Intelligent auxiliary vehicle searching method and system based on mobile robot

Abstract

The invention discloses an intelligent auxiliary vehicle searching method and system based on a mobile robot. The method comprises the following steps: acquiring attribute information and position information of a to-be-searched vehicle; vehicle searching path planning is carried out based on an improved A * algorithm, and an initial vehicle searching path is obtained; carrying out smoothing processing on the initial vehicle searching path; based on a dynamic obstacle avoidance TEB algorithm, designing a reward or penalty function for each influence factor by configuring the weight of each influence factor of an obstacle, a speed and an acceleration, scoring feasible routes, and selecting a local route with the highest score to send a speed instruction; the acceleration and the speed of the robot are restrained, so that the robot bypasses the obstacle to reach the target point fastest under kinematics constraint; the mobile robot can be used for assisting the user in querying the vehicle and autonomously guiding the user to complete rapid vehicle searching, compared with an intelligent vehicle searching platform or applet put in the current market, the experience feeling is better, the vehicle searching efficiency is higher, and equipment safety and use specifications are guaranteed.

Description

Technical Field

[0001] The present invention belongs to the field of artificial intelligence robot applications, and in particular relates to an intelligent assisted vehicle-finding method and system based on a mobile robot. Background Art

[0002] With the development of society and technological progress, people's living standards are constantly improving, and the number of private cars is increasing. Although the popularity of cars has facilitated people's travel life, it is also accompanied by many problems, the most common of which is "difficulty in parking". In order to solve this problem, a large number of parking lots have been built in the city, and the scale of parking lots has become increasingly large, which has led to the problem of "difficulty in finding cars". Especially in places with large traffic such as large supermarkets, general hospitals, tourist attractions, airports, and railway stations, underground parking lots not only have multi-layer problems, but also the area of ​​single-layer areas is also very large; the garage has a complex structure, poor light, wall obstruction, and no landmarks to guide, so it is easy to get lost, so it is difficult to find a vehicle. In addition, there are many elevators, and the exit locations of different elevators are different. Even if different partitions and parking space numbers are marked, it is still difficult to find, not to mention the frequent occurrence of misremembering or forgetting parking space numbers.

[0003] At present, in response to the application needs of the majority of users, many parking lots are equipped with intelligent car-finding platforms. Users can get the partition and parking space number information of the vehicle by entering the license plate number, but they cannot provide a specific route map for how to reach the parking location; or some smart terminal applets can display the car-finding path, but due to the complex garage environment, users are still easily lost, and for older users, it is difficult to operate the applets, which ultimately affects the efficiency of car-finding and reduces the user experience. Therefore, the intelligent car-finding platforms and program designs currently launched in the market have not really solved the problem of users' "difficulty in finding cars". Summary of the invention

[0004] In order to solve the problems existing in the prior art, the present invention provides an intelligent assisted car-finding system based on a mobile robot. By integrating a human-machine interaction panel on the mobile robot equipment, the human-machine interaction panel can be used by the user to input the license plate information of the vehicle to be found to query the parking space number of the parking place, an intelligent path planning module for quickly planning the shortest car-finding path, a navigation module for controlling the robot to move autonomously along the planned path, and a sensor module for sensing environmental changes and obstacle detection during the movement of the robot. The robot can complete the car-finding guidance according to the user's needs, solving the above-mentioned problems of difficulty in finding a car, low efficiency, and poor experience.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is: an intelligent assisted vehicle finding method based on a mobile robot, comprising the following steps: Obtain the property information and location information of the vehicle to be found; Car-searching path planning is performed based on the improved A* algorithm to obtain the initial car-searching path; Smoothing the initial vehicle-searching path; The dynamic obstacle avoidance TEB algorithm configures the weights of the influencing factors such as obstacles, speed and acceleration, designs a reward or penalty function for each influencing factor, scores the feasible routes, selects the local route with the highest score to issue a speed command, and constrains the acceleration and speed of the robot so that the robot can bypass obstacles and reach the target point as quickly as possible under its kinematic constraints.

