for a vehicle automatic parking system based on a scaled-down sand table experimental platform. The scaled-down sand table experimental platform includes a scaled-down vehicle and a sand table platform, and is connected to the vehicle automatic parking system through a wireless communication network. The device includes: a key parameter to be verified and vehicle control instruction determination module, which is used to determine the key parameters to be verified of the vehicle automatic parking system and the vehicle control instructions; among them, the key parameters to be verified include any one of communication delay, communication packet loss rate, image resolution, and perception accuracy; a simulated driving module of the scaled-down vehicle in the scaled-down sand table experimental platform, which is used to control the scaled-down vehicle to perform simulated driving in the sand table platform based on the scaled-down sand table experimental platform according to the key parameters to be verified and the vehicle control instructions, and collect the actual driving data of the scaled-down vehicle during the simulated driving process; a key parameter to be verified result acquisition module, which is used to obtain the verification result of the key parameter to be verified based on the actual driving data and the pre-stored simulated driving data; among them, the actual driving data includes the left distance between the left side of the scaled-down vehicle body and the left garage border line, and the simulated driving data includes the simulated left distance between the left side of the scaled-down vehicle body and the left garage border line.
[0013] In a third aspect, the present invention also provides a key parameter verification system for a vehicle automatic parking system based on a scaled-down sand table experimental platform, including: a scaled-down sand table experimental platform, which includes a physical sand table platform in the physical space and a twin sand table platform in the information space; among them, the physical sand table platform includes a scaled-down vehicle and a sand table platform, and the sand table platform includes structured roads, various roadside devices, and a workstation; the twin sand table platform is obtained by reconstructing the scene through three-dimensional modeling based on the physical sand table platform, and is used to reflect the real-time operation state of the physical sand table platform; a computing server, which includes a vehicle automatic parking system control unit and is wirelessly connected to the scaled-down sand table experimental platform, and is used to execute the key parameter verification method for the vehicle automatic parking system based on the scaled-down sand table experimental platform.
[0014] Fourthly, the present invention further provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the key parameter verification method for the vehicle automatic parking system based on the micro-scale sand table experimental platform as described in any one of the above.
[0015] The key parameter verification method for the vehicle automatic parking system based on the micro-scale sand table experimental platform provided by the present invention includes determining the key parameters to be verified for the vehicle automatic parking system and vehicle control instructions. Among them, the key parameters to be verified include any one of communication delay, communication packet loss rate, image resolution, and perception accuracy. Based on the micro-scale sand table experimental platform, according to the key parameters to be verified and vehicle control instructions, control the micro-scale vehicle to perform simulated driving on the sand table platform, and collect the actual driving data of the micro-scale vehicle during the simulated driving process. Based on the actual driving data and the pre-stored simulated driving data, obtain the verification result of the key parameters to be verified. Among them, the actual driving data includes the left distance between the left side of the micro-scale vehicle body and the left library boundary line, and the simulated driving data includes the simulated left distance between the left side of the micro-scale vehicle body and the left library boundary line. This method realizes accurate, safe, and efficient verification of the key parameters of vehicle automatic parking by using the micro-scale sand table experimental platform to verify the key parameters of the vehicle automatic parking system. Description of the Drawings
[0016] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.
[0017] Figure 1 It is a system schematic diagram of the key parameter verification system for the vehicle automatic parking system based on the micro-scale sand table experimental platform provided by the embodiments of the present invention.
[0018] Figure 2 It is a flowchart of the key parameter verification method for the vehicle automatic parking system based on the micro-scale sand table experimental platform provided by the embodiments of the present invention.
[0019] Figure 3 It is a structural schematic diagram of the key parameter verification device for the vehicle automatic parking system based on the micro-scale sand table experimental platform provided by the embodiments of the present invention.
[0020] Figure 4 It is a physical structure schematic diagram of the electronic device provided by the embodiments of the present invention. Detailed Embodiments
[0021] To make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings in the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without creative efforts shall fall within the protection scope of the present invention.
[0022] It is easy to understand that before elaborating on the method for verifying key parameters of a vehicle automatic parking system based on a scaled-down sand table experiment platform provided by the embodiments of the present invention in detail, the verification system used to implement this verification method is first described here.
[0023] Figure 1 The system schematic diagram of the verification system for key parameters of a vehicle automatic parking system based on a scaled-down sand table experiment platform provided by the embodiments of the present invention is shown.
[0024] As Figure 1 shown, the system includes: a scaled-down sand table experiment platform, including a physical sand table platform in the physical space and a twin sand table platform in the information space; wherein, the physical sand table platform includes a scaled-down vehicle and a sand table platform, and the sand table platform includes a structured road, a variety of roadside devices, multiple cameras and a workstation; the twin sand table platform is obtained by reconstructing the scene through 3D modeling based on the physical sand table platform, and is used to reflect the real-time operation state of the physical sand table platform; a computing server, including a control unit of the vehicle automatic parking system, is wirelessly connected to the scaled-down sand table experiment platform, and is used to execute the method for verifying key parameters of the vehicle automatic parking system based on the scaled-down sand table experiment platform.
