Parking Real-Time Obstacle Avoidance Method and Device

Abstract

The present invention relates to the technical field of automatic parking obstacle avoidance, and discloses a real-time parking obstacle avoidance method and device. The present invention discloses determining an obstacle avoidance grid map based on radar data information and an initial grid map; the radar data information is determined based on an ultrasonic input component of the vehicle; the initial grid map is determined based on a visual input component of the vehicle; using a preview algorithm to obtain the future movement trajectory of the vehicle in the obstacle avoidance grid map; determining whether there are obstacles at the coordinate points on the future movement trajectory in the obstacle avoidance grid map, and whether there is a collision risk between the obstacles and the vehicle, and determining an obstacle avoidance strategy for the vehicle according to the determination result. The present invention uses a preview algorithm with low computational overhead to predict the future movement trajectory of the vehicle in real time, which can achieve rapid response, thereby improving the real-time performance of parking obstacle avoidance.

Description

Technical Field

[0001] The present invention relates to the technical field of automatic parking obstacle avoidance, and particularly to a real-time parking obstacle avoidance method and device. Background Art

[0002] With the development of autonomous driving technology and the hardware upgrade of sensors, multi-sensor fusion technology has gradually become a research hotspot. By fusing the data of multiple sensors such as ultrasonic sensors and cameras, more comprehensive and accurate environmental perception capabilities can be provided for automatic parking. However, the existing multi-sensor data fusion algorithms used in the current state-of-the-art technologies have the problem of high computational complexity, which may lead to low real-time performance of obstacle avoidance prediction and better solutions are needed. Summary of the Invention

[0003] In view of this, the present invention provides a real-time parking obstacle avoidance method and device to solve the problem that the automatic parking obstacle avoidance method is not efficient enough.

[0004] On the one hand, the present invention provides a real-time parking obstacle avoidance method, the method includes determining an obstacle avoidance grid map based on radar data information and an initial grid map; the radar data information is determined based on the ultrasonic input component of the vehicle; the initial grid map is determined based on the visual input component of the vehicle; using a preview algorithm to obtain the future movement trajectory of the vehicle in the obstacle avoidance grid map; determining whether there are obstacles at the coordinate points on the future movement trajectory in the obstacle avoidance grid map, and whether there is a collision risk between the obstacles and the vehicle, and determining the obstacle avoidance strategy of the vehicle according to the judgment result.

[0005] On the other hand, the present invention also provides a real-time parking obstacle avoidance device, the device includes: a determination module, configured to determine an obstacle avoidance grid map based on radar data information and an initial grid map; the radar data information is determined based on the ultrasonic input component of the vehicle; the initial grid map is determined based on the visual input component of the vehicle; a prediction module, configured to use a preview algorithm to obtain the future movement trajectory of the vehicle in the obstacle avoidance grid map; an obstacle avoidance module, configured to determine whether there are obstacles at the coordinate points on the future movement trajectory in the obstacle avoidance grid map, and whether there is a collision risk between the obstacles and the vehicle, and determine the obstacle avoidance strategy of the vehicle according to the judgment result.

[0006] On the other hand, the present invention also provides a computer device, the computer device includes: a memory and a processor, the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor realizes the above real-time parking obstacle avoidance method by executing the computer instructions.

[0007] On the other hand, the present invention also provides a computer-readable storage medium, on which computer instructions are stored, and the computer instructions are used to enable a computer to implement the above-mentioned real-time parking obstacle avoidance method.

[0008] In this process, on the one hand, the preview algorithm is simple and efficient in calculation. Compared with complex multi-sensor data fusion algorithms, the preview algorithm has lower calculation overhead and can achieve fast response, thus improving the real-time performance of parking obstacle avoidance. On the other hand, by using the preview algorithm to predict the future motion trajectory of the vehicle in real time, potential obstacles and collision risks can be assisted in being discovered, and the safety of parking obstacle avoidance can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the related art, the following will briefly introduce the drawings required for use in the description of the specific embodiments or the related 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.

[0010] Figure 1 is a schematic flowchart of a real-time parking obstacle avoidance method provided by an embodiment of the present invention;

[0011] Figure 2 is a schematic flowchart of another real-time parking obstacle avoidance method provided by an embodiment of the present invention;

[0012] Figure 3 is a schematic structural diagram of a parking real-time obstacle avoidance device provided by an embodiment of the present invention;

[0013] Figure 4 is a schematic structural diagram of another parking real-time obstacle avoidance device provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0014] In the early development stage of vehicle autonomous driving, the parking obstacle avoidance system usually relies on a single type of sensor, such as a camera or an ultrasonic radar. However, this single sensor has obvious limitations. The detection range of the ultrasonic sensor is limited and its sensitivity to obstacles of certain materials (such as sound-absorbing materials) is relatively low. The detection effect of the camera will be affected under insufficient light or complex lighting conditions, and the image data processing of the camera is complex and has a large amount of calculation, resulting in poor real-time performance.

[0015] To overcome the limitations of a single sensor, multi-sensor fusion technology has gradually become a research hotspot. By fusing data from multiple sensors such as ultrasonic sensors and cameras, the deficiencies of a single sensor can be made up, providing more comprehensive and accurate environmental perception capabilities. Multi-sensor fusion technology not only improves the reliability and accuracy of the system but also enhances the system's adaptability in various complex environments.

[0016] However, the following problems may exist in existing multi-sensor fusion technologies: The computational complexity of multi-sensor data fusion algorithms is high, and there are deficiencies in the real-time performance of obstacle avoidance prediction.

[0017] To solve the above problems, in various embodiments of the present invention, a real-time parking obstacle avoidance method is provided, including: determining an obstacle avoidance grid map based on radar data information and an initial grid map; the radar data information is determined based on the ultrasonic input component of the vehicle; the initial grid map is determined based on the visual input component of the vehicle; using a preview algorithm to obtain the future movement trajectory of the vehicle in the obstacle avoidance grid map; determining whether there are obstacles at the coordinate points on the future movement trajectory in the obstacle avoidance grid map and whether there is a collision risk between the obstacles and the vehicle, and determining the obstacle avoidance strategy of the vehicle according to the judgment result.