[0006] Furthermore, obtaining the location information of the vehicle to be found includes: Extract multi-source single-frame images at the same moment from multi-source video streams; Complete the recognition of all license plate numbers from each frame of the image; Get the parking space number of the vehicle.

[0007] Furthermore, based on the improved A* algorithm, the vehicle search path planning is performed. When the initial vehicle search path is obtained, the dynamic weight is set according to the environmental characteristics and target requirements of the path planning to improve the heuristic function. In the path planning, the relationship between the search efficiency and the path quality is balanced according to the proportion of obstacles in the environment. The improved heuristic function is:

[0008] in, Represents the estimated cost from the initial state to the target state; It is the ratio of grids occupied by obstacles in the environment to grids occupied by the total map; Indicates the actual cost from the initial state to the current state; Represents the estimated cost from the current state to the target state.

[0009] Furthermore, based on the improved A* algorithm, the vehicle search path planning is performed to obtain the initial vehicle search path, which includes: expanding the number of search directions of the robot to 16 neighborhood searches, and deleting the search neighborhoods that exceed the steering angle threshold.

[0010] Furthermore, the vehicle search path planning is performed based on the improved A* algorithm to obtain the initial vehicle search path, which also includes: using a cubic spline interpolation path smoothing algorithm, dividing the known planned n+1 path point information into n intervals, constructing a cubic function in each interval, and the cubic function passes through all known nodes.

[0011] Furthermore, when constraining the acceleration and velocity of the robot, the position of the robot at each discrete moment and the time interval between the discrete moments are taken as vertices, and the arrival time and the distance to the obstacle are equal.

[0012] Furthermore, the routes close to obstacles are penalized, and the routes far from the middle point are penalized.

[0013] The present invention also provides an intelligent assisted vehicle-finding system based on a mobile robot, comprising a human-machine interactive display screen, a processor, a laser / radar sensor, a vision module, a navigation module and a motion module. The human-computer interactive display screen is used for the user to input the license plate information of the vehicle to be found to query the parking space number of the parking place; The processor is used to plan the shortest intelligent path to find the vehicle and issue acceleration and speed commands; Laser / radar sensors are used to obtain obstacle information; The visual module is used to assist in obtaining the attribute information and location information of the vehicle to be found; The motion module is used to execute the acceleration and velocity instructions of the processor.

[0014] The present invention can also provide a computer device, including a processor and a memory, the memory is used to store a computer executable program, the processor reads the computer executable program from the memory and executes it, and when the processor executes the computer executable program, it can implement the intelligent assisted vehicle finding method based on a mobile robot described in the present invention.

[0015] At the same time, a computer-readable storage medium is provided, in which a computer program is stored. When the computer program is executed by a processor, the intelligent assisted vehicle-finding method based on a mobile robot described in the present invention can be implemented.

[0016] Compared with the prior art, the present invention has at least the following beneficial effects: The present invention proposes an intelligent assisted car-finding robot path planning method based on improved A*, targeting two major technical challenges faced in robot path planning research: one is that the internal structure of the underground garage is complex, with various passages and entrances and exits crisscrossing, which increases the difficulty of path planning; the other is that there are many pedestrians and vehicles in the garage, and the robot needs to quickly solve the problem of dynamic obstacle avoidance while driving. First, from the perspective of theoretical algorithms, with the goal of shortening the path search time and distance and adjusting the path smoothness, in the design of the intelligent path planning module, by improving the heuristic function of the A* algorithm and adding a dynamic weight factor, the algorithm can dynamically adjust the heuristic function according to the environment of different complexity, so as to reduce the number of nodes that need to be traversed in the path search, further reduce the search time, optimize the path quality, and improve the planning efficiency. By expanding the search neighborhood and refining the search direction, the scope of path search coverage can be increased, while avoiding the redundancy of calculation to explore a better path; secondly, in order to solve the problem of dynamic obstacle avoidance during movement, an A* algorithm integrating TEB local path planning is proposed, and more efficient dynamic obstacle avoidance is further achieved by improving TEB; finally, from the perspective of practical application, based on the ROS platform equipped with a human-computer interaction interface, path planning module, sensor equipment, motion module, etc., the design and development of the intelligent assisted car search system is completed. By integrating the optimized and improved path planning method into the intelligent assisted car search robot equipment, safe and fast car search guidance in large and complex underground garages can be achieved to meet public needs.

