CN110275538A - Intelligent cruise vehicle navigation method and system - Google Patents

Abstract

The invention discloses a navigation method for an intelligent cruising vehicle, which includes the following steps: selecting a global navigation route according to the travel of the vehicle body; moving the vehicle body according to the global navigation route and generating a motion path; data; judge whether the car body encounters an obstacle on the moving path according to the depth of field visual data; the car body encounters an obstacle on the moving path, and continues to move according to the global navigation route after the car body avoids the obstacle; Smart cruiser navigation system. The invention of the present application collects the depth of field visual data of the vehicle body on the moving path, and establishes a multi-layer environment-aware navigation map in the navigation route, which can sense obstacles in the multi-layer environment and realize the non-collision navigation of the vehicle body.

Description

Navigation method and system for intelligent cruising vehicle

technical field

The invention relates to the technical field of automatic guided transport vehicles, in particular to a navigation method and system for an intelligent cruise vehicle.

Background technique

In order to improve the efficiency of picking, it is a new warehouse management mode adopted by modern enterprises to replace manual work with robots to complete the complicated picking work in the warehouse system. As for the navigation of the robot, although the magnetic guide and line inspection scheme adopted in the prior art is relatively low in cost, it is necessary to re-arrange the guidance line of the robot when adjusting the layout of the production line, which is time-consuming and laborious, and cannot be quickly adjusted to adapt to the layout of the production line. Change. However, the navigation solution with laser radar as the core can be quickly adjusted according to changes in the layout of the production line, but it cannot accurately avoid obstacles during the navigation process, for example, avoiding obstacles outside the scanning range of the laser, There is a risk of accidents occurring.

Contents of the invention

Aiming at the deficiencies of the prior art, the present invention provides a navigation method and system for an intelligent cruising vehicle.

A kind of intelligent cruising vehicle navigation method comprises:

Select the global navigation route according to the itinerary of the car body;

The car body moves according to the global navigation route and generates a motion path;

Collect the depth of field visual data of the car body on the motion path;

Judging whether the car body encounters obstacles on the moving path according to the depth of field visual data;

When the car body encounters an obstacle on the moving path, the car body continues to move according to the global navigation route after avoiding the obstacle.

According to one embodiment of the present invention, it also includes:

The car body moves to the target position according to the global navigation route, and obtains the correction information of the target position setting;

Adjust the pose of the car body according to the correction information.

According to an embodiment of the present invention, obtaining the correction information set by the target location includes:

Vision scan correction information;

Get the corrected pose of the car body.

According to an embodiment of the present invention, adjusting the pose of the car body according to the correction information includes:

Get the real-time pose of the car body;

Drive the car body so that the real-time pose of the car body is consistent with the corrected pose.

According to an embodiment of the present invention, the vehicle body avoiding obstacles includes:

The car body avoids obstacles according to the trajectory of the obstacle.

According to an embodiment of the present invention, the vehicle body avoiding the obstacle according to the movement track of the obstacle includes:

Obtain the obstacle distance of the obstacle relative to the vehicle body according to the depth of field visual data;

Calculate the trajectory of the obstacle according to the distance of the obstacle;

Plan the local navigation route of the car body according to the movement trajectory of the obstacle;

The car body avoids obstacles according to the local navigation route.

According to an embodiment of the present invention, selecting the global navigation route according to the itinerary of the vehicle body includes:

Construct a working scene; the target position is located in the working scene;

Plan the global navigation route according to the real-time position of the vehicle body in the working scene and the target position;

Select a global navigation route.

According to an embodiment of the present invention, while collecting the depth-of-field visual data of the vehicle body on the moving path, it also includes:

Collect the ultrasonic ranging data of the car body on the moving path;

After judging whether the car body encounters obstacles on the motion path according to the depth of field visual data, it also includes:

According to the ultrasonic ranging data, it is judged whether the vehicle body encounters an obstacle on the moving path.

According to an embodiment of the present invention, while collecting the depth of field visual data of the vehicle body on the moving path,

Also includes:

Collect the laser ranging data of the car body on the moving path;

After judging whether the car body encounters obstacles on the motion path according to the depth of field visual data, it also includes:

According to the laser ranging data, it is judged whether the vehicle body encounters obstacles on the moving path.

A navigation system for an intelligent cruising car includes:

The car body moves according to the global navigation route and generates a motion path;

A visual navigation module, which is used to collect the depth of field visual data of the vehicle body on the motion path;

The main control module judges whether the car body encounters obstacles on the moving path according to the depth of field visual data; the main control module controls the car body to continue to move according to the global navigation route after avoiding obstacles.

According to an embodiment of the present invention, it further includes a visual recognition module; the visual recognition module acquires correction information of the target position; and the main control module adjusts the pose of the vehicle body according to the correction information.

According to an embodiment of the present invention, it also includes a laser navigation module; the laser navigation module plans a global navigation route according to the real-time position of the vehicle body in the working scene and the target position.

According to one embodiment of the present invention, it also includes an ultrasonic ranging module; the ultrasonic ranging module is used to collect ultrasonic ranging data of the car body on the moving path, and transmits it to the main control module; the main control module judges according to the ultrasonic ranging data Whether the car body encounters obstacles on the motion path.

According to one embodiment of the present invention, it also includes a laser ranging module; the laser ranging module is used to collect the laser ranging data of the car body on the moving path, and transmits it to the main control module; the main control module judges the distance according to the laser ranging data Whether the car body encounters obstacles on the motion path.

Compared with the prior art, this application collects the depth of field visual data of the vehicle body on the moving path, and establishes a multi-layer environment-aware navigation map in the navigation route, which can sense obstacles in the multi-layer environment and realize the vehicle body touchless navigation. Moreover, through the correction information set at the target position, the pose of the car body is adjusted at the target position to achieve precise docking between the car and the application environment.