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of 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 of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0019] According to an embodiment of the present invention, a real-time parking obstacle avoidance method is provided. Figure 1 It is a schematic flowchart of a real-time parking obstacle avoidance method provided by an embodiment of the present invention, as Figure 1 shown, and the process includes the following steps:

[0020] Step S101, determining an obstacle avoidance grid map based on radar data information and an initial grid map; the radar data information is determined based on the ultrasonic input component of the vehicle; the initial grid map is determined based on the visual input component of the vehicle.

[0021] Step S102, using a preview algorithm to obtain the future movement trajectory of the vehicle in the obstacle avoidance grid map.

[0022] Step S103, determining whether there are obstacles at the coordinate points on the future movement trajectory in the obstacle avoidance grid map and whether there is a collision risk between the obstacles and the vehicle, and determining the obstacle avoidance strategy of the vehicle according to the judgment result.

[0023] In a possible implementation, the ultrasonic input component may refer to a distance sensor on a vehicle that can measure the distance by emitting ultrasonic pulses and receiving their reflected signals, and is used to measure the distance between the vehicle and surrounding obstacles. For example, the ultrasonic input component may include, but is not limited to, a single-point ultrasonic sensor and a multi-point ultrasonic sensor. Among them, the single-point ultrasonic sensor can be used to measure the distance in a single direction, and the multi-point ultrasonic sensor can measure the distances in multiple directions simultaneously or sequentially.

[0024] Furthermore, the radar data information determined by the ultrasonic input component may refer to the data obtained through the ultrasonic input component. The functions of the radar data information may include, but are not limited to, measuring distance, detecting obstacles, and perceiving the environment.

[0025] In a possible implementation, the visual input component may refer to a visual sensor on a vehicle that is used to obtain visual information of the vehicle's surrounding environment. For example, the visual input component may include, but is not limited to, a camera, an infrared camera, and a panoramic camera.

[0026] In a possible implementation, a grid map may refer to a map representation method that divides the geographical space into regular grids or cells and associates each cell with a specific area or location on the map.

[0027] In a possible implementation, the initial grid map determined by the vehicle's visual input component can be implemented based on the following steps:

[0028] Obtain image data of the vehicle's surrounding environment according to a preset number of cameras installed on the vehicle. The image data may include, but is not limited to, photos and videos.

[0029] Preprocess the image data to improve the effect of subsequent analysis.

[0030] Use image processing and computer vision techniques to divide the image data into grids.

[0031] Determine the position and size of the grids according to the vehicle's surrounding environment, and determine the initial grid map according to the position and size of the grids.

[0032] Among them, image processing may refer to the techniques and methods for operating and processing image data, and image processing may include, but is not limited to, denoising, enhancement, filtering, and geometric transformation. Computer vision technology may refer to a technology for analyzing and understanding image data by a computer, and computer vision technology may include, but is not limited to, object recognition, image classification, and scene segmentation.

[0033] In a possible implementation, based on the radar data information determined by the ultrasonic input component of the vehicle and the initial grid map determined by the visual input component of the vehicle, the obstacle avoidance grid map can be determined, which can be implemented based on the following steps:

[0034] Adopt a perception algorithm to perform target detection and recognition on the acquired image data, determine the target information corresponding to the image data, and map the obstacles into the initial grid map;

[0035] According to the image data obtained by the camera, measure the distance between the obstacles in the surrounding environment and the vehicle, and determine the position and size of the obstacles in the initial grid map according to the distance;

[0036] Determine the obstacle avoidance grid map according to the radar data information, the position and size of the obstacles in the initial grid map.

[0037] Among them, the obstacle avoidance grid map can be used to represent the obstacles in the surrounding environment of the vehicle, and each grid in the obstacle avoidance grid map can represent the position coordinates, type and size of the obstacles; the types of obstacles can include but are not limited to: other vehicles, pedestrians, building structures, ground obstacles, movable obstacles.

[0038] In a possible implementation, the preview algorithm can refer to predicting the future motion trajectory or behavior of the vehicle within a certain period of time according to the current state of the vehicle (such as position, speed, direction), environmental conditions (such as road conditions, traffic conditions) and a pre-set motion model.

[0039] Exemplarily, according to the running state of the vehicle and the parking environment where the vehicle is located, the preview algorithm can be used to predict the future motion trajectory of the vehicle during parking.

[0040] In a possible implementation, to determine whether there are obstacles at the coordinate points on the future motion trajectory in the obstacle avoidance grid map and whether there is a collision risk between the obstacles and the vehicle, and determine the obstacle avoidance strategy of the vehicle according to the judgment result, it can be implemented based on the following steps:

[0041] Map the future motion trajectory of the vehicle into the obstacle avoidance grid map to determine whether there is obstacle information on the predicted future motion trajectory;

[0042] For the detected obstacles, evaluate whether there is a collision risk between the vehicle and the obstacles, and determine the corresponding obstacle avoidance strategy according to the evaluation result of the collision risk.

[0043] Exemplarily, the obstacle avoidance strategy can include but is not limited to: adjusting the path, reducing the vehicle speed, braking and waiting, emergency braking.

[0044] By the above method, the future motion trajectory of the vehicle can be obtained in real time by using the preview algorithm, which can assist in discovering potential obstacles and collision risks and improve the safety of parking and obstacle avoidance.

[0045] In a specific embodiment, based on the radar data information determined by the ultrasonic input component of the vehicle, it is implemented based on the following steps:

[0046] Use the ultrasonic input component of the vehicle to determine the obstacle map based on the historical parking data of the vehicle; the obstacle map includes but is not limited to: obstacle distance information, radar blind area information.

[0047] In a possible implementation manner, the historical parking data of the vehicle may include but is not limited to the obstacle information detected by the ultrasonic input component; the historical parking data may include but is not limited to: detection timestamp, obstacle distance, obstacle angle, obstacle position, obstacle shape, obstacle size.