[0017] The present invention can utilize mobile robots to assist users in searching for vehicles and autonomously guide users to quickly find vehicles. Compared with the smart vehicle-finding platforms or applets currently on the market, the present invention has a better user experience, higher vehicle-finding efficiency, and ensures equipment safety and standardized use. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The drawings herein are incorporated into the specification and constitute a part of the specification, showing embodiments consistent with the present application, and together with the specification, are used to explain the principles of the present application. Obviously, the drawings described below are only some embodiments of the present application, and for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work. In the drawings: Figure 1 A schematic diagram of the architecture of an intelligent assisted vehicle-finding system based on a mobile robot provided in an embodiment of the present application; Figure 2 A schematic diagram of a process flow of an intelligent assisted vehicle-finding system based on a mobile robot provided in an embodiment of the present application; Figure 3 A flowchart of a license plate and parking space matching process provided in an embodiment of the present application; Figure 4A schematic diagram of an improved A* algorithm neighborhood expansion provided in an embodiment of the present application; Figure 5 A schematic diagram of cubic spline path smoothing provided in an embodiment of the present application; Figure 6 A schematic diagram of a dynamic obstacle avoidance TEB algorithm provided in an embodiment of the present application; Figure 7 A schematic diagram of an implementation process of an intelligent vehicle-finding robot device provided in an embodiment of the present application; Figure 8 A test result diagram of an improved A* algorithm heuristic function provided in an embodiment of the present application; Fig. 9 A test result diagram after searching the direction of an extended A* algorithm provided in an embodiment of the present application; Fig.10 A test result diagram after optimizing the search direction of the A* algorithm provided in an embodiment of the present application; Fig.11 A schematic diagram of a path smoothing provided in an embodiment of the present application; Fig.12 A schematic diagram of static navigation of a robot in a ROS simulation environment provided in an embodiment of the present application; DETAILED DESCRIPTION

[0019] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. The drawings in this application are only for the purpose of illustration and description and are not intended to limit the scope of protection of the present application. The flowcharts used in this application are only used to illustrate the operations implemented according to some embodiments of the present application. Those skilled in the art may add one or more other operations to the flowcharts, or may remove one or more operations from the flowcharts, under the guidance of the contents of this application.

[0020] Based on the concept of the method, the present invention can provide an intelligent assisted vehicle-finding system based on a mobile robot, characterized in that it includes a human-machine interactive display screen, a processor, a laser / radar sensor, a vision module, a navigation module and a motion module; The human-computer interactive display screen is used for the user to input the license plate information of the vehicle to be found to query the parking space number of the parking place; The processor is used to plan the shortest intelligent path to find the vehicle and issue acceleration and speed commands; Laser / radar sensors are used to obtain obstacle information; The visual module is used to assist in obtaining the attribute information and location information of the vehicle to be found; The motion module is used to execute the acceleration and velocity instructions of the processor.

[0021] First, the workflow of the intelligent assisted vehicle search system based on mobile robots is introduced from the perspective of the overall system architecture. Figure 1 As shown in the figure, the intelligent assisted vehicle search system based on mobile robots is divided into two parts: the front end and the back end. The back end is used for vehicle information collection and data processing, and the front end is used for user query and vehicle search guidance. For the specific process, see Figure 2 When the user needs to find a vehicle, he only needs to enter the license plate information in the interactive interface of the intelligent assisted car-finding robot. The system will query the parking space information that matches the license plate number through the intelligent car-finding module; then, the system will input the target parking space information into the intelligent path planning module, and use the pre-built garage environment. Based on the intelligent path planning algorithm, it can quickly generate an optimal path from the current robot position to the target parking position; finally, the navigation module of the robot can realize autonomous movement along the planned path, and avoid unknown obstacles based on the perception module to achieve safe passage.