Description of drawings

The drawings described here are used to provide a further understanding of the application and constitute a part of the application. The schematic embodiments and descriptions of the application are used to explain the application and do not constitute an improper limitation to the application. In the attached picture:

Fig. 1 is the control block diagram of intelligent cruising vehicle navigation system in embodiment one;

Fig. 2 is the flow chart of the intelligent cruising vehicle navigation method in embodiment two;

Fig. 3 is the flowchart of selecting the global navigation route according to the itinerary of the car body in the second embodiment;

FIG. 4 is a flow chart of avoiding obstacles according to the trajectory of the obstacle in Embodiment 2;

FIG. 5 is a flow chart of obtaining the correction information of the target position setting in the second embodiment;

Fig. 6 is a flow chart of adjusting the vehicle body pose according to the correction information in the second embodiment;

Fig. 7 is the control block diagram of the intelligent cruising car navigation system in embodiment three;

Fig. 8 is the flow chart of the intelligent cruising vehicle navigation method in the fourth embodiment;

Fig. 9 is the control block diagram of the intelligent cruising vehicle navigation system in the fifth embodiment;

Fig. 10 is a flow chart of the navigation method of the intelligent cruising vehicle in the sixth embodiment.

Explanation of reference signs:

1. Vehicle body; 2. Visual navigation module; 3. Main control module; 31. Storage unit; 32. Control unit; 4. Visual recognition module; 5. Laser navigation module; 6. Drive module; 7. IMU inertial test module ; 10. Ultrasonic ranging module; 20. Laser ranging module.

Detailed ways

A number of embodiments of the present invention will be disclosed in the following figures. For the sake of clarity, many practical details will be described together in the following description. It should be understood, however, that these practical details should not be used to limit the invention. That is, in some embodiments of the invention, these practical details are not necessary. In addition, for the sake of simplifying the drawings, some well-known and commonly used structures and components will be shown in a simple schematic manner in the drawings.

It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain each If the relative positional relationship, movement conditions, etc. between the components change, the directional indication will also change accordingly.

In addition, in the present invention, descriptions such as "first", "second" and so on are only for the purpose of description, and do not refer to the meaning of sequence or sequence, nor are they intended to limit the present invention, but are only for the purpose of distinguishing the following The components or operations described by the same technical terms are only used, but should not be understood as indicating or implying their relative importance or implying the number of indicated technical features. Thus, a feature defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions of the various embodiments can be combined with each other, but it must be based on the realization of those skilled in the art. When the combination of technical solutions is contradictory or cannot be realized, it should be considered that the combination of technical solutions does not exist , nor within the scope of protection required by the present invention.

In order to further understand the content, characteristics and effects of the present invention, the following examples are given, and detailed descriptions are as follows in conjunction with the accompanying drawings:

Embodiment one

Referring to FIG. 1 , FIG. 1 is a control block diagram of an intelligent cruising vehicle navigation system according to an embodiment. The intelligent cruising car navigation system in this embodiment includes a car body 1 , a visual navigation module 2 , a main control module 3 , a visual recognition module 4 , a laser navigation module 5 , a drive module 6 and an IMU inertial measurement module 7 . Wherein, the visual navigation module 2 , the main control module 3 , the visual recognition module 4 , the laser navigation module 5 , the driving module 6 and the IMU inertial measurement module 7 are respectively arranged on the vehicle body 1 . The main control module 3 is respectively connected with the visual navigation module 2, the visual recognition module 4, the laser navigation module 5, the driving module 6 and the IMU inertial measurement module 7.

The vehicle body 1 moves according to the global navigation route and generates a movement path. Wherein, the global navigation route is selected according to the itinerary of the car body 1 . The itinerary here is the task itinerary for the car body 1 to complete the task, and the end of the trip is the target position for the car body 1 to perform the task. For example, if the car body 1 is a storage robot, its task is to move the goods in front of the warehouse door to the shelf in the warehouse, the distance between the front of the warehouse door and the shelf is the task schedule of car body 1, and the shelf is the end of the trip, which is the target position. The global navigation route is that car body 1 moves from the front of the warehouse door to The optimal route for completing the trip among the multiple routes that the shelf may pass is the global navigation route to be selected by the car body 1, for example, the shortest route, the car body 1 moves according to the global navigation route, and the places it moves will generate motion path. The vehicle body 1 in this embodiment can be a storage robot or a sweeping robot, or other mobile platforms, which are not limited here. In a specific application, the car body 1 has omnidirectional movement function. Specifically, the four corners of the bottom of the car body 1 are respectively provided with driving wheels. The driving wheels are Mecanum wheels with omnidirectional movement functions. Each driving wheel An independent motor is installed for driving, and the driving motors of the driving wheels are respectively controlled to realize the independent driving of each driving wheel, and the cooperative driving of the four driving wheels, so as to realize the omnidirectional displacement, steering or in-situ rotation of the car body 1, etc. Function.

The visual navigation module 2 is used to collect the depth of field visual data of the vehicle body 1 on the moving path. It can be understood that when the car body 1 moves along the global navigation route, it is inevitable to encounter some obstacles on the moving path, for example, accidentally dropped goods, pedestrians, or other car bodies 1 and so on. Therefore, these obstacles need to be avoided when the vehicle body 1 is moving to ensure the normal execution of the mission schedule. In order to achieve the above purpose, it is first necessary to find out whether obstacles are encountered on the moving path of the car body 1, and then avoid them. The visual navigation module 2 analyzes and judges by collecting the visual data of the depth of field of the car body 1 on the moving path Whether an obstacle is encountered on the moving path of the car body 1, and avoid it according to the depth of field visual data. The depth of field visual data in this embodiment is visual data formed by collecting images on the moving path of the vehicle body 1 , such as photos or videos. When specifically setting, the visual navigation module 2 can be arranged on the front end of the vehicle body 1, preferably, the visual navigation module 2 is located below the laser navigation module 5, preferably, the number of the visual navigation modules 2 can also be set to multiple, A plurality of visual navigation modules 2 are respectively located on the periphery of the vehicle body 1 in order to collect visual data of depth of field in the surrounding environment of the vehicle body 1 . The visual navigation module 2 in this embodiment can use a binocular camera. The depth of field visual data collected by the visual navigation module 2 will be transmitted to the main control module 3 .