[0025] Specifically, the scaled-down sand table test platform consists of a physical sand table platform in the physical space and a twin sand table platform in the information space. The physical sand table platform in the physical space is mainly composed of a structured road, roadside devices, a workstation and a scaled-down vehicle. Among them, the roadside devices and the scaled-down vehicle are the main traffic elements to be considered.
[0026] The roadside devices all belong to the type of devices with unidirectional information transmission, mainly including street lights, traffic lights, parking lot lifting rods and cameras with full-area coverage. They are connected to the workstation through wired or serial ports. The workstation can adjust the states of roadside devices such as street lights, traffic lights and parking lot lifting rods through serial communication. In addition, through wired connection, the workstation can read the images taken by the global cameras at a rate not lower than 10fps, and then calculate the relevant state information of the scaled-down vehicle based on the images.
[0027] In a specific embodiment, the sand table includes typical and rich road scenarios such as intersections, roundabouts, parking lots, and bus stops, and can be used to simulate various typical urban traffic scenarios. Among them, the length of the sand table is 9m, and the width is 5m. The roads on the sand table are composed of strictly structured roads. The width of a single lane is 240mm, and there are road structures such as two-way four-lane and two-way two-lane. The road yellow lines and green belts restrict the driving directions, and it is stipulated that the counterclockwise direction outside the yellow line near the edge of the sand table is the driving direction.
[0028] At the same time, 11 traffic lights and 73 street lights are deployed on the side of the roads on the sand table, and 3 motor-controlled lifting rods are deployed at the entrance of the parking lot and on the green belt on the center line parallel to the short axis. The traffic lights, street lights, and lifting rods communicate with the workstation responsible for overall information through serial ports. The workstation can change the real-time start and stop status by issuing serial port instructions, and each device can be independently controlled.
[0029] Four cameras are deployed on the ceiling about 2.1m above the plane of the sand table. They are connected to the workstation through USB cables, achieving full coverage of the observation of the plane of the sand table. Among them, there is a certain overlapping area between each camera, and the width of the overlapping area is greater than or equal to the vehicle length. This deployment method ensures that vehicles can be seamlessly detected even when crossing the camera boundaries during driving.
[0030] The miniature vehicle belongs to a device with two-way information transmission. It is connected to the workstation through a local wireless network. The miniature vehicle can convert its own real-time motor speed into real-time speed and actively report it to the workstation; the workstation monitors the real-time operating status of the miniature vehicle and sends the desired front wheel angle and desired vehicle speed to the miniature vehicle. The miniature vehicle can quickly complete the response without exceeding the allowed threshold range.
[0031] In a specific embodiment, the weight of the miniature vehicle used in this embodiment is 1.4kg, and the length, width, and height are 200mm, 180mm, and 130mm respectively. The wheelbase length is 140mm, and the wheel diameter is 60mm. The miniature vehicle is equipped with a camera with a resolution of 640×480 and a single-line lidar. In addition, the miniature vehicle is also equipped with an IMU (Inertial Measurement Unit).
[0032] The scaled vehicle communicates with the sand table workstation via a local wireless network. By obtaining the motor speed, it converts it into the real-time speed of the scaled vehicle and reports it. Generally, the maximum speed of the scaled vehicle can reach 1 m / s, and the battery can continuously supply power for about 3.5 hours. In terms of in-vehicle computing power, the host computer of the scaled vehicle is a Raspberry Pi 4B installed with Ubuntu 1804, and the main control chip of the lower computer is STM32F103RBT6. The CPU of the host computer is a 64-bit quad-core processor with a main frequency of 1.5 GHz, having 2G of memory and 16G of storage. Based on the in-vehicle sensing devices and computing units, the scaled vehicle can achieve a certain accuracy of in-vehicle environment perception.
[0033] The twin sand table platform in the information space mainly consists of roadside devices in the physical space, the one-to-one twin entities (twin scaled vehicles) of the scaled vehicles, and virtual vehicles. Through 3D modeling and interface design, the twin entities complete scene reconstruction based on real-time communication, enabling them to reflect the real-time operating states of the corresponding physical elements. The virtual vehicles are used to represent the physical scaled vehicles through similar dynamics and kinematics modeling.
[0034] This embodiment can complete the modeling of the information space based on the game engine Unity3D. By obtaining the real-time operating states of various traffic elements in the physical space, it adjusts the corresponding state parameters of the twin entities to achieve real-time state mapping. For virtual vehicles, their information flow interaction is similar to that of the scaled vehicles in the physical space. By reporting their real-time state information, including but not limited to speed, position, and orientation, to the host running Unity (i.e., Figure 1 the Unity host in
[0035] The computing server (i.e., Figure 1 the cloud Java server in
[0036] establishes connections with the workstations in the physical space and the Unity host in the information space, obtains the real-time states of various traffic elements in the physical space and virtual vehicles in the information space, and aligns and encapsulates all the state information in terms of time through time synchronization and a predetermined protocol.