[0048] Among them, the detection timestamp may refer to the time when the ultrasonic input component detects an obstacle, the obstacle distance may refer to the straight-line distance from the ultrasonic input component to the obstacle, the obstacle angle may refer to the angle of the obstacle relative to the vehicle, and the obstacle shape may refer to the specific contour of the obstacle.

[0049] Furthermore, determining the obstacle map based on the historical parking data of the vehicle can be implemented based on the following steps:

[0050] Convert the historical parking data detected at different times to the same time point through the vehicle motion model of the vehicle, and convert the historical parking data at different positions to the same global coordinate system;

[0051] According to the detection results of the historical parking data at the same position, determine the probability of the existence of an obstacle at this position;

[0052] Map the processed historical parking data to the grid map, and use the grids in the grid map to represent whether there is an obstacle in this area to determine the obstacle map;

[0053] Update the obstacle map in real time according to the data obtained by using the ultrasonic input component during each parking process.

[0054] Among them, the vehicle motion model may refer to a mathematical model used to describe the motion characteristics of the vehicle; for example, the vehicle motion model may include but is not limited to: bicycle model, multi-rigid body dynamics model; the bicycle model can simplify the vehicle into a "bicycle" with front and rear axles, and describe the steering and side-slip behavior of the vehicle by considering the front-wheel steering angle and speed of the vehicle; the multi-rigid body dynamics model can consider the dynamic characteristics of each component of the vehicle, such as the body, tires, and suspension system, for high-precision simulation and complex motion prediction.

[0055] Through the above method, by determining the obstacle map, even if the obstacle is in the radar blind area of the vehicle, the obstacle can be recognized based on the obstacle map, thereby improving the accuracy of obstacle avoidance; on the other hand, by determining the obstacle map through historical parking data, the real-time performance of obstacle avoidance detection during the current parking process can be improved.

[0056] In a specific embodiment, based on the radar data information and the initial grid map, an obstacle avoidance grid map is determined, which is implemented based on the following steps:

[0057] Use the visual input component of the vehicle to obtain the image of the environment where the vehicle is located, determine the current environment information corresponding to the environment where the vehicle is located based on the environment image, and update the current environment information to the environment information set in the data storage system. The data storage system includes but is not limited to: distributed file system, cloud storage system;

[0058] Use computer vision algorithms to identify obstacles based on the environment information set in the data storage system, and obtain the obstacle information in the environment information; the obstacle information includes but is not limited to: obstacle type, obstacle coordinates;

[0059] Fuse the initial grid map, obstacle information and obstacle map to determine the obstacle avoidance grid map.

[0060] In a possible implementation, the distributed file system can be a system jointly provided by at least one computer for file storage services, and the cloud storage system can be a storage service system based on the cloud computing architecture; computer vision algorithms can be a class of algorithms used to process and analyze image or video data.

[0061] For example, computer vision algorithms can include but are not limited to: image processing algorithms, image feature extraction algorithms, target detection algorithms.

[0062] In a possible implementation, determining the current environment information corresponding to the environment where the vehicle is located based on the environment image may include:

[0063] Use perception algorithms to identify the environment information in the environment image. The environment information can include but is not limited to: buildings, other vehicles, weather, parking road environment, parking road traffic conditions; the environment image can include but is not limited to: pictures, videos.

[0064] In a possible implementation, using computer vision algorithms to identify obstacles based on the environment information set in the data storage system and obtain the obstacle information in the environment information may include:

[0065] Using a machine learning algorithm, obstacle recognition is performed based on the set of environmental information in the data storage system to obtain the obstacle information in the environmental information.

[0066] Here, the machine learning algorithm can refer to a class of algorithms that learn patterns and rules from data and use these learned patterns and rules for prediction or decision-making.

[0067] For example, the machine learning algorithm can include but is not limited to: linear regression algorithm, decision tree algorithm, support vector machine (SVM) algorithm.

[0068] In a possible implementation, fusing the initial grid map, obstacle information, and obstacle map to determine the obstacle avoidance grid map may include:

[0069] Overlay the obstacle information and the obstacle map on the initial grid map to determine the obstacle avoidance grid map.

[0070] By the above method, updating the current environmental information to the set of environmental information in the data storage system and continuously iterating the perception algorithm can enable the vehicle to improve the success rate of obstacle recognition as the number of parking operations increases.

[0071] In a specific embodiment, using a preview algorithm to obtain the future motion trajectory of the vehicle in the obstacle avoidance grid map includes:

[0072] Using a preview algorithm, according to the current driving data of the vehicle and the vehicle motion model of the vehicle, predict the trajectory of the vehicle driving a preset distance in the obstacle avoidance grid map; the current driving data includes but is not limited to: the steering angle of the vehicle, the position of the vehicle; the trajectory is used to represent the continuous path of the vehicle determined by prediction.

[0073] Average-segment the trajectory according to a preset number, take the endpoints of each segment as the future positions of the vehicle, and determine the future motion trajectory of the vehicle according to the sequence of future positions composed of the future positions.

[0074] Based on the volume information of the vehicle, determine the four corner points corresponding to the vehicle in the standard coordinate system.

[0075] Using a preset transformation matrix, perform matrix transformation on the four corner points of the vehicle to obtain the position coordinates of the four corner points corresponding to the obstacle avoidance grid map, so as to represent the position coordinates of the vehicle corresponding to the obstacle avoidance grid map.

[0076] Using a preset transformation matrix, perform matrix transformation on at least one future position in the sequence of future positions to obtain the position coordinates of the sequence of future positions corresponding to the obstacle avoidance grid map, so as to represent the position coordinates of the future motion trajectory corresponding to the obstacle avoidance grid map.

[0077] In a possible implementation, a preview algorithm is adopted. Based on the current driving data of the vehicle and the vehicle motion model of the vehicle, the trajectory of the vehicle traveling a preset distance in the obstacle avoidance grid map is predicted, which can be implemented based on the following steps:

[0078] Obtain the current driving data of the vehicle and determine the preset distance as the preview distance of the preview algorithm; the preview distance is used to represent the extension distance of the predicted trajectory;

[0079] Predict the trajectory of the vehicle within the preview distance according to the vehicle motion model of the vehicle.