After the main control module 3 receives the depth of field visual data transmitted by the visual navigation module 2, it judges whether the vehicle body 1 encounters an obstacle on the moving path according to the depth of field visual data. , the main control module 3 then controls the car body 1 to avoid obstacles and continue to move according to the global navigation route. During specific setting, the main control module 3 is the control center of the navigation system, which is connected with the visual navigation module 2, the visual recognition module 4, the laser navigation module 5, the drive module 6 and the IMU inertial measurement module 7 through various interfaces and lines, In order to receive and send data information, for example, it can be connected to the visual navigation module 2 through the USB3.0 interface, and connected to the visual recognition module 4, laser navigation module 5, drive module 6 and IMU inertial measurement module 7 through Ethernet communication. The main control module 3 receives the data information of the visual navigation module 2, the visual recognition module 4, the laser navigation module 5 and the IMU inertial measurement module 7, and calculates and processes the received data information, and sends out various processing according to the analysis and processing results command, and control the vehicle body 1 to complete the corresponding stroke movement through the drive module 6 .

Specifically, the main control module 3 includes a storage unit 31 and a control unit 32 . The storage unit 31 is used for data information storage, software storage or executable program storage, etc., for example, to the storage of data information transmitted by the visual navigation module 2, the visual recognition module 4, the laser navigation module 5 and the IMU inertial measurement module 7, Another example is the storage of work scenes and navigation maps, another example is the storage of system software that executes navigation maps, another example is the storage of other setting information such as coordinate data, and another example is the storage of other executable programs. The control unit 32 analyzes and processes the data information transmitted by the visual navigation module 2, the visual recognition module 4, the laser navigation module 5 and the IMU inertial measurement module 7, and calls the data information stored by the storage unit 31 for combined analysis and processing when necessary, and According to the analysis or processing results, call the software stored in the storage unit 31 to perform navigation tasks, or call other executable programs to perform travel tasks, or control the driving module 6 to control the movement and posture adjustment of the car 1 . The control unit 32 in this embodiment may use an STM32 chip.

For example, the visual navigation module 2 collects the depth of field visual data on the movement path of the vehicle body 1 and transmits it to the control unit 32. The control unit 32 judges whether the vehicle body 1 encounters an obstacle on the movement path according to the depth of field data information. When an obstacle is encountered, the distance information of the obstacle relative to the vehicle body 1 is further calculated, that is, the obstacle distance, and then the control unit 32 determines that the obstacle is a static obstacle according to the change of the obstacle distance in a unit time It is still a dynamic obstacle, and then plan the local navigation route according to the obstacle distance of the static obstacle or the moving track of the dynamic obstacle, and then control the car 1 to move according to the local navigation route through the drive module 6 to avoid the obstacle, and then continue according to the The global navigation route moves toward the target position.

The visual recognition module 4 acquires the correction information of the row target position setting, and the control unit 32 adjusts the pose of the vehicle body 1 according to the correction information. The car body 1 will continue to move towards the target position after avoiding obstacles. When the car body 1 moves to the target position, it needs to perform tasks, for example, place the goods to the designated position of the shelf, and at this time the distance between the car body 1 and the shelf and The angle and pose of the car body 1 facing the shelf may not allow the car body 1 to accurately place the goods on the designated position of the shelf, which requires correction of the pose of the car body 1. The above correction information is just the car body 1 The pose information that can accurately perform the task, that is, the distance of the car body 1 relative to the shelf and the angle facing the shelf are just so that the goods can be accurately placed on the designated position of the shelf. In the specific setting, the correction information is set on the ground or the wall through a carrier capable of storing data. In this embodiment, a two-dimensional code is used as the carrier of the correction information. The visual recognition module 4 obtains the correction information by scanning the two-dimensional code. And transmit it to the control unit 32, the control unit 32 adjusts the pose of the vehicle body 1 according to the correction information to complete the corresponding task. In the specific setting, the position where the visual recognition module 4 is set on the vehicle body 1 corresponds to the setting position of the correction information. For example, when the correction information is set on the ground, the visual recognition module 4 is set on the lower part of the The periphery of the car body 1, and the identification end of the visual recognition module 4 should face the ground. The identification end should face the wall. The visual recognition module 4 in this embodiment may use a monocular camera, an image sensor or an optical code reader, which is not limited here. In this way, the visual recognition module 4 obtains the correction information stored in the two-dimensional code by scanning the two-dimensional code set at the target position, so that the car body 1 is adjusted according to the correction information, and adjusted to the best pose for performing tasks, realizing the integration with the application environment. Accurate docking. The correction information in this embodiment is a correction pose.