[0037] According to Figure 1It can be seen that the key parameter verification system for the vehicle automatic parking system based on the scaled sand table experimental platform provided in this embodiment can provide the state data of the vehicle to the external controller, and on the premise of ensuring safety, the converted external control instructions can also be applied to the scaled vehicle terminal. For the external controller, it generates the control instructions for the scaled vehicle after receiving the real-time state data of the scaled vehicle without having to pay attention to the specific implementation details inside the system.
[0038] To meet the access requirements of external controllers with different needs, three types of vehicle control modes are opened in the cloud in this embodiment. The parameters and meanings corresponding to each vehicle control mode are as follows: (1) Desired front wheel angle and speed; (2) Waypoint. The waypoint is the desired position of the scaled vehicle. Generally, the waypoint is often a discrete point on the center line of the road, and the set of waypoints constitutes the desired trajectory of the scaled vehicle over a period of time. To enable the scaled vehicle to reach the position indicated by the waypoint while meeting the road geometric constraints, the waypoint usually includes information such as coordinates and reference speed; (3) Map node. The map node referred to in this embodiment is a large-scale diversion or confluence point of the vehicle queue, usually referring to the starting point and ending point of the lane under the structured road. To enable the scaled vehicle to reach the position indicated by the map node at the minimum cost while meeting the road geometric constraints, the map node usually includes information such as number, coordinates, and the number of the downstream node connected to it.
[0039] The above three vehicle control modes opened to the outside provide a certain degree of flexibility while ensuring the driving safety of the vehicle, and can meet the vehicle control requirements of different external controllers.
[0040] According to Figure 1 It can also be seen that in addition to the external controller, the external input can also be applied to the human-computer interaction device. To realize the input of the user's intention, this embodiment can capture the behavior of the wearer through the Hololens device and convert the behavior of the wearer into the corresponding intention input based on the preset behavior / logic correspondence. Similarly, to realize the visualization of the information space, the established model can be projected into the three-dimensional space in the form of a hologram based on the Hololens device, providing a three-dimensional immersive observation perspective for the wearer.
[0041] To realize the HDV simulation and provide the driver with a first-person driving perspective, this embodiment is based on a driving simulator, updates the parameters of each model by obtaining the real-time operating state of the scene, and provides the driver with a first-person driving perspective; at the same time, by collecting the driver's input control information, including throttle opening information, gear information, steering wheel angle information, etc., and feeding it back to the scaled vehicle in the system, thereby simulating human driving of a car.
[0042] In this embodiment, a key parameter verification system for a vehicle automatic parking system based on a scaled-down sand table experiment platform includes a scaled-down sand table experiment platform and a computing server. The scaled-down sand table experiment platform includes a physical sand table platform in the physical space and a twin sand table platform in the information space. Among them, the physical sand table platform includes a scaled-down vehicle and a sand table platform, and the sand table platform includes a structured road, a variety of roadside devices, and a workstation. The twin sand table platform is obtained by reconstructing the scene through three-dimensional modeling based on the physical sand table platform, and is used to reflect the real-time operating state of the physical sand table platform. The computing server includes a control unit for the vehicle automatic parking system, and is wirelessly connected to the scaled-down sand table experiment platform, and is used to execute the key parameter verification method for the vehicle automatic parking system based on the scaled-down sand table experiment platform. This system effectively improves the safety, efficiency, and controllability of the key parameter verification of the vehicle automatic parking system.
[0043] Furthermore, based on the key parameter verification system for a vehicle automatic parking system based on a scaled-down sand table experiment platform provided in the above embodiment, a key parameter verification method for a vehicle automatic parking system based on a scaled-down sand table experiment platform is executed. Specifically, Figure 2 Fig. 5 shows a schematic flowchart of the key parameter verification method for a vehicle automatic parking system based on a scaled-down sand table experiment platform provided in an embodiment of the present invention.
[0044] As Figure 2 shown, this method includes steps S210-S230, and the following will elaborate on steps S210-S230 and related steps in detail.
[0045] S210, determine the key parameters to be verified of the vehicle automatic parking system and vehicle control commands; among them, the key parameters to be verified include any one of communication delay, communication packet loss rate, image resolution, and perception accuracy.
[0046] It should be noted that the key parameter verification method for a vehicle automatic parking system based on a scaled-down sand table experiment platform provided in an embodiment of the present invention takes the computing server in the system as the execution subject.
[0047] It is easy to understand that based on the key parameter verification system for a vehicle automatic parking system based on a scaled-down sand table experiment platform provided in the above embodiment, by obtaining the current operating state information and environmental state information of the scaled-down vehicle in the scaled-down sand table experiment platform, the computing server can generate corresponding vehicle control commands. At the same time, the computing server will also determine the key parameters of the vehicle automatic parking system to be verified currently, that is, the key parameters to be verified.