[0080] Among them, the ways to obtain the current driving data may include but are not limited to: in-vehicle sensors, in-vehicle computers, vehicle control units; the in-vehicle sensors may include: inertial measurement units, global positioning systems.

[0081] Exemplarily, the preset distance can be specifically set according to actual needs. For example, the preset distance can be 1 meter; the size of the set preview distance is proportional to the speed of the vehicle.

[0082] In a possible implementation, the trajectory is evenly segmented according to a preset quantity, and the endpoints of each segment are taken as the future positions of the vehicle. The future motion trajectory of the vehicle is determined according to the future position sequence composed of the future positions, which can be implemented based on the following steps:

[0083] Adopt a preview algorithm to predict the trajectory of the vehicle within the preset preview distance, and divide the trajectory into corresponding numbers of segments according to the preset quantity; the preset quantity can be specifically set according to the accuracy requirement;

[0084] Take the endpoints of each segment as the future positions of the vehicle, and form a future position sequence by arranging the endpoints in the order in the trajectory, so as to represent the position path during the future driving process of the vehicle.

[0085] Exemplarily, the preset quantity can be 10. The trajectory with a length of 1 meter is divided into 10 segments according to the preset quantity, and the 10 endpoints corresponding to the 10 segments are taken as the future positions of the vehicle.

[0086] Here, the future position sequence representing the future motion trajectory of the vehicle is determined through the preview algorithm. By determining discrete points, it provides data support for subsequent obstacle avoidance recognition. Compared with continuous trajectories, the calculation of discrete points is simpler and more efficient. Each discrete point can be considered separately, which is convenient for real-time detection and adjustment, thereby improving the real-time performance of obstacle avoidance recognition.

[0087] In a possible implementation, based on the volume information of the vehicle, the four corner points corresponding to the vehicle in the standard coordinate system are determined, which can be implemented based on the following steps:

[0088] Based on the volume information of the vehicle, determine the length, width, and height corresponding to the vehicle;

[0089] Taking the center of the vehicle set as the origin, the vehicle's traveling direction as the positive x-axis direction, the lateral direction as the positive y-axis direction, and the vertically upward direction as the positive z-axis direction, determine the local coordinate system corresponding to the vehicle;

[0090] According to the length, width, and height of the vehicle and the local coordinate system, determine the relative positions of the four corner points of the vehicle in the local coordinate system;

[0091] According to the current position of the vehicle and the coordinates of the four corner points in the local coordinate system, perform coordinate transformation to determine the coordinates of the four corner points in the standard coordinate system.

[0092] In a possible implementation, a preset transformation matrix is used to perform matrix transformation on the four corner points of the vehicle to obtain the position coordinates of the four corner points corresponding to the obstacle avoidance grid map for representing the position coordinates of the vehicle corresponding to the obstacle avoidance grid map, which can be implemented based on the following steps:

[0093] Determine the preset first transformation matrix, and use the first transformation matrix to perform coordinate transformation on the coordinates of each corner point in the standard coordinate system to determine the transformed corner point coordinates;

[0094] Determine the size of each grid in the obstacle avoidance grid map, and divide the transformed corner point coordinates by the grid size to determine the position coordinates of the four corner points in the obstacle avoidance grid map.

[0095] Further, when the determined position coordinates are outside the valid range of the obstacle avoidance grid map, perform preset processing on the position coordinates that exceed the valid range. The preset processing includes but is not limited to: cropping, ignoring.

[0096] In a possible implementation, a preset transformation matrix is used to perform matrix transformation on at least one future position in the future position sequence to obtain the position coordinates of the future position sequence corresponding to the obstacle avoidance grid map for representing the position coordinates of the future movement trajectory corresponding to the obstacle avoidance grid map, which can be implemented based on the following steps:

[0097] Determine the preset second transformation matrix, and use the second transformation matrix to perform coordinate transformation on the coordinates of each future position in the standard coordinate system to determine the transformed future position coordinates;

[0098] Determine the size of each grid in the obstacle avoidance grid map, and divide the transformed future position coordinates by the grid size to determine the position coordinates of each future position in the obstacle avoidance grid map.

[0099] Through the above method, by combining the preview algorithm with the vehicle motion model, the motion trajectory of the vehicle in the future for a period of time can be predicted in real time, enabling the vehicle to react in advance during the parking operation, enhancing the real-time performance and accuracy of obstacle avoidance; enabling the vehicle to adjust the driving strategy in a timely manner, avoiding potential collisions, and greatly improving the safety of the parking process; the preview algorithm can adjust the predicted trajectory in real time according to the dynamic state of the vehicle and environmental changes, improving the adaptability of obstacle avoidance.

[0100] In a specific embodiment, it is determined whether there are obstacles at the coordinate points on the future motion trajectory in the obstacle avoidance grid map, and whether there is a collision risk between the obstacles and the vehicle. According to the judgment result, the obstacle avoidance strategy of the vehicle is determined based on the following steps:

[0101] If it is determined according to the obstacle avoidance grid map that there are obstacles on the future motion trajectory of the vehicle, it is determined whether there is a collision risk between the obstacles and the vehicle;

[0102] If it is determined that there is a collision risk between the obstacles and the vehicle, the position coordinates corresponding to the obstacles in the obstacle avoidance grid map are determined;

[0103] Based on the position coordinates of the vehicle and the position coordinates of the obstacles, a first distance is determined; the first distance is used to represent the maximum moving distance before the vehicle collides with the obstacles;

[0104] An ultrasonic input component of the vehicle is used to obtain a second distance; the second distance is used to represent the minimum distance from the obstacles on the future motion trajectory of the vehicle;

[0105] Based on the first distance and the second distance, the obstacle avoidance strategy of the vehicle is determined.