The laser navigation module 5 plans a global navigation route according to the real-time position and target position of the vehicle body 1 in the working scene. It can be understood that the planning and determination of the navigation route needs to know the starting position and the end position, and also needs to know the situation of each route between the starting position and the end position, which needs to include the starting position and the end position. The scene is detected, in this embodiment, it is the working scene where the car body 1 performs the task, and the end point is the target position. The working scene is constructed through the laser navigation module 5, and the real-time position of the car 1 is determined through the laser navigation module 5, that is, the starting position. When the target position is known, the global navigation route can be planned. The laser navigation module 5 can be arranged on the top of the car body 1, and the laser navigation module 5 can perform 360-degree rotation scanning to obtain the geographical environment of the work scene, and construct a map of the work scene according to the geographical environment information. The working scene in can be a warehouse environment where shelves are placed. The laser navigation module 5 in this embodiment can use laser radar. In a specific application, the laser navigation module 5 first scans the working scene in all directions, and then builds a map of the entire working scene, and stores the built map in the storage unit 31. The software is called by the control unit 32 together. The target position is set in advance, and the laser navigation module 5 continuously performs 360-degree laser scanning to obtain the working scene around the car body 1, and then determines the position of the car 1 in the working scene, obtains the real-time positioning of the car body 1, and then The navigation software plans a global navigation route according to the real-time positioning information of the car body 1 and the position information of the target position.

The drive module 6 includes four drive units, each drive unit is correspondingly connected to the drive motor, and the control unit 32 controls the rotation of the four drive wheels through the four drive units, thereby realizing the control of the omnidirectional movement of the car body 1. This enables the vehicle body 1 to move and rotate in any direction under the control of the control unit 32, so as to achieve the motion effect of going straight, obliquely, sideways or rotating on the spot. Moreover, each driving wheel is correspondingly provided with an IMU inertial measurement module 7, and the IMU inertial measurement module 7 can obtain the odometer information and the heading angle information of the vehicle body 1 and transmit them to the control unit 32, and the control unit 32 according to the IMU inertial measurement module 7 The obtained data and the real-time positioning information of the car body 1 are used to obtain the current position and attitude information of the car body 1. In this way, it is possible to determine whether the real-time pose of the car body 1 is consistent with the corrected pose when performing tasks. After the car body 1 is driven to move and steer so that the real-time pose of the car body 1 is in agreement with the corrected pose, the control unit 32 then controls the car body 1 to perform tasks.

Embodiment two

Referring to Fig. 1 and Fig. 2, Fig. 2 is a flow chart of the navigation method of the intelligent cruising vehicle in the second embodiment. In the present embodiment, the intelligent cruising vehicle navigation method comprises the following steps:

Step S1, select a global navigation route according to the itinerary of the car body 1 .

Step S2, the vehicle body 1 moves according to the global navigation route, and generates a movement path.

Step S3, collecting visual depth data of the vehicle body 1 on the moving path.

Step S4, judging whether the car body 1 encounters an obstacle on the moving path according to the depth of field visual data;

Step S5, the car body 1 encounters an obstacle on the moving path, and the car body 1 continues to move according to the global navigation route after avoiding the obstacle.

Step S6, the car body 1 moves to the target position according to the global navigation route, and acquires the correction information of the target position setting.

Step S7, adjusting the pose of the vehicle body 1 according to the correction information.

By collecting the depth-of-field visual data of the car body 1 on the moving path, a multi-layer environment-aware navigation map is established in the navigation route, which can sense obstacles in the multi-layer environment and realize the non-collision navigation of the car body. Moreover, through the correction information set at the target position, the pose of the car body is adjusted at the target position to achieve precise docking between the car and the application environment.

For ease of understanding, the navigation method of the smart cruiser in this embodiment is further described based on the navigation system of the smart cruiser in the first embodiment.

Continue to refer to FIG. 3 , which is a flow chart of selecting a global navigation route according to the travel of the vehicle body in the second embodiment. Further, in step S1, selecting the global navigation route according to the itinerary of the car body 1 includes the following sub-steps:

In step S11, a working scene is constructed, and the target position is located in the working scene. The laser navigation module 5 installed on the car body 1 first scans the entire working scene, and transmits the data collected by scanning to the control unit 32, and the control unit 32 obtains the point cloud data in the working scene, and then incrementally builds a map of the working scene , and stored in the storage unit 31. In specific applications, the laser navigation module 5 continuously rotates 360 degrees on the car body 1, so that the laser beam can cover a plane, so that the distance information of an object on a plane in the work scene relative to the car body 1 can be measured , and then construct a map of the surrounding environment through the SLAM algorithm. Wherein, the laser navigation module 5 measures the distance information of the object relative to the vehicle body 1 by measuring the flight time of the laser when scanning the entire working scene, specifically, the following formula can be used: d=ct/2, where d is the distance between the vehicle body 1 and the vehicle body 1. The distance of various objects in the working scene, c is the speed of light, and t is the time interval from laser emission to reception. When the working scene changes, the laser navigation module 5 re-scans the new working scene to quickly build a map of the new working scene, so that it can be quickly adjusted according to the change of the production line layout to adapt to various working scenes, which is flexible and changeable. The target position is located in the working scene. The target position here is the end of the journey where the car body 1 needs to perform the task, which is determined according to the task to be performed by the car body 1. For example, the task to be performed by the car body 1 is to move goods to the warehouse The warehouse is the working scene, the target shelf is the target position, and it is also the end of the task journey of car 1.

In step S12, a global navigation route is planned according to the real-time position of the car body 1 in the working scene and the target position. In the working scene, the target position is artificially set according to the task itinerary. When the car body 1 enters the working scene, the laser navigation module 5 scans the surrounding environment information of the car body 1 and transmits it to the control unit 32. The control unit 32 can use the SLAM The algorithm realizes autonomous positioning based on the perception of the surrounding environment of the car body 1, determines the real-time position of the car body 1 in the working scene, and then uses the real-time position of the car body 1 in the working scene and the target position, and the control unit 32 calls The navigation software in the storage unit 31 executes the navigation task. The navigation software plans a global navigation route according to the working scene, the real-time position of the car body 1 and the target position. The global navigation route planning of the shortest path, and then through the A* algorithm, the improved A* algorithm optimizes the planned global navigation route to form an optimal global navigation route. Alternatively, the A* algorithm or the improved A* algorithm can be used to directly plan the global navigation route. Of course, other algorithms can also be used as software navigation algorithms, which are not limited here. Preferably, when there are multiple car bodies 1, the navigation software can also plan the navigation route of each car body 1 in a unified manner, so as to achieve the function of traffic control, so that each car 1 can move smoothly.