[0048] Specifically, through the cameras with global coverage in the scaled sandbox experiment platform, the state information of the entire sandbox platform and the scaled vehicles can be captured, including but not limited to road network structure information, lane geometry information, the correspondence information between lane IDs and colors, and the overall global image. Based on these perceived state information, the pose information and speed information of the scaled vehicles, that is, the current running state information of the scaled vehicles, can be calculated through vehicle detection and state output algorithms.
[0049] Meanwhile, the start-stop state information of traffic lights, street lights, and parking lot lifting rods can be obtained from the workstation. For this information, an edge cloud (cloud computing model) approach can be used for integration to obtain the integrated environment information, that is, the environmental state information.
[0050] Subsequently, based on the current running state information of the scaled vehicles and the environmental state information, the desired front wheel angle and speed of the scaled vehicles can be determined, thereby generating corresponding vehicle control instructions.
[0051] In addition, the key parameters to be verified in the vehicle automatic parking system in this embodiment include but are not limited to communication delay, communication packet loss rate, image resolution, and perception accuracy. However, when verifying the key parameters of the vehicle automatic parking system, only one of communication delay, communication packet loss rate, image resolution, and perception accuracy is selected to be changed each time. That is to say, the key parameters to be verified such as communication delay, communication packet loss rate, image resolution, and perception accuracy are in an "or" relationship during verification, rather than an "and" relationship.
[0052] Among them, the type of the key parameters to be verified can be determined according to actual requirements / situations, and no specific limitation is made here. The values of the key parameters to be verified are obtained according to a predetermined simulation test algorithm.
[0053] Communication delay refers to the time required for the information to be completely received at the receiving end from the sending end during the network data transmission process, which can be controlled by using a network emulator or built-in tools of the operating system. Communication packet loss rate refers to the proportion of data packets that fail to reach the destination successfully in the total sent data packets during the network data transmission process, which can be controlled by using a network emulator. Image resolution refers to the number of pixels contained in an image. A higher resolution means that the image contains more details and information, which can be controlled by changing the performance parameters of the camera. Perception accuracy refers to the accuracy of the perception model used to detect and predict the state of the scaled vehicles, which can be controlled by changing the model parameters of the perception model.
[0054] Specifically, the control of communication latency / communication packet loss rate mainly targets the scenario of unreliable communication. By relying on the camera to detect the distance between the vehicle body and the parking space line, there is a time delay in the position. This time delay causes the parking position received by the automatic parking system to lag behind the actual parking position. As a result, the actual parking position of the vehicle lags behind the theoretical parking position, leading to a continuous decrease in the left-side distance, and thus the vehicle presses the line.
[0055] In addition, in a communication network, data transmission usually has burstiness. That is, after one data is lost during transmission, the probability of the next data being lost is greater than the probability of successful transmission. Therefore, packet loss may occur. The control of image resolution mainly considers detecting the distance between the vehicle body and the parking space line through the camera. So, the camera resolution will affect the parking performance of the system. The control of the perception accuracy of the perception model mainly considers that the accuracy of detecting the relative distance between the vehicle body and the parking space line based on the camera in the system will directly affect the output accuracy of the system.
[0056] After determining the key parameters to be verified and the vehicle control instructions for the vehicle automatic parking system, the calculation controller will transmit the key parameters to be verified and the vehicle control instructions to the scaled-down sand table experiment platform to execute step S220.
[0057] S220. Based on the scaled-down sand table experiment platform, according to the key parameters to be verified and the vehicle control instructions, control the scaled-down vehicle to perform simulated driving on the sand table platform and collect the actual driving data of the scaled-down vehicle during the simulated driving process.
[0058] It is easy to understand that after receiving the key parameters to be verified, the values of the key parameters to be verified are controlled / adjusted, such as increasing the communication latency, increasing the communication packet loss rate, reducing the image resolution, or reducing the perception accuracy of the perception model, etc.
[0059] After adjusting the key parameters to be verified, according to the vehicle control instructions, adjust the front-wheel angle of the scaled-down vehicle with the desired angle as the target and adjust the speed of the scaled-down vehicle with the desired speed as the target, thereby controlling the scaled-down vehicle to perform simulated driving on the sand table platform. At the same time, based on the cameras with global coverage and various installed sensor devices, collect the actual driving data of the scaled-down vehicle during the entire simulated driving process. Here, the actual driving data includes the left-side distance between the left side of the scaled-down vehicle body and the left-side parking space line.
[0060] After collecting the actual driving data of the scaled-down vehicle during the simulated driving process, transmit the actual driving data to the calculation server to execute step S230.
[0061] S230. Obtain the verification result of the key parameter to be verified based on the actual driving data and the pre-stored simulated driving data, where the actual driving data includes the left distance between the left side of the vehicle body of the scaled vehicle and the left library border line, and the simulated driving data includes the simulated left distance between the left side of the vehicle body of the scaled vehicle and the left library border line.