[0022] The intelligent car search module adopts intelligent car search technology based on computer vision. Through the video surveillance equipment installed in the current underground parking lot and the cameras installed above some parking spaces, the collected multi-source video streams and digital images are transmitted to the image processor through the network to identify the license plate number and the parking space partition number, and the license plate number is intelligently matched with the parking space information. The multi-source video streams are synchronized in time, the timestamps of each stream are aligned, and the frames closest in time in each video stream are extracted to ensure that the single-frame image of each video source at the same time is extracted. For the extracted single-frame image, the license plate number position and the parking space number position are located according to the color features, and pre-processing (such as binarization, denoising, etc.) is performed to enhance the image quality; then the pixel distribution histogram of the binary image is analyzed by the projection segmentation method to find the boundary points of adjacent characters for segmentation; finally, the convolutional neural network is used to classify and recognize the input characters through feature extraction and mapping, and the recognized license plate number and the parking space number are matched and stored. The implementation process is as follows: Figure 3 As shown. Due to the high mobility of vehicles in the garage, the vehicles parked in the same parking space will change, and information needs to be updated in a timely manner. Regularly capture parking space images, identify vehicle license plate numbers, compare matching information in the database, determine whether there is a vehicle replacement, and ensure real-time data synchronization. Finally, when the user needs to find a car, he can enter the license plate number in the car search system interface at any time, and the intelligent car search system model will query the parking space number information corresponding to the license plate number from the current database.

[0023] The intelligent path planning module adopts an intelligent path planning method based on improved A*, with the parking location number output by the intelligent car search module as the end point and the robot's location as the starting point, to quickly plan a shortest car search route. The robot can move along the planned path using the built-in navigation module to guide the user to find the vehicle. There are two major technical difficulties in implementing robot path planning in a garage environment: one is that the internal structure of the underground garage is complex, with various passages and entrances and exits crisscrossing, which makes path planning more difficult; the other is that there are many pedestrians and vehicles in the garage, and the robot needs to quickly solve dynamic obstacle avoidance while driving. The traditional A* algorithm has defects such as many search nodes, long planning time, and large turning angles. Therefore, this application aims to improve the algorithm with the goal of improving path planning efficiency, optimizing the driving path, and enhancing the adaptability of the algorithm in complex environments. The specific plan is as follows: As a key part of the A* algorithm, the heuristic function can help guide the correct expansion direction by estimating the cost of the current position and the target position, which determines the efficiency of the search process and the quality of the final path. Existing heuristic functions all perform search planning by calculating the distance from the current node to the target node and the starting position. Once the heuristic function is determined, the search breadth remains fixed. When faced with complex large-scale environments, the effect is sometimes good and sometimes bad. The present invention proposes to improve the heuristic function by setting dynamic weights according to the environmental characteristics and target requirements of path planning, as shown in formula (1). In path planning, the relationship between search efficiency and path quality is balanced according to the proportion of obstacles in the environment, which improves the task adaptability and the generalization ability of the algorithm.

[0024] (1) in, Represents the estimated cost from the initial state to the target state; It is the ratio of grids occupied by obstacles in the environment to grids occupied by the total map; Indicates the actual cost from the initial state to the current state; Represents the estimated cost from the current state to the target state. Therefore, for complex environments with many obstacles, the heuristic function mainly relies on the estimated cost from the current state to the target state for path search.