Step S13, selecting a global navigation route. When the number of global road navigation routes planned by the navigation software is multiple, select the optimal global navigation route according to the itinerary of the car body 1. When in a work scene, choose a smooth route that does not intersect with other car bodies 1 .

In step S2, the car body 1 moves according to the global navigation route, and generates a moving path. After the global navigation route is determined, the control unit 32 controls the vehicle body 1 to move along the global navigation route through the driving module 6 . When the car body 1 moves along the global navigation route without encountering obstacles, the car body 1 will directly move to the target position, and then execute the task.

In step S3, the depth-of-field visual data of the vehicle body 1 on the moving path is collected. It can be understood that when the car body 1 encounters an obstacle on the moving path, the car body 1 needs to avoid the obstacle before moving to the target position, which requires collecting the depth of field visual data on the moving path of the car body 1 as a judgment Information Sources. Although the laser navigation module 5 can scan the surrounding objects of the car body 1, it only scans in one plane. When performing work scene construction scanning or laser navigation, it cannot scan and identify obstacles outside the plane, so That is, obstacles will be missed, especially obstacles whose height is below the scanning plane of the laser navigation module 5 . Through the visual navigation module 2 arranged on the vehicle body 1, the shooting recognition area of the visual navigation module 2 can cover the area outside the laser radar scanning plane, so as to obtain the depth of field visual data on the movement path of the vehicle body 1. The depth of field vision The data can form a three-dimensional image, and then combined with the map of the laser navigation module 5 in one plane, a multi-layer environment perception map can be constructed. In this way, the visual three-dimensional information is constructed for the space outside the laser scanning plane, and then added In the navigation map, when navigating in this way, the obstacle in the multi-layer environment can be perceived and predicted, especially, the visual perception of the obstacle whose height is lower than the scanning plane of the laser navigation module 5, in order to realize the vehicle Body 1's touchless navigation provides a prerequisite. In addition, compared with laser scanning, the accuracy of the distance information between the obstacle and the vehicle body 1 calculated by visual recognition is higher. The visual navigation module 2 in this embodiment is a binocular camera, which acquires the depth of field visual data on the moving path of the vehicle body 1 through the binocular camera. Specifically, the binocular camera of the visual navigation module 2 is to simulate the scene seen by the eyes of the human body from two different directions, so as to increase the three-dimensional stereoscopic effect of the image. Specifically, the binocular camera shoots an obstacle from two points , taking the images of the obstacle at different viewing angles, according to the pixel matching relationship between the two images, the offset between the pixels is calculated by the principle of triangulation to obtain the three-dimensional information of the obstacle, and then the depth information of the obstacle is obtained , and the distance between the obstacle and the binocular camera can be calculated, and the obstacle distance between the car body 1 and the obstacle can be obtained.

In step S4, it is judged according to the depth vision data whether the vehicle body 1 encounters an obstacle on the moving path. Specifically, the depth of field visual data is the image information on the moving path of the vehicle body 1. After the image information is transmitted to the control unit 32, the control unit 32 analyzes whether there is an object in the image. When the image contains an object, the vehicle body 1 Encountered an obstacle, otherwise, it is not encountered an obstacle.

Continuing to refer to FIG. 4 , which is a flow chart of avoiding obstacles according to their movement trajectories in the second embodiment. Further, in step S5, the vehicle body 1 encounters an obstacle on the moving path, and the vehicle body 1 continues to move according to the global navigation route after avoiding the obstacle. Wherein, the vehicle body 1 avoids the obstacle according to the movement trajectory of the obstacle, which includes the following sub-steps:

S51. Obtain the obstacle distance of the obstacle relative to the vehicle body 1 according to the depth vision data. The control unit 32 receives the depth of field visual data transmitted by the visual navigation module 2, and calculates the distance of the obstacle relative to the vehicle body 1, which is the obstacle distance. The calculation of obstacle distance is based on the principle of binocular distance measurement. Binocular distance measurement mainly utilizes the difference between the horizontal coordinates of the obstacle on the left and right images, that is, there is a difference between the parallax and the distance Z from the obstacle to the imaging plane. Inverse proportional relationship, the specific formula is as follows: Z=Bf/h, where B is the baseline, expressed as the distance between the two cameras, f is the focal length of the two cameras, h is the parallax, that is, the pixel coordinate difference of the object in the two cameras .

S52. Calculate the movement track of the obstacle according to the distance of the obstacle. The control unit 32 calculates the movement trajectory of the obstacle according to the numerical change of the obstacle distance; the vehicle body 1 stops moving, and when the obstacle distance does not change within a unit time, it means that the obstacle does not move and is a static obstacle; If the distance of the inner obstacle changes, it means that the obstacle is in a moving state and is a dynamic obstacle. The obstacle distance is calculated once per unit time, and the trajectory of the obstacle can be known according to the change of the obstacle distance. Alternatively, the visual navigation module 2 acquires the abscissas in multiple consecutive images, and performs curve fitting on the abscissas of the obstacle in the multiple consecutive images according to the curve fitting algorithm to obtain the movement track of the obstacle. Specifically, the control unit 32 calculates the moving track of the obstacle by calling the software in the storage unit 31 .