[0062] It is easy to understand that the system pre-stores the simulated driving data corresponding to the key parameter to be verified, and the simulated driving data includes the simulated left distance between the left side of the vehicle body of the scaled vehicle and the left library border line. After receiving the actual driving data (including the left distance) of the key parameter to be verified transmitted by the scaled sand table experiment platform, the actual driving data is compared with the corresponding simulated driving data, and thus the verification result of the key parameter to be verified can be obtained.
[0063] Specifically, calculate the difference between the left distance obtained in the verification process and the simulated left distance obtained in the simulated test process. If the difference between the two is less than or equal to the set threshold, it means that the verification result of the current key parameter to be verified is qualified; otherwise, if the difference between the two is greater than the set threshold, it means that the verification result of the current key parameter to be verified is unqualified and needs to be adjusted again for the simulated test process.
[0064] Among them, the set threshold can be adjusted according to the actual situation and is not specifically limited here.
[0065] It is worth mentioning that this embodiment verifies the key parameters of the vehicle automatic parking system, but this embodiment also verifies the simulation test algorithm for determining the key parameters to be verified. If the simulation test method is accurate and effective, the key parameters to be verified determined by it must be verified to be qualified; otherwise, if the simulation test method has poor effects, the key parameters to be verified determined by it will cause the vehicle automatic parking system to be unable to complete automatic parking or the automatic parking effect is poor.
[0066] In this embodiment, the key parameters to be verified and the vehicle control instructions of the vehicle automatic parking system are determined; among them, the key parameters to be verified include any one of communication delay, communication packet loss rate, image resolution, and perception accuracy; based on the micro sand table experiment platform, according to the key parameters to be verified and the vehicle control instructions, the micro vehicle is controlled to perform simulated driving on the sand table platform, and the actual driving data of the micro vehicle during the simulated driving process is collected; based on the actual driving data and the pre-stored simulated driving data, the verification result of the key parameter to be verified is obtained; among them, the actual driving data includes the left distance between the left side of the micro vehicle body and the left side parking line, and the simulated driving data includes the simulated left distance between the left side of the micro vehicle body and the left side parking line. This method realizes the accurate, safe, and efficient verification of the key parameters of the vehicle automatic parking by using the micro sand table experiment platform to verify the key parameters of the vehicle automatic parking system.
[0067] On the basis of the above embodiment, further, the process of obtaining the current running state information of the micro vehicle will be described in detail below.
[0068] It is easy to understand that the current running state information of the micro vehicle includes pose information and speed information. Therefore, this embodiment mainly describes in detail the process of obtaining the pose information and speed information.
[0069] To realize the state perception of the micro vehicle, real-time perception of the micro vehicle is realized based on the global camera and the color block design scheme on the top of the micro vehicle. Its essence is an image processing algorithm, that is, the detection and state perception of the vehicle are realized by detecting different color block combinations of different micro vehicles.
[0070] In a specific embodiment, the process of obtaining the pose information of the micro vehicle is described in detail.
[0071] The steps of obtaining the current running state information of the micro vehicle in the micro sand table experiment platform specifically include: obtaining the global image of the sand table platform and the micro vehicle; performing color space conversion on the global image to obtain an HSV image; performing binarization according to the H value of the pixel points in the HSV image to obtain a binarized image; performing convex polygon fitting on the binarized image and calculating the minimum bounding rectangle of the convex polygon; when the aspect ratio of the minimum bounding rectangle is within the set interval and the area of the minimum bounding rectangle is greater than the set area, the minimum bounding rectangle is used as the color block area on the top of the micro vehicle to obtain the pose information of the micro vehicle, and the pose information includes the position of the micro vehicle.
[0072] Specifically, the perception of the pose information of the micro vehicle can be realized based on a single-frame global image, and the following steps are required: color space conversion, edge extraction and graphic filtering, coordinate and orientation calculation and output.
[0073] A color space is one of the ways to describe colors. By abstractly representing colors in a high-dimensional space, each color corresponds to a point in the color space, making the description of colors more intuitive. Since the global images captured by the global camera are based on the RGB color space, they need to be converted to the HSV color space during subsequent processing. For the color values of any pixel point in the global image, convert them from RGB to the HSV color space to obtain the HSV image.
[0074] To extract the information of the scaled-down vehicles, that is, the color patch combination information at the top of each vehicle, in the image represented by the HSV color space (HSV image), search for the colors at the rear, front, and middle of the scaled-down vehicle body in the HSV image in sequence. During the search process, the main parameter is the H value of the pixel point in the HSV image. Based on the H value of the pixel point, convert the HSV image into the required binary image. Due to the possible presence of the colors to be detected and noise in the scene background, the initial binary image often contains multiple regions to be screened. To screen out the target vehicles from multiple regions to be screened, it is necessary to filter the regions to be screened.