[0106] Specifically, according to the future motion trajectory of the vehicle, the motion direction of the vehicle is determined, and it is judged whether there are obstacles in the motion direction of the vehicle; when there are obstacles, each obstacle is traversed one by one to judge whether there is a collision risk between the obstacle and the vehicle; when there is a collision risk, according to the position coordinates of the vehicle and the position coordinates of the obstacles in the obstacle avoidance grid map, the first distance representing the maximum moving distance before the collision is calculated, and according to the first distance and the second distance determined by the ultrasonic input component, the obstacle avoidance strategy of the vehicle is determined.

[0107] Through the above method, the potential collision risk with the obstacles can be determined according to the future motion path of the vehicle, improving the real-time performance of the obstacle avoidance decision-making, enabling the vehicle to quickly respond to environmental changes; by determining the minimum distance between the obstacles and the vehicle and the maximum distance that the vehicle can move safely, the collision risk can be accurately evaluated and corresponding obstacle avoidance decisions can be made, thereby improving the accuracy of the obstacle avoidance decision-making.

[0108] In a specific embodiment, when an obstacle exists on the future movement trajectory of the vehicle determined according to the obstacle avoidance grid map, it is determined whether there is a collision risk between the obstacle and the vehicle, which is implemented based on the following steps:

[0109] Use the visual input component of the vehicle to obtain the motion state information of the obstacle, and determine the state type of the obstacle as stationary or moving according to the motion state information; the operating state information includes but is not limited to: motion direction, motion speed;

[0110] When it is determined that the state type of the obstacle is moving, predict the future travel trajectory of the obstacle according to the motion state information of the obstacle, and map the future travel trajectory of the obstacle into the obstacle avoidance grid map;

[0111] Based on the future movement trajectory of the vehicle and the future travel trajectory of the obstacle in the obstacle avoidance grid map, determine whether there is a collision risk between the obstacle and the vehicle.

[0112] Specifically, use a camera to obtain the operating state information of the obstacle, and determine whether the obstacle is in a stationary state or a moving state through a computer vision algorithm; when the obstacle is in a moving state, use a motion model to predict the future travel trajectory of the obstacle, and map the future travel trajectory into the obstacle avoidance grid map so that the future travel trajectory can accurately reflect the movement path of the obstacle; compare the future movement trajectory of the vehicle with the future travel trajectory of the obstacle in the obstacle avoidance grid map, determine whether each coordinate point on the future movement trajectory intersects or is adjacent to the future travel trajectory, and determine whether there is a collision risk between the obstacle and the vehicle.

[0113] Through the above method, use a computer vision algorithm to update the motion state information of the obstacle in a timely manner, and use an operating model to predict the future travel trajectory, thereby improving the real-time performance and accuracy of the obstacle avoidance decision-making; by comparing the future movement trajectory of the vehicle with the future travel trajectory of the obstacle, the collision avoidance rate of the vehicle for moving obstacles can be effectively improved.

[0114] In a specific embodiment, based on the first distance and the second distance, determine the obstacle avoidance strategy of the vehicle, including:

[0115] Take the minimum value of the first distance and the second distance as the obstacle avoidance distance;

[0116] According to the size relationship between the obstacle avoidance distance and the obstacle avoidance threshold, determine the obstacle avoidance strategy of the vehicle for the obstacle avoidance distance; the obstacle avoidance strategy includes but is not limited to: braking, decelerating, adjusting the motion direction.

[0117] Specifically, take the minimum value of the first distance and the second distance as the obstacle avoidance distance to ensure that the vehicle can safely avoid the obstacle in the predicted worst case; when the obstacle avoidance distance is less than or equal to the obstacle avoidance threshold, it indicates that the vehicle needs to adopt an obstacle avoidance strategy.

[0118] In a possible implementation, when the difference between the obstacle avoidance threshold and the obstacle avoidance distance is greater than the first threshold, it indicates that the vehicle will definitely collide with the obstacle. At this time, the obstacle avoidance strategy can be braking and / or direction adjustment; when the difference between the obstacle avoidance threshold and the obstacle avoidance distance is less than the first threshold and greater than the second threshold, it indicates that the vehicle may collide with the obstacle. At this time, the obstacle avoidance strategy can be deceleration.

[0119] Through the above method, by determining the obstacle avoidance distance in real time, the probability of collision between the vehicle and the obstacle can be effectively reduced, thereby improving driving safety; when the obstacle avoidance distance permits, the system can smoothly adjust the vehicle speed and direction, which can improve the comfort of maintaining driving.

[0120] Figure 2 It is a schematic flowchart of another real-time obstacle avoidance method for parking provided by an embodiment of the present invention. As Figure 2 shown, the process includes the following steps:

[0121] Step S201, output a preliminary grid map;

[0122] Here, the vehicle's vision input component is used to determine the preliminary grid map (i.e., the initial grid map);

[0123] Step S202, introduce big data and machine learning algorithms to identify the obstacle type and output the obstacle coordinates;

[0124] Here, the big data method is used to update the current environmental information to the environmental information set in the data storage system. The machine learning algorithm is used to perform obstacle recognition based on the environmental information set in the data storage system to obtain the obstacle information in the environmental information;

[0125] Step S203, combine historical data to draw an obstacle map;

[0126] Here, based on the vehicle's historical parking data, the obstacle map is determined;

[0127] Step S204, fuse the grid map, obstacle type, and radar obstacle map data, and output a grid map for obstacle avoidance;

[0128] Here, fuse the grid map (i.e., the initial grid map), obstacle type, and radar obstacle map data (i.e., the obstacle map), and output a grid map for obstacle avoidance (i.e., the obstacle avoidance grid map);

[0129] Step S205, predict the trajectory of the vehicle traveling 1 m according to the current gear and steering wheel angle (i.e., the current driving data), divide it into 10 equal segments on average, and take the endpoints as the future positions of the vehicle;

[0130] Here, it is equivalent to predicting the trajectory of the vehicle traveling a preset distance in the obstacle avoidance grid map according to the current driving data of the vehicle and the vehicle motion model of the vehicle, evenly segmenting the trajectory according to a preset number, taking the endpoints of each segment as the future positions of the vehicle, and determining the future motion trajectory of the vehicle according to the future position sequence composed of the future positions;