S53, planning a local navigation route of the vehicle body 1 according to the movement track of the obstacle. When the obstacle is a static obstacle, the control unit 32 analyzes and calculates according to the obstacle distance information, and plans a local navigation route, so that the vehicle body 1 can avoid the obstacle navigation route. For the planning of the local navigation route, the control unit 32 can plan by invoking the navigation software. After the local navigation route planning is completed, the car body 1 will continue to move towards the target position according to the local navigation route after bypassing static obstacles. . For example, for static obstacles, the navigation software can directly plan a local navigation route that can avoid obstacles. When the obstacle is dynamic, the navigation software controls the car body 1 to move according to the local navigation route after planning a local navigation route that can avoid the obstacle. At this time, the visual navigation module 2 continues to collect the depth of field visual data of the obstacle, and Pass it to the control unit 32, and the control unit 32 monitors whether the trajectory of the obstacle changes according to the new depth of field visual data, so as to call the navigation software again to avoid the local navigation route again, and ensure that the vehicle body 1 can avoid the obstacle. The navigation software of the local navigation route is consistent with the navigation software of the global navigation route. The starting position of the plan is the real-time position of the car body 1, and the end position is the position behind the obstacle on the global navigation route. An image of a motion trajectory.

S54, the vehicle body 1 avoids obstacles according to the local navigation route.

Referring to Fig. 5, Fig. 5 is a flow chart of obtaining the correction information of the target position setting in the second embodiment; in step S6, obtaining the correction information of the target position setting includes the following sub-steps:

S61, vision scanning correction information. After the vehicle body 1 moves to the target position, the vision-corrected carrier two-dimensional code is visually scanned and recognized by the visual recognition module 4 . In a specific application, if the visual recognition module 4 cannot accurately scan the two-dimensional code, the target position is used as the center point, and the center point is used as the center to scan within a similar range until the two-dimensional code is scanned.

S62. Obtain the corrected pose of the car body 1 . The visual recognition module 4 reads the correction information in the two-dimensional code carrier, that is, the correction pose. It can be understood that the target position is the end point of the movement of the car body 1 in the working scene, that is, the end of the stroke. The car body 1 needs to perform tasks. For example, after the warehouse robot delivers the goods to the destination, it needs to be placed on the target point of the shelf, and In order to accurately move the cargo to the target point, the vehicle body 1 needs to be in a position and posture facing the target point to accurately move the cargo to the target point. However, when the car body 1 moves to the target position, it cannot ensure that it is just in the position and attitude of the target point. The correction information set at the target position is just to correct the position and attitude of the car body 1, so that the car body 1 can To accurately move the goods to the target point, the correction information is exactly the position and attitude that the car body 1 can accurately move the goods to the target. The correction information uses a two-dimensional code as a storage medium, in which the corrected pose is stored, and the corrected pose is exactly the position and posture where the vehicle body 1 faces the target point and can complete the task. In addition, the two-dimensional code also stores ID information, which is used for the identification of the two-dimensional code itself. When setting the itinerary task, the ID information can be used as the identification information at the end of the itinerary to avoid interference from other two-dimensional codes. At the same time, it is also easy to manage. For example, the shelf corresponds to the ID information of the two-dimensional code, and the goods correspond to the shelf. In this way, the information of the goods and the shelf can be known through the ID information, which is convenient for management.

Continue to refer to Fig. 6, Fig. 6 is the flowchart of adjusting the pose of the car body according to the correction information in the second embodiment. In step S7, adjusting the pose of the car body 1 according to the correction information includes the following sub-steps:

S71, acquiring real-time attitude information of the car body 1 . Specifically, the mileage information and heading angle information of the vehicle body 1 are obtained through the IMU inertial measurement module 7 provided on the vehicle body 1, and then the real-time pose information of the vehicle body 1 is obtained.

S72, driving the car body 1 so that the real-time pose of the car body 1 is consistent with the corrected pose. Specifically, the control unit 32 judges whether the real-time pose information of the car body 1 is consistent with the corrected pose. If they are consistent, the car body 1 does not need to be adjusted. The driving wheel rotates, and the IMU inertial measurement module 7 performs real-time feedback until the real-time pose of the car body 1 is consistent with the corrected pose, so that the task can be performed correctly.

Implementation three

Referring to Fig. 7, Fig. 7 is a control block diagram of the navigation system of the intelligent cruiser in the third embodiment. The difference between the intelligent cruising vehicle navigation system in this embodiment and the first embodiment is that it also includes an ultrasonic distance measuring module 10 . The ultrasonic ranging module 10 is connected with the control unit 32 of the main control module 3 . The ultrasonic ranging module 10 is used to collect the ultrasonic ranging data of the vehicle body 1 on the moving path, and transmit it to the control unit 32 of the main control module 3, and the control unit 32 of the main control module 3 judges the vehicle body 1 according to the ultrasonic ranging data. Whether an obstacle is encountered on the motion path.

It is understandable that there are various objects as obstacles, and some obstacles are obstacles that cannot be detected by optical sensors such as lidar or cameras, for example, glass walls; for this category of obstacles, ultrasonic to identify. In this embodiment, the ultrasonic ranging data on the moving path of the vehicle body 1 or around the vehicle body 1 is collected by adding the ultrasonic ranging module 10 on the vehicle body 1, especially the ultrasonic ranging data of obstacles that cannot be detected by the optical sensor Collect and form a comprehensive identification of obstacle types to avoid omission of obstacles. Then, the ultrasonic distance measurement module 10 transmits the collected ultrasonic distance measurement data to the control unit 32, and the control unit 32 calculates the obstacle distance of the obstacle relative to the vehicle body 1 according to the ultrasonic distance measurement data, and forms the movement track of the obstacle , and then control the driving module 6 to drive the car body 1 to avoid obstacles according to the obstacle distance and the obstacle movement trajectory, and then continue to move according to the global navigation route. Time-of-flight is used to measure the distance information of the object relative to the vehicle body 1, which will not be repeated here. The calculation principle and method of the same obstacle movement trajectory are also consistent with those described above, and will not be repeated here. In a specific setting, the ultrasonic ranging module 10 can be set at the front of the vehicle body 1 . Preferably, there are multiple ultrasonic distance measuring modules 10, and the plurality of ultrasonic distance measuring modules 10 are arranged in a 360-degree ring around the periphery of the vehicle body 1, so as to facilitate ultrasonic distance measurement for all obstacles around the vehicle body 1. The ultrasonic ranging module 10 in this embodiment may use an ultrasonic sensor.