[0075] When filtering, first perform convex polygon fitting on the binary image at the pixel level, then calculate the minimum bounding rectangle of the convex polygon, and finally screen based on the geometric features of the minimum bounding rectangle. Considering that the imaging diagram of the designed rectangular color patch may be deformed as the distance between the scaled-down vehicle and the global camera changes during the driving process of the scaled-down vehicle, the geometric parameters of the minimum bounding rectangle may fluctuate. For the result of whether the region to be screened is the target color patch, when the aspect ratio of the minimum bounding rectangle is within a certain range and the area is greater than the threshold, it is determined that the region to be selected is the color patch region at the top of the vehicle, and thus the pose information of the scaled-down vehicle can be determined, including the position and orientation of the scaled-down vehicle.
[0076] In another specific embodiment, the process of obtaining the speed information of the scaled-down vehicle is described in detail.
[0077] It is easy to understand that there are two schemes for obtaining the real-time running speed of the scaled-down vehicle. One is to perform differencing on the positions of the scaled-down vehicle in adjacent frames of global images to obtain the change in the central point coordinates of the scaled-down vehicle; the other is based on the communication between the scaled-down vehicle and the sand table workstation, and the scaled-down vehicle converts its own motor speed into the real-time speed and reports it.
[0078] In the solution where the miniature vehicle actively reports its speed, the host computer of the miniature vehicle is a Raspberry Pi 4B with the Ubuntu 18.04 system, and the main control chip of the lower computer is an STM32. The Raspberry Pi and the sand table host are in the same local area network and communicate wirelessly. The sand table host sends the desired speed and front wheel angle to the Raspberry Pi at a frequency of 20Hz, and the Raspberry Pi sends the desired speed and front wheel angle commands to the lower computer based on serial communication at a frequency of 120Hz. The lower computer sends corresponding control commands to the execution device of the miniature vehicle at a frequency of 20Hz. The vehicle obtains the states of mechanical devices such as motors, such as rotational speed, and then converts them into the corresponding vehicle speed. Finally, the estimated vehicle speed is reported to the sand table host at a frequency of 20Hz based on wireless communication.
[0079] In this embodiment, by determining the key parameters to be verified and vehicle control commands of the vehicle automatic parking system; wherein, the key parameters to be verified include any one of communication delay, communication packet loss rate, image resolution, and perception accuracy; based on the miniature sand table experimental platform, according to the key parameters to be verified and vehicle control commands, control the miniature vehicle to perform simulated driving on the sand table platform, and collect the actual driving data of the miniature vehicle during the simulated driving process; based on the actual driving data and the pre-stored simulation driving data, obtain the verification results of the key parameters to be verified; wherein, the actual driving data includes the left distance between the left side of the body of the miniature vehicle and the left side parking line, and the simulation driving data includes the simulated left distance between the left side of the body of the miniature vehicle and the left side parking line. This method realizes the accurate, safe, and efficient verification of the key parameters of the vehicle automatic parking system by using the miniature sand table experimental platform to verify the key parameters of the vehicle automatic parking system.
[0080] In some other embodiments, if the left distance is greater than the first set threshold, the vehicle automatic parking system control unit in the calculation server will send the first prompt message of successful vehicle entry to the miniature vehicle; if the left distance is greater than zero and less than the first set threshold, the vehicle automatic parking system control unit in the calculation server will send the second prompt message of vehicle entry and pressing the line to the miniature vehicle; if the left distance is greater than the second set threshold and less than zero, the vehicle automatic parking system control unit in the calculation server will send the third prompt message of vehicle not entering the warehouse to the miniature vehicle.
[0081] Among them, the first set threshold and the second set threshold can be adjusted according to the actual situation, and no specific limitation is made here.
[0082] Corresponding to the method for verifying the key parameters of the vehicle automatic parking system based on the miniature sand table experimental platform described in the above embodiments, the present invention also provides a device for verifying the key parameters of the vehicle automatic parking system based on the miniature sand table experimental platform.
[0083] Specifically, Figure 3The structural schematic diagram of the key parameter verification device for the vehicle automatic parking system based on the micro - scale sand table experiment platform provided by the embodiments of the present invention is shown.
[0084] As Figure 3 shown, the device includes: a key parameter to be verified and vehicle control instruction determination module 310, configured to determine the key parameter to be verified and the vehicle control instruction of the vehicle automatic parking system; wherein, the key parameter to be verified includes any one of communication delay, communication packet loss rate, image resolution, and perception accuracy; a micro - scale vehicle simulation driving module 320 in the micro - scale sand table experiment platform, configured to, based on the micro - scale sand table experiment platform, control a micro - scale vehicle to perform simulated driving on the sand table platform according to the key parameter to be verified and the vehicle control instruction, and collect the actual driving data of the micro - scale vehicle during the simulated driving process; a key parameter to be verified verification result acquisition module 330, configured to obtain the verification result of the key parameter to be verified based on the actual driving data and the pre - stored simulation driving data; wherein, the actual driving data includes the left - hand side distance between the body of the micro - scale vehicle and the left - hand side parking line, and the simulation driving data includes the simulated left - hand side distance between the body of the micro - scale vehicle and the left - hand side parking line.