[0025] The traditional A* algorithm usually uses the four directions of up, down, left, and right, or adds the neighborhood of the upper left, lower left, upper right, and lower right, a total of eight directions. The search space is not comprehensive, and the agent cannot move at any angle, resulting in a large turning angle and it is difficult to achieve the optimal path. Figure 4The neighborhood expansion shown in the figure expands the traditional 8-neighborhood search to a 16-neighborhood search, increases the number of optional nodes when the algorithm searches, and can cover more potential path choices. However, this expansion may also increase the complexity of path planning and cause the algorithm to encounter more obstacles and spend more search time during the search process. To this end, it is necessary to further optimize the search direction. The specific optimization strategy is shown in Table 1. The angle in Table 1 is the angle between the line between the current node and the target point and the east direction. Node direction 1 represents the upper left neighborhood, node direction 2 represents the upper neighborhood, node direction 3 represents the upper right neighborhood, node direction 4 represents the right neighborhood, node direction 5 represents the lower right neighborhood, node direction 6 represents the lower neighborhood, node direction 7 represents the lower left neighborhood, and node direction 8 represents the left neighborhood. Usually, the algorithm will traverse all searchable neighborhoods according to the heuristic function when searching for a path. However, based on the principle of robot motion and trajectory planning experience, larger turns are a challenge for robot motion. Therefore, the present invention will delete the search neighborhood with larger turns, that is, appropriately exclude some search neighborhoods with turning angles exceeding 150° under the current robot orientation, so as to reduce the search amount of nodes, further speed up the path search and improve the planning efficiency. Table 1 is a schematic diagram of an improved A* algorithm optimization search direction provided in an embodiment of the present application.

[0026] Table 1

[0027] As an important step in path planning, the purpose of smoothing is to adjust the path obtained by the preliminary planning to be smoother and more natural, more in line with the kinematic characteristics of the robot, thereby improving the feasibility and execution effect of the path. It is proposed to use the cubic spline interpolation path smoothing algorithm, according to the known planned n+1 path point information, divided into n intervals, and construct a cubic function in each interval. The cubic function must pass through all known nodes. Figure 5 As shown, a cubic function is constructed based on discrete data points passing through known nodes.

[0028] Due to the complex environment of the garage, with many vehicles and pedestrians, the risk of the robot colliding with dynamic obstacles during movement is high, so it is necessary to introduce a dynamic obstacle avoidance mechanism in the robot motion planning. The Timed Elastic Band (TEB) algorithm has a strong forward-looking nature and performs well in achieving dynamic obstacle avoidance. It is used to solve the real-time obstacle avoidance problem of the car-seeking robot. In actual scenarios, the robot's walking route will be affected by many factors. In order to keep a certain safe distance between the robot's walking route and dynamic obstacles, the obstacles are expanded. Combined with the current state of the robot (such as speed, acceleration) and the dynamic changes of the surrounding obstacles, the trajectory is optimized while following the existing planned path as much as possible, so that the path not only avoids obstacles in space, but also meets dynamic constraints in time, and can move along the planned shortest path.

[0029] The TEB algorithm adjusts the weight of each influencing factor and designs a reward and penalty function for each influencing factor. By adjusting the parameters, the speed and acceleration thresholds can be set to limit the robot's driving speed and prevent the robot from accelerating or decelerating sharply. The robot's position at each discrete moment and the time interval between discrete moments are taken as the vertices, and the acceleration, speed, arrival time and distance to obstacles are constrained, so that the robot can safely bypass obstacles and reach the target point as quickly as possible under its kinematic constraints. Figure 6 The following is a schematic diagram of the robot implementing local obstacle avoidance path planning based on the TEB algorithm. First, in order to stay away from obstacles, the robot needs to set the obstacle distance, obstacle expansion distance, and obstacle weight, and penalize the route close to the obstacle, refer to formula (2). Secondly, in order to pass through the middle point of the route as much as possible, parameters are set with the opposite purpose, such as the middle point weight, middle point interval, and middle point order, and the route far from the middle point is penalized, refer to formula (3).

[0030] (2) (3) Where, d min is the actual minimum distance between the TEB path and the obstacle, and r is the minimum distance defined by the parameters. It can be seen that how to reasonably configure the parameters has become the most important part of the TEB path optimization algorithm.