Embodiment four

Referring to FIG. 8 , FIG. 8 is a flow chart of the navigation method of the intelligent cruising vehicle in the fourth embodiment. The difference between the intelligent cruising vehicle navigation method in the present embodiment and the second embodiment is:

In step S3, while collecting the depth of field visual data of the vehicle body 1 on the moving path, the ultrasonic ranging data of the vehicle body 1 on the moving path is also collected. That is, the depth of field visual data collected by the visual navigation module 2 in real time and the ultrasonic ranging data actually collected by the ultrasonic ranging module 10 are respectively transmitted to the control unit 32, and then the control unit 32 performs overall analysis and processing on the above data, and the different types of obstacles The obstacle distance is calculated.

In step S4, after judging whether the vehicle body 1 encounters an obstacle on the moving path according to the depth of field visual data, it also includes: judging whether the vehicle body 1 encounters an obstacle on the moving path according to the ultrasonic ranging data. That is, the control unit 32 successively judges whether obstacles on the moving path of the vehicle body 1 are encountered according to the depth of field visual data and ultrasonic ranging data, including ordinary obstacles and obstacles that cannot be recognized by optical sensors, such as glass walls, so as to realize the detection of obstacles. comprehensive identification of objects. In actual application, the above two judgments may be performed simultaneously, which is not limited here.

In step S51, while obtaining the obstacle distance of the obstacle relative to the vehicle body 1 according to the depth vision data, it also includes obtaining the obstacle distance of the obstacle relative to the vehicle body 1 according to the ultrasonic ranging data.

Embodiment five

Referring to Fig. 9, Fig. 9 is a control block diagram of the navigation system of the intelligent cruising vehicle in the fifth embodiment. The difference between the intelligent cruising vehicle navigation system in this embodiment and the third embodiment is that it also includes a laser distance measuring module 20 . The laser ranging module 20 is used to collect the laser ranging data of the vehicle body 1 on the moving path, and transmit it to the control unit 32 of the main control module 3, and the control unit 32 of the main control module 3 judges the vehicle body 1 according to the laser ranging data. Whether an obstacle is encountered on the motion path.

It can be understood that, compared with laser ranging, the range and sensitivity of the visual navigation module 2 and the ultrasonic ranging module 10 are slightly worse, and the main function of the laser navigation module 5 is to carry out navigation and real-time positioning, and The distance can only be measured in one plane. By adding the laser distance measurement module 20 to perform additional laser distance measurement on the movement path of the vehicle body 1 or around the vehicle body, the laser distance measurement data is directly transmitted to the control unit 32, and then the control unit 32 Calculate the obstacle distance of the obstacle relative to the vehicle body 1, and form the trajectory of the obstacle, and then avoid the obstacle according to the obstacle distance and the obstacle trajectory, which will reduce the reaction time of the vehicle body 1 to avoid obstacles, and increase The sensitivity of the vehicle body 1 to avoid obstacles at a relatively long distance facilitates obstacle avoidance in advance. The calculation principle and method of calculating the obstacle distance and obstacle movement trajectory based on the laser ranging data are consistent with those described above, and will not be repeated here. And through the cooperation of the laser ranging data collected by the laser ranging module 20 and the depth of field visual data collected by the visual navigation module 2, the control unit 32 can first judge obstacles in advance through laser ranging, when the vehicle body 1 is closer to the obstacle , then the second confirmation by visual distance measurement also ensures the accuracy of the entire distance measurement and obstacle avoidance. In a specific setting, the laser ranging module 20 can be installed on the head of the vehicle body 1 to measure the distance of the 270-degree area of the head of the vehicle body 1 . Preferably, two laser ranging modules 20 may be arranged at diagonal positions of the vehicle body 1 to perform laser ranging in a 360-degree direction around the vehicle body 1 . The laser ranging module 20 in this embodiment may use a laser sensor. Preferably, a 360-degree surrounding anti-collision edge can also be set around the vehicle body 1 to physically prevent unavoidable collisions.

Embodiment six

Referring to FIG. 10 , FIG. 10 is a flow chart of the navigation method of the intelligent cruising vehicle in the sixth embodiment. The difference between the intelligent cruising vehicle navigation method in this embodiment and the fourth embodiment is:

In step S3, while collecting the depth of field visual data and ultrasonic ranging data of the vehicle body 1 on the moving path, laser ranging data on the moving path of the vehicle body 1 are also collected. That is, the depth of field visual data collected in real time by the visual navigation module 2, the ultrasonic ranging data actually collected by the ultrasonic ranging module 10, and the laser ranging data collected by the laser ranging module 20 are respectively transmitted to the control unit 32, and then the control unit 32 controls the above-mentioned The data is analyzed and processed as a whole, and the obstacle distance of obstacles of different types and different areas is calculated.

In step S4, after judging whether the car body 1 encounters an obstacle on the moving path according to the depth of field visual data and judging whether the car body 1 encounters an obstacle on the moving path according to the ultrasonic ranging data, it also includes: After judging whether the vehicle body 1 encounters an obstacle on the moving path based on the distance data. That is, the control unit 32 successively judges whether obstacles are encountered on the moving path of the car body 1 according to the depth of field visual data, ultrasonic ranging data and laser ranging data, including ordinary obstacles and obstacles that cannot be recognized by optical sensors, such as glass Walls, including far and near obstacles, realize comprehensive recognition of obstacles and full area recognition. In actual application, the above three judgments may be performed simultaneously, which is not limited here.