[0085] In this embodiment, the key parameter to be verified and vehicle control instruction determination module 310 determines the key parameter to be verified and the vehicle control instruction of the vehicle automatic parking system; wherein, the key parameter to be verified includes any one of communication delay, communication packet loss rate, image resolution, and perception accuracy; the micro - scale vehicle simulation driving module 320 in the micro - scale sand table experiment platform controls a micro - scale vehicle to perform simulated driving on the sand table platform according to the key parameter to be verified and the vehicle control instruction based on the micro - scale sand table experiment platform, and collects the actual driving data of the micro - scale vehicle during the simulated driving process; the key parameter to be verified verification result acquisition module 330 obtains the verification result of the key parameter to be verified based on the actual driving data and the pre - stored simulation driving data; wherein, the actual driving data includes the left - hand side distance between the body of the micro - scale vehicle and the left - hand side parking line, and the simulation driving data includes the simulated left - hand side distance between the body of the micro - scale vehicle and the left - hand side parking line. This device realizes accurate, safe, and efficient verification of the key parameters of vehicle automatic parking by using the micro - scale sand table experiment platform to verify the key parameters of the vehicle automatic parking system.
[0086] It should be noted that the key parameter verification device for the vehicle automatic parking system based on the micro - scale sand table experiment platform provided by the embodiments of the present invention can be correspondingly referred to the key parameter verification method for the vehicle automatic parking system based on the micro - scale sand table experiment platform described in the above embodiments, and will not be elaborated here.
[0087] Figure 4 Illustrates the physical structure schematic diagram of an electronic device, as Figure 4As shown in the figure, the electronic device may include: a processor 410, a communications interface 420, a memory 430, and a communication bus 440. Among them, the processor 410, the communications interface 420, and the memory 430 complete mutual communication through the communication bus 440. The processor 410 may call logical instructions in the memory 430 to execute a method for verifying key parameters of a vehicle automatic parking system based on a micro-scale sand table experiment platform. The method includes: determining the key parameters to be verified and vehicle control instructions of the vehicle automatic parking system; where the key parameters to be verified include any one of communication delay, communication packet loss rate, image resolution, and perception accuracy; based on the micro-scale sand table experiment platform, according to the key parameters to be verified and the vehicle control instructions, controlling a micro-scale vehicle to perform simulated driving on the sand table platform, and collecting actual driving data of the micro-scale vehicle during the simulated driving process; based on the actual driving data and pre-stored simulation driving data, obtaining a verification result of the key parameters to be verified; where the actual driving data includes the left distance between the left side of the body of the micro-scale vehicle and the left library border line, and the simulation driving data includes the simulated left distance between the left side of the body of the micro-scale vehicle and the left library border line.
[0088] In addition, when the logical instructions in the above-mentioned memory 430 can be implemented in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to enable a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.
[0089] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. A person of ordinary skill in the art can understand and implement it without creative labor.
[0090] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.
[0091] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A method for verifying key parameters of a vehicle automatic parking system based on a miniature sandbox experimental platform, characterized in that: The miniature sand table experimental platform includes a miniature vehicle and a sand table platform, which is connected to the vehicle automatic parking system via a wireless communication network; The method comprises: Determining key parameters to be verified of the vehicle automatic parking system and vehicle control instructions; wherein the key parameters to be verified include any one of communication delay, communication packet loss rate, image resolution and perception accuracy; Based on the miniature sandbox experimental platform, according to the key parameters to be verified and the vehicle control instructions, the miniature vehicle is controlled to perform simulated driving in the sandbox platform, and actual driving data of the miniature vehicle during the simulated driving process is collected; Based on actual driving data and pre-stored simulated driving data, a verification result of the key parameter to be verified is obtained; wherein the actual driving data includes the left side distance between the left side of the miniature vehicle body and the left side of the garage edge line, and the simulated driving data includes the simulated left side distance between the left side of the miniature vehicle body and the left side of the garage edge line.
2. The key parameter verification method of the vehicle automatic parking system based on the miniature sandbox experimental platform according to claim 1 is characterized in that: The step of determining key parameters to be verified of the vehicle automatic parking system and vehicle control instructions includes: Obtain the current operating status information and environmental status information of the miniature vehicle in the miniature sandbox experimental platform; Generate vehicle control instructions according to the current operating state information and environmental state information of the miniature vehicle; Among them, the current running state information includes posture information and speed information; the environmental state information includes signal light state information, street light state information and lifting pole state information; the vehicle control instruction includes the expected turning angle and expected speed of the miniature vehicle.