[0031] By combining path planning, motion navigation, and dynamic obstacle avoidance methods, the present invention designs an intelligent assisted vehicle-finding path planning system based on a mobile robot, which can realize unmanned, intelligent path generation and motion navigation, as well as safe obstacle avoidance and other functions. For the specific implementation process, see Figure 7 .

[0032] Based on the ROS platform, it is equipped with a human-computer interactive display screen, a microprocessor, a laser / radar sensor, a vision module, a navigation module, and a motion module, and integrates an intelligent auxiliary vehicle search path planning system to improve the intelligent auxiliary vehicle search robot equipment, helping users to find and guide vehicles. When the user enters the vehicle volume information on the display interface, the processor will match the vehicle position and automatically plan an effective path from the current position to the vehicle's target position. Based on the fusion information of each sensor, the mobile robot can automatically detect and avoid dynamic obstacles in motion, and achieve safe obstacle avoidance in the process of guiding users to find vehicles.

[0033] By testing the improved A* path planning algorithm in different environments and comparing parameters such as search nodes, path length, search time, and number of inflection points, the advantages and feasibility of the solution are verified.

[0034] In an environment with obstacles, the path planning effect after using different heuristic functions is as follows Figure 8 As shown in the figure, the left figure is the result of using Euclidean distance as the heuristic function, and the right figure is the result of using Manhattan distance and diagonal distance combined according to the obstacle ratio. By comparing the results, it can be seen that when using the heuristic function search combining Manhattan distance and diagonal distance, the number of nodes that need to be traversed and the number of inflection points are significantly reduced.

[0035] The path planning effect after expanding the search neighborhood from the traditional 8 directions to 16 directions is as follows Figure 8 As shown, Fig. 9 The left-center picture uses the traditional 8-direction search neighborhood, while the right-center picture uses a search neighborhood expanded to 16 directions. By comparing the test results, we can see that neighborhood expansion can cover more hidden paths and further explore a better shortest path.

[0036] In complex environments, a strategy for optimizing the search direction is proposed to address the problem of increased search time after neighborhood expansion. The planned path effect is as follows: Fig.10 As shown in the figure, the expandable search direction is selected in real time according to the current position, and some search nodes with turning angles exceeding 150° are appropriately deleted to complete the search direction optimization. By reducing unnecessary node searches, the number of nodes that need to be traversed in the path search is successfully reduced, saving search time.

[0037] The effect after path smoothing is as follows Fig.11 As shown, the blue path is the smoothed path, which has no obvious turning points and is more consistent with the kinematic characteristics of the robot.

[0038] The static obstacle avoidance effect diagram obtained based on Gazebo environment modeling and simulation in the ROS platform is as follows: Fig.12As shown, the robot can detect obstacles in real time and plan safe obstacle avoidance areas to achieve safe avoidance and smooth passage.

[0039] On the other hand, the present invention provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the intelligent assisted vehicle finding method based on a mobile robot described in the present invention can be implemented.

[0040] The present invention can also provide a computer device, including a processor and a memory, the memory is used to store a computer executable program, the processor reads the computer executable program from the memory and executes it, and when the processor executes the computer executable program, it can implement the intelligent assisted vehicle finding method based on a mobile robot described in the present invention.

[0041] The computer device may be a laptop computer, a desktop computer or a workstation.

[0042] The processor can be a central processing unit (CPU), a digital signal processor (DSP), an application specific integrated circuit (ASIC), or an off-the-shelf field programmable gate array (FPGA).

[0043] The memory of the present invention may be an internal storage unit of a laptop computer, a desktop computer or a workstation, such as a memory or a hard disk; or an external storage unit, such as a mobile hard disk or a flash memory card.

[0044] Computer-readable storage media may include computer storage media and communication media. Computer storage media include volatile and non-volatile, removable and non-removable media implemented by any method or technology for storing information such as computer-readable instructions, data structures, program modules or other data. Computer-readable storage media may include: read-only memory (ROM), random access memory (RAM), solid-state drive (SSD) or optical disk, etc. Among them, random access memory may include resistance random access memory (ReRAM) and dynamic random access memory (DRAM).