[0131] Step S206, according to the vehicle size (i.e., volume information) and the future position coordinates of the vehicle (i.e., the future position sequence), perform matrix calculations to obtain 4 sides of the vehicle and convert them into coordinates within the grid map;

[0132] Here, it is equivalent to determining the four corner points corresponding to the vehicle in the standard coordinate system based on the volume information of the vehicle; using a preset transformation matrix to perform matrix transformation on the four corner points corresponding to the vehicle to determine the position coordinates corresponding to the vehicle in the obstacle avoidance grid map;

[0133] Step S207, according to the vehicle motion direction, determine whether there are obstacles at the coordinate points of the side in the vehicle motion direction; if so, go to step S208;

[0134] Here, it is equivalent to determining whether there are obstacles at the coordinate points on the future motion trajectory in the obstacle avoidance grid map;

[0135] Step S208, according to the coordinate of the current vehicle position (i.e., the vehicle position coordinate) and the coordinate of the current point (i.e., the obstacle position coordinate), calculate the distance S1 (i.e., the first distance) that the vehicle can move before collision;

[0136] Here, it is equivalent to determining the first distance based on the vehicle position coordinate and the obstacle position coordinate, and the first distance is used to represent the maximum moving distance before the vehicle collides with the obstacle;

[0137] Step S209, output the distance in real-time by ultrasonic wave;

[0138] Here, an ultrasonic input component is used to obtain the ultrasonic output distance;

[0139] Step S210, according to the distance output by the ultrasonic wave and the vehicle motion direction (i.e., the future motion trajectory), take the minimum distance S2 (i.e., the second distance) from the obstacle in the vehicle motion direction;

[0140] Here, it is equivalent to using the ultrasonic input component of the vehicle to obtain the second distance, and the second distance is used to represent the minimum distance from the obstacle on the future motion trajectory of the vehicle;

[0141] Step S211, take the minimum value S (i.e., the obstacle avoidance distance) of S1 and S2, and perform different strategies (i.e., obstacle avoidance strategies) according to S, such as decelerating or braking;

[0142] Here, the minimum value of the first distance and the second distance is used as the obstacle avoidance distance. According to the magnitude relationship between the obstacle avoidance distance and the obstacle avoidance threshold, the obstacle avoidance strategy of the vehicle for the obstacle avoidance distance is determined. The obstacle avoidance strategy includes, but is not limited to: braking, decelerating, and adjusting the direction.

[0143] Figure 3 It is a schematic structural diagram of a real-time parking obstacle avoidance device provided by an embodiment of the present invention. As Figure 3 shown, the device can be applied to intelligent electronic devices such as servers and computers; the device includes: a determination module 301, a prediction module 302, and an obstacle avoidance module 303;

[0144] Among them, the determination module 301 is used to determine an obstacle avoidance grid map based on radar data information and an initial grid map; the radar data information is determined based on the ultrasonic input component of the vehicle; the initial grid map is determined based on the visual input component of the vehicle;

[0145] The prediction module 302 is used to adopt a preview algorithm to obtain the future motion trajectory of the vehicle in the obstacle avoidance grid map;

[0146] The obstacle avoidance module 303 is used to determine whether there are obstacles at the coordinate points on the future motion trajectory in the obstacle avoidance grid map, and whether there is a collision risk between the obstacles and the vehicle. According to the judgment result, the obstacle avoidance strategy of the vehicle is determined.

[0147] In a specific embodiment, the determination module 301 is used to use the ultrasonic input component of the vehicle to determine an obstacle map based on the historical parking data of the vehicle; the obstacle map includes, but is not limited to: obstacle distance information and radar blind area information.

[0148] In a specific embodiment, the determination module 301 is used to use the visual input component of the vehicle to obtain an image of the environment where the vehicle is located, determine the current environment information corresponding to the environment where the vehicle is located based on the environment image, and update the current environment information to the environment information set in the data storage system. The data storage system includes, but is not limited to: a distributed file system and a cloud storage system;

[0149] Adopt a computer vision algorithm to perform obstacle recognition based on the environment information set in the data storage system to obtain obstacle information in the environment information; the obstacle information includes, but is not limited to: obstacle type and obstacle coordinates;

[0150] Fuse the initial grid map, obstacle information, and obstacle map to determine the obstacle avoidance grid map.

[0151] In a specific embodiment, the prediction module 302 is configured to use a preview algorithm to predict the trajectory of the vehicle traveling a preset distance in the obstacle avoidance grid map according to the current driving data of the vehicle and the vehicle motion model of the vehicle; the current driving data includes but is not limited to: the steering angle of the vehicle, the position of the vehicle; the trajectory is used to characterize the continuous path of the predicted vehicle.

[0152] Average the trajectory into segments according to a preset quantity, take the endpoints of each segment as the future positions of the vehicle, and determine the future motion trajectory of the vehicle according to the sequence of future positions formed by the future positions.

[0153] Based on the volume information of the vehicle, determine the four corner points corresponding to the vehicle in the standard coordinate system.

[0154] Use a preset transformation matrix to perform matrix transformation on the four corner points of the vehicle to obtain the position coordinates of the four corner points corresponding to the obstacle avoidance grid map, so as to represent the position coordinates of the vehicle corresponding to the obstacle avoidance grid map.

[0155] Use a preset transformation matrix to perform matrix transformation on at least one future position in the sequence of future positions to obtain the position coordinates of the sequence of future positions corresponding to the obstacle avoidance grid map, so as to represent the position coordinates of the future motion trajectory corresponding to the obstacle avoidance grid map.

[0156] In a specific embodiment, the obstacle avoidance module 303 is configured to determine whether there is a collision risk between the obstacle and the vehicle if there is an obstacle on the future motion trajectory of the vehicle determined according to the obstacle avoidance grid map.

[0157] If it is determined that there is a collision risk between the obstacle and the vehicle, determine the position coordinates of the obstacle corresponding to the obstacle avoidance grid map.

[0158] Based on the position coordinates of the vehicle and the position coordinates of the obstacle, determine a first distance; the first distance is used to represent the maximum moving distance before the vehicle collides with the obstacle.