3. The key parameter verification method of the vehicle automatic parking system based on the miniature sandbox experimental platform according to claim 2 is characterized in that: The miniature sandbox experimental platform includes one or more miniature vehicles, and the miniature vehicles are distinguished by different color block designs; Accordingly, the step of obtaining the current running status information of the miniature vehicle in the miniature sandbox experimental platform specifically includes: Get the global image of the sandbox platform and miniature vehicles; Performing color space conversion on the global image to obtain an HSV image; Binarize the image according to the H value of the pixel in the HSV image to obtain a binary image; Performing convex polygon fitting on the binary image and calculating the minimum envelope rectangle of the convex polygon; When the aspect ratio of the minimum envelope rectangle is within a set range and the area of the minimum envelope rectangle is greater than the set area, the minimum envelope rectangle is used as the color block area on the top of the miniature vehicle to obtain the posture information of the miniature vehicle, wherein the posture information includes the position of the miniature vehicle.
4. The key parameter verification method of the vehicle automatic parking system based on the miniature sandbox experimental platform according to claim 2 is characterized in that: The steps of obtaining the current running status information of the miniature vehicle in the miniature sandbox experimental platform specifically include: Obtaining the position of the miniature vehicle in the global image of the adjacent frames; Perform differential processing on the position of the miniature vehicle in the global image of adjacent frames to obtain the change of the coordinates of the center point of the miniature vehicle; The speed information of the miniature vehicle is obtained according to the change of the coordinates of the center point of the miniature vehicle.
5. The key parameter verification method of the vehicle automatic parking system based on the miniature sandbox experimental platform according to claim 1 is characterized in that: The obtaining the verification result of the key parameter to be verified based on the actual driving data and the pre-stored simulated driving data includes: When the difference between the left side distance and the simulated left side distance is less than or equal to a set threshold, the verification result of the key parameter to be verified is determined as qualified; When the difference between the left side distance and the simulated left side distance is greater than a set threshold, the verification result of the key parameter to be verified is determined as verification failure.
6. The key parameter verification method of the vehicle automatic parking system based on the miniature sandbox experimental platform according to claim 5 is characterized in that: Also includes: When the left side distance is greater than a first set threshold, a first prompt message of successful vehicle entry is sent to the miniature vehicle; When the left side distance is greater than zero and less than a first set threshold, a second prompt message of a vehicle entering the warehouse and pressing the line is sent to the miniature vehicle; When the left distance is greater than the second set threshold value and less than zero, a third prompt message indicating that the vehicle has not entered the warehouse is sent to the miniature vehicle.
7. The method for verifying key parameters of an automatic parking system of a vehicle based on a miniature sandbox experimental platform according to any one of claims 1 to 6, characterized in that: The sandbox platform includes a structured road, a variety of roadside equipment and a workstation, wherein: The structured road is used to provide a simulated driving scenario for the miniature vehicle; The roadside equipment is arranged on the structured road and includes at least street lights, traffic lights, parking lot lifting poles and cameras with global coverage; The workstation is connected to the roadside equipment and the miniature vehicle at the same time, and is used to adjust the start and stop status of various road test equipment, read the global image taken by the camera, and send the expected turning angle and expected vehicle speed to the miniature vehicle according to the read global image, so that the miniature vehicle can complete the response.
8. A key parameter verification device for a vehicle automatic parking system based on a miniature sandbox experimental platform, characterized in that: The miniature sand table experimental platform includes a miniature vehicle and a sand table platform, which is connected to the vehicle automatic parking system via a wireless communication network; The device comprises: A module for determining key parameters to be verified and vehicle control instructions, used to determine key parameters to be verified and vehicle control instructions of the vehicle automatic parking system; wherein the key parameters to be verified include any one of communication delay, communication packet loss rate, image resolution and perception accuracy; The miniature vehicle simulation driving module in the miniature sandbox experimental platform is used to control the miniature vehicle to perform simulated driving in the sandbox platform based on the miniature sandbox experimental platform according to the key parameters to be verified and the vehicle control instructions, and collect the actual driving data of the miniature vehicle during the simulated driving process; The module for obtaining the verification result of the key parameter to be verified is used to obtain the verification result of the key parameter to be verified based on the actual driving data and the pre-stored simulated driving data; wherein the actual driving data includes the left side distance between the left side of the body of the miniature vehicle and the left side of the garage edge line, and the simulated driving data includes the simulated left side distance between the left side of the body of the miniature vehicle and the left side of the garage edge line.
9. A key parameter verification system for a vehicle automatic parking system based on a miniature sandbox experimental platform, comprising: A miniature sandbox experimental platform, including a physical sandbox platform in physical space and a twin sandbox platform in information space; wherein the physical sandbox platform includes a miniature vehicle and a sandbox platform, and the sandbox platform includes a structured road, a variety of roadside equipment, and a workstation; the twin sandbox platform is obtained by reconstructing the scene through three-dimensional modeling based on the physical sandbox platform, and is used to reflect the real-time operation status of the physical sandbox platform; The computing server includes a vehicle automatic parking system control unit, which is wirelessly connected to the miniature sandbox experimental platform and is used to execute the vehicle automatic parking system key parameter verification method based on the miniature sandbox experimental platform described in any one of claims 1-7.
10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the key parameter verification method of the vehicle automatic parking system based on the miniature sandbox experimental platform as described in any one of claims 1 to 7 is implemented.