Claims

1. An intelligent assisted vehicle-finding method based on a mobile robot, characterized in that: The following steps are involved: Obtain the property information and location information of the vehicle to be found; Car-searching path planning is performed based on the improved A* algorithm to obtain the initial car-searching path; Smoothing the initial vehicle-searching path; The dynamic obstacle avoidance TEB algorithm configures the weights of the influencing factors such as obstacles, speed and acceleration, designs a reward or penalty function for each influencing factor, scores the feasible routes, selects the local route with the highest score to issue a speed command, and constrains the acceleration and speed of the robot so that the robot can bypass obstacles and reach the target point as quickly as possible under its kinematic constraints.

2. The intelligent assisted vehicle finding method based on mobile robots according to claim 1 is characterized in that: Obtaining the location information of the vehicle to be found includes: Extract multi-source single-frame images at the same moment from multi-source video streams; Complete the recognition of all license plate numbers from each frame of the image; Get the parking space number of the vehicle.

3. The intelligent assisted vehicle finding method based on mobile robots according to claim 1 is characterized in that: Based on the improved A* algorithm, the vehicle search path planning is carried out. When the initial vehicle search path is obtained, the dynamic weight is set according to the environmental characteristics and target requirements of the path planning to improve the heuristic function. In the path planning, the relationship between the search efficiency and the path quality is balanced according to the proportion of obstacles in the environment. The improved heuristic function is: in, Represents the estimated cost from the initial state to the target state; It is the ratio of grids occupied by obstacles in the environment to grids occupied by the total map; Indicates the actual cost from the initial state to the current state; Represents the estimated cost from the current state to the target state.

4. The intelligent assisted vehicle finding method based on mobile robot according to claim 1 is characterized in that: Based on the improved A* algorithm, the vehicle search path planning is performed to obtain the initial vehicle search path, including: expanding the number of search directions of the robot to 16 neighborhood searches, and deleting the search neighborhoods that exceed the steering angle threshold.

5. The intelligent assisted vehicle finding method based on mobile robot according to claim 1 is characterized in that: Based on the improved A* algorithm, the vehicle search path planning is performed to obtain the initial vehicle search path, which also includes: using the cubic spline interpolation path smoothing algorithm, dividing the known planned n+1 path point information into n intervals, constructing a cubic function in each interval, and the cubic function passes through all known nodes.

6. The intelligent assisted vehicle finding method based on mobile robots according to claim 1 is characterized in that: When constraining the acceleration and velocity of the robot, the position of the robot at each discrete moment and the time interval between discrete moments are taken as vertices, and the arrival time and the distance to the obstacle are equal.

7. The intelligent assisted vehicle finding method based on mobile robots according to claim 1 is characterized in that: Penalize routes that are close to obstacles and penalize routes that are far from the middle point.

8. An intelligent assisted vehicle-finding system based on a mobile robot, characterized in that: Including human-computer interaction display, processor, laser / radar sensor, vision module, navigation module and motion module, The human-computer interactive display screen is used for the user to input the license plate information of the vehicle to be found to query the parking space number of the parking place; The processor is used to plan the shortest intelligent path to find the vehicle and issue acceleration and speed commands; Laser / radar sensors are used to obtain obstacle information; The visual module is used to assist in obtaining the attribute information and location information of the vehicle to be found; The motion module is used to execute the acceleration and velocity instructions of the processor.

9. A computer device, characterized in that: It includes a processor and a memory, the memory is used to store computer executable programs, the processor reads part or all of the computer executable programs from the memory and executes them, and when the processor executes part or all of the computer executable programs, it can implement the intelligent assisted vehicle finding method based on a mobile robot as described in any one of claims 1-6.

10. A computer-readable storage medium, characterized in that: A computer program is stored in a computer-readable storage medium, and when the computer program is executed by a processor, it can implement the intelligent assisted vehicle finding method based on a mobile robot as described in any one of claims 1-6.

Images