[0159] Use the ultrasonic input component of the vehicle to obtain a second distance; the second distance is used to represent the minimum distance from the obstacle on the future motion trajectory of the vehicle.

[0160] Based on the first distance and the second distance, determine the obstacle avoidance strategy of the vehicle.

[0161] In a specific embodiment, the obstacle avoidance module 303 is configured to use the visual input component of the vehicle to obtain the motion state information of the obstacle, and determine the state type of the obstacle as stationary or moving according to the motion state information; the operating state information includes but is not limited to: the motion direction, the motion speed.

[0162] When it is determined that the state type of the obstacle is moving, based on the motion state information of the obstacle, predict the future travel trajectory of the obstacle, and map the future travel trajectory of the obstacle into the obstacle avoidance grid map;

[0163] Based on the future motion trajectory of the vehicle and the future travel trajectory of the obstacle in the obstacle avoidance grid map, determine whether there is a collision risk between the obstacle and the vehicle.

[0164] In a specific embodiment, the obstacle avoidance module 303 is used to take the minimum value of the first distance and the second distance as the obstacle avoidance distance;

[0165] According to the magnitude relationship between the obstacle avoidance distance and the obstacle avoidance threshold, determine the obstacle avoidance strategy of the vehicle for the obstacle avoidance distance; the obstacle avoidance strategy includes but is not limited to: braking, decelerating, and adjusting the motion direction.

[0166] It should be noted that: when the above-mentioned parking real-time obstacle avoidance device implements the corresponding parking real-time obstacle avoidance method, only the above-mentioned division of each program module is used for illustration. In practical applications, the above-mentioned processing can be allocated to different program modules according to needs, that is, the internal structure of the device is divided into different program modules to complete all or part of the above-mentioned processing. In addition, the device provided in the above-mentioned embodiment and the corresponding Figure 1 The embodiment of the method shown belongs to the same concept, and the specific implementation process can be seen in the method embodiment, which will not be repeated here.

[0167] The embodiment of the present invention further provides a computer device having the above Figure 3 Shown parking real-time obstacle avoidance device.

[0168] Please refer to Figure 4 , Figure 4 is a schematic structural diagram of another parking real-time obstacle avoidance device provided by the embodiment of the present invention. As Figure 4 shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Each component communicates with each other using different buses and can be installed on a common motherboard or installed in other ways according to needs. The processor can process instructions executed within the computer device, including instructions stored in the memory or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some alternative embodiments, if necessary, multiple processors and / or multiple buses can be used with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a set of blade servers, or a multi-processor system). Figure 4 In

[0169] The processor 10 may be a central processing unit, a network processor, or a combination thereof. Among them, the processor 10 may further include a hardware chip. The above-mentioned hardware chip may be an application-specific integrated circuit, a programmable logic device, or a combination thereof. The above-mentioned programmable logic device may be a complex programmable logic device, a field programmable gate array, a generic array logic, or any combination thereof.

[0170] Among them, the memory 20 stores instructions executable by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiments.

[0171] The memory 20 may include a program storage area and a data storage area. Among them, the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may further include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some alternative embodiments, the memory 20 may optionally include a memory remotely provided with respect to the processor 10, and these remote memories may be connected to the computer device through a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an enterprise intranet, a local area network, a mobile communication network, and combinations thereof.

[0172] The memory 20 may include a volatile memory, for example, a random access memory; the memory may also include a non-volatile memory, for example, a flash memory, a hard disk, or a solid-state drive; the memory 20 may further include a combination of the above types of memories.

[0173] The computer device further includes a communication interface 30 for the computer device to communicate with other devices or communication networks.

[0174] Embodiments of the present invention also provide a computer-readable storage medium. The method according to the embodiments of the present invention can be implemented in hardware, firmware, or be implemented as computer code that can be recorded on a storage medium, or be implemented as computer code that is originally stored in a remote storage medium or a non-transitory machine-readable storage medium and downloaded through a network and will be stored in a local storage medium. Thus, the method described herein can be stored as such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, or a solid-state drive, etc.; further, the storage medium can also include a combination of the above-mentioned types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by the computer, the processor, or the hardware, the method shown in the above embodiments is implemented.

[0175] A part of the present invention can be applied as a computer program product, such as computer program instructions. When executed by a computer, through the operation of the computer, the methods and / or technical solutions according to the present invention can be invoked or provided. Those skilled in the art should be able to understand that the forms of existence of computer program instructions in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executes the instructions, or the computer compiles the instructions and then executes the corresponding compiled program, or the computer reads and executes the instructions, or the computer reads and installs the instructions and then executes the corresponding installed program. Herein, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to the computer.

[0176] Although the embodiments of the present invention are described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A real-time parking obstacle avoidance method, characterized in that: The method comprises: Determining an obstacle avoidance grid map based on radar data information and an initial grid map; the radar data information is determined based on an ultrasonic input component of the vehicle; the initial grid map is determined based on a visual input component of the vehicle; Using a preview algorithm to obtain the future motion trajectory of the vehicle in the obstacle avoidance grid map; Determine whether there is an obstacle at the coordinate point on the future motion trajectory in the obstacle avoidance grid map, and whether there is a risk of collision between the obstacle and the vehicle, and determine the obstacle avoidance strategy of the vehicle according to the determination result; The method of using a preview algorithm to obtain the future motion trajectory of the vehicle in the obstacle avoidance grid map includes: A preview algorithm is used to predict the trajectory of the vehicle traveling a preset distance in the obstacle avoidance grid map based on the current driving data of the vehicle and the vehicle motion model of the vehicle; the current driving data includes but is not limited to: the steering angle of the vehicle and the position of the vehicle; the trajectory is used to characterize the predicted continuous path of the vehicle; The trajectory is evenly divided into segments according to a preset number, the endpoints of each segment are taken as the future position of the vehicle, and the future motion trajectory of the vehicle is determined according to a future position sequence composed of the future positions; Based on the volume information of the vehicle, determining four corner points corresponding to the vehicle in a standard coordinate system; Using a preset transformation matrix, the four corner points of the vehicle are subjected to matrix transformation to obtain position coordinates of the four corner points corresponding to the obstacle avoidance grid map, so as to represent the position coordinates of the vehicle corresponding to the obstacle avoidance grid map; The preset transformation matrix is ​​used to perform a matrix transformation on at least one future position in the future position sequence to obtain the position coordinates of the future position sequence corresponding to the obstacle avoidance grid map, so as to characterize the position coordinates of the future motion trajectory corresponding to the obstacle avoidance grid map.