In step S51, while obtaining the obstacle distance of the obstacle relative to the vehicle body 1 according to the depth of field vision data and obtaining the obstacle distance of the obstacle relative to the vehicle body 1 according to the ultrasonic ranging data, it also includes: obtaining the obstacle according to the laser vision data Obstacle distance relative to vehicle body 1.

To sum up, by collecting the depth of field visual data of the car body on the moving path, a multi-layer environment perception navigation map is established in the navigation route, which can sense obstacles in the multi-layer environment and realize the non-collision navigation of the car body. Moreover, through the correction information set at the target position, the pose of the car body is adjusted at the target position to achieve precise docking between the car and the application environment. In addition, triple anti-collision perception is carried out through ultrasonic, laser and vision, and all-round and all-type identification of obstacles on the movement path of the car body is carried out to further ensure the collision-free navigation of the car body.

The above is only an embodiment of the present invention, and is not intended to limit the present invention. Various modifications and variations of the present invention will occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the scope of the claims of the present invention.

Claims (14)

1. a kind of intelligent cruise vehicle air navigation aid characterized by comprising

Global navigation routine is selected according to the stroke of car body；

Car body (1) is moved according to the global navigation routine, and generates motion path；

Acquire depth of field vision data of the car body (1) on the motion path；

Judge whether the car body (1) encounters barrier on the motion path according to the depth of field vision data；

The car body encounters the barrier on the motion path, and the car body (1) continues root after evading the barrier It is moved according to the global navigation routine.

2. intelligent cruise vehicle air navigation aid according to claim 1, which is characterized in that further include:

The car body (1) moves to target position according to the global navigation routine, and obtains repairing for the target position setting Positive information；

The pose of the car body (1) is adjusted according to the update information.

3. intelligent cruise vehicle air navigation aid according to claim 2, which is characterized in that obtain the target position setting Update information includes:

Update information described in visual scanning；

Obtain the amendment pose of the car body (1).

4. intelligent cruise vehicle air navigation aid according to claim 3, which is characterized in that adjust institute according to the update information The pose for stating car body (1) includes:

Obtain the real-time pose of the car body (1)；

The car body (1) is driven, so that the real-time pose of the car body (1) is consistent with the amendment pose.

5. intelligent cruise vehicle air navigation aid according to any one of claims 1 to 4, which is characterized in that the car body (1) is evaded The barrier includes:

The car body (1) evades the barrier according to the motion profile of the barrier.

6. intelligent cruise vehicle air navigation aid according to claim 5, which is characterized in that the car body (1) is according to the barrier Hinder the motion profile of object to evade the barrier to include:

Distance of obstacle of the barrier relative to the car body (1) is obtained according to the depth of field vision data；

The motion profile of the barrier is calculated according to the distance of obstacle；

Go out the Local Navigation route of the car body (1) according to the Motion trajectory of the barrier；

The car body (1) evades the barrier according to the Local Navigation route.

7. according to any intelligent cruise vehicle air navigation aid of claim 2 to 4, which is characterized in that according to the stroke of car body The global navigation routine is selected to include:

Construct operative scenario；The target position is located in the operative scenario；

It cooks up the overall situation according to real time position of the car body (1) in the operative scenario and the target position and leads Air route line；

Select the global navigation routine.

8. intelligent cruise vehicle air navigation aid according to claim 1, which is characterized in that acquire the car body (1) described While depth of field vision data on motion path, further includes:

Acquire ultrasonic distance measurement data of the car body (1) on the motion path；

After judging whether the car body (1) encounters barrier on the motion path according to the depth of field vision data, also Include:

Judge whether the car body (1) encounters barrier on the motion path according to the ultrasonic distance measurement data.

9. intelligent cruise vehicle air navigation aid according to claim 1, which is characterized in that acquire the car body (1) described While depth of field vision data on motion path, further includes:

Acquire laser ranging data of the car body (1) on the motion path；

After judging whether the car body (1) encounters barrier on the motion path according to the depth of field vision data, also Include:

Judge whether the car body (1) encounters barrier on the motion path according to the laser ranging data.

10. a kind of intelligent cruise vehicle navigation system characterized by comprising

Car body (1) moves according to global navigation routine, and generates motion path；

Vision guided navigation module (2) is used to acquire depth of field vision data of the car body (1) on the motion path；

Main control module (3) judges whether the car body (1) encounters on the motion path according to the depth of field vision data Barrier；The main control module (3), which controls after the car body (1) evades the barrier, continues according to the global navigation routine Movement.

11. intelligent cruise vehicle navigation system according to claim 10, which is characterized in that it further includes visual identity module (4)；The visual identity module (4) obtains the update information of the target position；The main control module (3) is according to the amendment Information adjusts the pose of the car body (1).

12. intelligent cruise vehicle navigation system according to claim 11, which is characterized in that it further includes laser navigation module (5)；The laser navigation module (5) is according to real time position of the car body (1) in operative scenario and the target position Cook up the global navigation routine.

13. intelligent cruise vehicle navigation system according to claim 10, which is characterized in that it further includes ultrasonic distance measurement mould Block (10)；The ultrasonic distance measuring module (10) is for acquiring ultrasonic distance measurement of the car body (1) on the motion path Data, and it is transferred to the main control module (3)；The main control module (3) judges the vehicle according to the ultrasonic distance measurement data Whether body (1) encounters barrier on the motion path.

14. intelligent cruise vehicle navigation system according to claim 10, which is characterized in that it further includes laser ranging module (20)；The laser ranging module (20) is used to acquire laser ranging data of the car body (1) on the motion path, and It is transferred to the main control module (3)；The main control module (3) judges the car body (1) in institute according to the laser ranging data It states and whether encounters barrier on motion path.