2. The method according to claim 1, characterized in that Based on the radar data information determined by the vehicle's ultrasonic input component, this is achieved based on the following steps: The ultrasonic input component of the vehicle is used to determine an obstacle map based on the historical parking data of the vehicle; the obstacle map includes but is not limited to: obstacle distance information and radar blind spot information.

3. The method according to claim 2, characterized in that Based on the radar data information and the initial grid map, the obstacle avoidance grid map is determined, which is implemented based on the following steps: Using a visual input component of the vehicle to obtain an image of the environment in which the vehicle is located, determining current environmental information corresponding to the environment in which the vehicle is located based on the environmental image, and updating the current environmental information to an environmental information set in a data storage system, wherein the data storage system includes but is not limited to: a distributed file system and a cloud storage system; Using a computer vision algorithm, obstacle identification is performed based on the environmental information set in the data storage system to obtain obstacle information in the environmental information; The obstacle information includes but is not limited to: obstacle type and obstacle coordinates; The initial grid map, the obstacle information and the obstacle map are integrated to determine an obstacle avoidance grid map.

4. The method according to claim 1, characterized in that: Determine whether there is an obstacle at the coordinate point on the future motion trajectory in the obstacle avoidance grid map, and whether there is a risk of collision between the obstacle and the vehicle, and determine the obstacle avoidance strategy of the vehicle based on the judgment result, based on the following steps: If it is determined according to the obstacle avoidance grid map that there is an obstacle on the future motion trajectory of the vehicle, determining whether there is a risk of collision between the obstacle and the vehicle; If it is determined that there is a risk of collision between the obstacle and the vehicle, determining the position coordinates corresponding to the obstacle in the obstacle avoidance grid map; Determining a first distance based on the position coordinates of the vehicle and the position coordinates of the obstacle; the first distance is used to represent the maximum moving distance before the vehicle collides with the obstacle; Using an ultrasonic input component of the vehicle to obtain a second distance; the second distance is used to represent the minimum distance from the obstacle on the future motion trajectory of the vehicle; An obstacle avoidance strategy for the vehicle is determined based on the first distance and the second distance.

5. The method according to claim 4, characterized in that If it is determined based on the obstacle avoidance grid map that there is an obstacle on the future motion trajectory of the vehicle, it is determined whether there is a risk of collision between the obstacle and the vehicle, which is achieved based on the following steps: The visual input component of the vehicle is used to obtain the motion state information of the obstacle, and the state type of the obstacle is determined to be static or moving according to the motion state information; the motion state information includes but is not limited to: motion direction, motion speed; When it is determined that the state type of the obstacle is motion, predicting the future moving trajectory of the obstacle according to the motion state information of the obstacle, and mapping the future moving trajectory of the obstacle into the obstacle avoidance grid map; Based on the future motion trajectory of the vehicle and the future travel trajectory of the obstacle in the obstacle avoidance grid map, it is determined whether there is a collision risk between the obstacle and the vehicle.

6. The method according to claim 4, characterized in that The determining, based on the first distance and the second distance, an obstacle avoidance strategy of the vehicle includes: The minimum value between the first distance and the second distance is used as the obstacle avoidance distance; According to the magnitude relationship between the obstacle avoidance distance and the obstacle avoidance threshold, an obstacle avoidance strategy of the vehicle for the obstacle avoidance distance is determined; the obstacle avoidance strategy includes but is not limited to: braking, deceleration, and movement direction adjustment.

7. A parking real-time obstacle avoidance device, characterized in that: The device comprises: A determination module, configured to determine an obstacle avoidance grid map based on radar data information and an initial grid map; the radar data information is determined based on an ultrasonic input component of the vehicle; and the initial grid map is determined based on a visual input component of the vehicle; A prediction module, used to obtain the future motion trajectory of the vehicle in the obstacle avoidance grid map by using a preview algorithm; An obstacle avoidance module is used to determine whether there is an obstacle at the coordinate point on the future motion trajectory in the obstacle avoidance grid map, and whether there is a risk of collision between the obstacle and the vehicle, and determine the obstacle avoidance strategy of the vehicle according to the judgment result; The prediction module is used to use a preview algorithm to predict the trajectory of the vehicle traveling a preset distance in the obstacle avoidance grid map based on the current driving data of the vehicle and the vehicle motion model of the vehicle; the current driving data includes but is not limited to: the steering angle of the vehicle and the position of the vehicle; the trajectory is used to characterize the continuous path of the vehicle determined by the prediction; The trajectory is evenly divided into segments according to a preset number, the endpoints of each segment are taken as the future position of the vehicle, and the future motion trajectory of the vehicle is determined according to a future position sequence composed of the future positions; Based on the volume information of the vehicle, determining four corner points corresponding to the vehicle in a standard coordinate system; Using a preset transformation matrix, the four corner points of the vehicle are subjected to matrix transformation to obtain position coordinates of the four corner points corresponding to the obstacle avoidance grid map, so as to represent the position coordinates of the vehicle corresponding to the obstacle avoidance grid map; The preset transformation matrix is ​​used to perform a matrix transformation on at least one future position in the future position sequence to obtain the position coordinates of the future position sequence corresponding to the obstacle avoidance grid map, so as to characterize the position coordinates of the future motion trajectory corresponding to the obstacle avoidance grid map.

8. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the real-time parking obstacle avoidance method according to any one of claims 1 to 6 by executing the computer instructions.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the real-time parking obstacle avoidance method according to any one of claims 1 to 6.

Images